Late-stage systemic immune effectors in Plasmodium berghei ANKA infection: biopterin and oxidative stress

Introduction

Malaria infection due to Plasmodium falciparum (P. falciparum) remains a devastating health problem all over the world, with an estimated 300–500 million cases occurring annually that lead to 1.5–3 million deaths [1]. Plasmodium berghei ANKA (P. berghei ANKA) infection is one of the best available mouse models for human P. falciparum-mediated cerebral malaria, which kills nearly 2.5 million people every year because of destructive neurological syndrome [2]. Indeed, infection with P. berghei ANKA, a rodent malarial parasite, produces a disease that closely mimics the features of malarial infection in man and induces neurological symptoms [3]. Despite the innate and adaptive immune response of the host against malarial infection, Plasmodium spp. causes the worst parasitic diseases in humans and evades host immunity in complicated ways [4].

It is thought that two factors are mainly responsible for the inability of the host to resist malarial infection. First, the parasite can escape from host immunity by antigenic variation, via its capacity to continuously alter the surface-exposed antigenic proteins that are vulnerable to antibody recognition and attack [5]. Actually, antigenic variation at the Plasmodium-infected erythrocyte surface plays a critical role in malaria disease severity and host immune evasion. Small variant gene families comprising the “Plasmodium interspersed repeat (pir) multigene gene family” of P. vivax, P. knowlesi and the rodent malarial types P. chabaudi, P. berghei and P. yoelii also show the same features, suggesting a role in antigenic variation and immune evasion [6]. Second, inappropriate immune activation following P. berghei ANKA infection might be due to the activation and expansion of T cells. The immune escape of malarial parasites requires the activation of regulatory T cells (Tregs) [7]. Furthermore, Walther et al. [8] found that up-regulation of Tregs correlates with rapid parasite growth during human malarial infection. Consequently, T-cell immune responses are critical for the protection of the host and for disease pathogenesis during infection with Plasmodium species [9].

Moreover, enhanced catabolism of tryptophan (Trp) is proposed as a sequential defense mechanism for the immune response against invading cells [10]. In this respect, interferon-γ (IFN-γ)-induced indoleamine 2,3-dioxygenase (IDO) activity and its antimicrobial effect may be taken into consideration. However, Villegas-Mendez et al. [11] showed that IFN-γ is not required for the activation of splenic CD4+ and CD8+T cells during P. berghei ANKA infection, but that it is essential for the contraction of the splenic effector T-cell population through the induction of apoptosis. Actually, IDO is the first and rate-limiting enzyme of the l-tryptophan-l-kynurenine (Kyn) pathway and IDO-mediated Trp depletion has a predominantly antimicrobial effect during the early phase of infection [12]. Although similar amounts of plasma, brain and spleen Trp concentrations were found in P. berghei ANKA-uninfected subjects, non-cerebral malarial and cerebral malarial (CM) groups, Sanni et al. [9] suggested that IDO expression in the brain has been implicated in the late-stage immunopathology of CM. IDO inhibition slightly suppresses parasite density in association with enhanced proliferation of CD4+ T cells in response to malarial parasites, but its biological significance remains unclear [4].

In addition, host immune response to malaria also involves phagocytosis as well as production of nitric oxide (NO) and oxygen radicals that form part of the host defense system and also contribute to the pathology of the disease. In this context, hemoglobin degradation by malarial parasites produces redox active by-products–free heme, superoxide and hydrogen peroxide (H₂O₂)–that enhance lipid peroxidation and confer oxidative insult on the host cells [13]. Since activation of IDO-mediated Trp metabolism is strongly redox sensitive [14], the present study was designed to investigate whether the dependence of systemic Trp degradation and concomitant oxidative stress increases the susceptibility to P. berghei ANKA 6653 strain parasitemia or not, on day 7 post-inoculation.

Materials and methods

In this study, six male Swiss Albino mice aged between 14 and 16 weeks were used to determine the oxidative stress response and Trp degradation profile in a murine malarial model. The mice were infected with the P. berghei ANKA 6653 strain. The control group also consisted of eight male mice with the same characteristics [mean±standard error of the mean (SEM): 44.90±1.33 g]. All experimental procedures on mice were consistent with the International Guiding Principle for Biomedical Research Involving Animals, and our research plan was approved by the Local Ethics Committee for Experiments on Animals of Gazi University. To create the malarial infection model, 4.8×10⁵ parasites were injected intraperitoneally into each mouse in the study group. The course of infection was followed by a daily determination of parasitemia by Giemsa-stained blood smear. When the parasitemia reached approximately 5% of the initial inoculation (at the onset of the CM syndrome), treatment groups were assigned. Seven days after the injection, blood samples were obtained from the tail veins of the mice and Giemsa-stained thin layer preparations were made. The preparations were evaluated microscopically for parasitemia. Malarial infection was confirmed clinically, as well as microscopically (data not shown). The liver, spleen and brain tissue samples and serum samples were kept at –80°C until evaluation.

As markers of oxidative stress in tissue and serum samples, the levels of NOx (the stable end products of NO) and the levels of malondialdehyde (MDA; the end product of lipid peroxidation) were studied. In addition, to determine the antioxidant response in the tissues and serum, superoxide dismutase (SOD) and glutathione peroxidase (GPx) enzyme activities were evaluated.

To determine the serum MDA levels (nmol/mL), the quantity of thiobarbituric acid reactive substances (TBARS) in the serum samples as an index of MDA production and hence lipid peroxidation was determined by the method described by Yoshioka et al. [15]. The standard MDA levels were expressed as nanomoles per milliliter (nmol/mL). The levels of TBARS were determined in tissue samples homogenized at a ratio of 1:10 (w/v) in 1.5% (w/v) cold potassium chloride solution by a thiobarbituric acid method, and the results were obtained in nanomoles per gram (nmol/g) of tissue weight by the method described by Mihara and Uchiyama [16].

NOx (μmol/mL) levels were determined according to the principle of Griess reaction [17]. SOD activity was determined using an SOD assay kit (Cayman Chemical, Ann Arbor, MI, USA). This kit uses a tetrazolium salt to detect the superoxide radicals generated by xanthine oxidase and hypoxanthine. One unit of SOD was defined as an amount of enzyme needed to exhibit 50% dismutation of the superoxide radical [18]. GPx activity was determined using a GPx assay kit (Cayman Chemical). This kit measures GPx activity indirectly by a coupled reaction with glutathione reductase [19].

Cardiac blood samples and urine samples of the control and the P. berghei ANKA-infected mice were collected to detect urinary biopterin, serum Trp and Kyn levels, and then the samples were stored at –80°C until the analysis. The biopterin and creatinine levels in the urine samples were analyzed by high-performance liquid chromatography (HPLC) [20]. The Trp and Kyn concentrations in serum were determined by reversed-phase HPLC. In order to estimate the Trp degradation, Kyn-to-Trp ratio (Kyn/Trp) was calculated by dividing the Kyn concentration (μmol/L) by the Trp concentration (mmol/L) [21].

Results

On the seventh day, malarial infection was confirmed clinically, as well as microscopically (data not shown). The liver and spleen of the infected mice (1.52±0.06 g and 0.09±0.01 g, respectively) were significantly heavier than those of the control group (2.24±0.15 g and 0.54±0.10 g, respectively) (p<0.05), confirming the pathological changes, while the overall weight of the study group (31.67±1.08 g) was significantly lower than the that of the healthy control group (44.90±1.33 g) (p<0.05). Parasitemia was also detected in the erythrocytes and brain of the infected mice. In the P. berghei ANKA-infected mice, neither the serum Trp nor the Kyn concentration was different compared to the control group (p>0.05). Despite the unchanged Kyn/Trp, urinary biopterin excretion, the oxidized tetrahydrobiopterin product, significantly increased to 1710.78±213.91 μmol biopterin/mol creatinine in the infected mice compared to the control group: 777.49±102.98 μmol biopterin/mol creatinine (p<0.05) (Table 1). This increase in the excretion of biopterin was consistent with the elevated levels of NOx, as tetrahydrobiopterin is the cofactor of NO synthase. The infection caused a significant rise in the serum, liver, spleen and brain NOx concentrations (2.65±0.22 μmol/mL, 42.61±5.84 μmol/mL, 27.25±1.57 μmol/mL and 39.58±4.37 μmol/mL, respectively) when compared with the matched control group (0.78±0.08 μmol/mL, 20.44±1.51 μmol/mL, 16.47±1.65 μmol/mL and 13.63±2.67 μmol/mL, respectively) (p<0.05) (Figure 1). However, we could not observe a correlation between urinary biopterin levels and serum Kyn/Trp in either the P. berghei ANKA-infected group or the healthy mice (p>0.05). Malarial infection caused a significant increase in the MDA lipid peroxidation marker and GPx in the liver, brain and serum (MDA: 12.40±0.86 nmol/g, 19.60±2.44 nmol/g and 24.40±3.07 nmol/mL, respectively; GPx: 155.01±14.41 U/mL, 373.97±19.45 U/mL and 20.51±1.10 U/mL, respectively) of the infected mice (p<0.05) and a decrease in SOD levels (0.71±0.04 U/mL, 4.42±0.23 U/mL and 0.25±0.004 U/mL, respectively) compared to the control group (MDA: 1.86±0.26 nmol/g, 7.00±1.23 nmol/g and 11.43±1.32 nmol/mL; GPx: 78.89±4.37 U/mL, 216.11±24.42 U/mL and 5.04±0.80 U/mL; SOD: 1.04±0.04 U/mL, 10.94±5.34 U/mL and 0.44±0.02 U/mL, respectively) (p<0.05) (Figures 2–4). Moreover, the levels of SOD and MDA in the spleen remained unchanged compared to the control group (p>0.05). Despite the increased activity of GPx in the other organs, spleen GPx was significantly decreased in the infected group (p<0.05).

Figure 1:

Comparison between the nitric oxide (NOx) levels of the control group and the P. berghei ANKA-infected mice.

Figure 2:

Comparison between the malondialdehyde (MDA) levels of the control group and the P. berghei ANKA-infected mice.

Figure 3:

Comparison between the glutathione peroxidase (GP) levels of the control group and the P. berghei ANKA-infected mice.

Figure 4:

Comparison between the superoxide dismutase (SOD) levels of the control group and the P. berghei ANKA-infected mice.

Discussion

Despite the known pathologies in the tissues, the precise mechanism leading to severe distress in CM is poorly understood.

Because of the life cycle of malarial parasite, infection of red blood cells with Plasmodium subjects them to oxidative stress. Hemoglobin degradation by malarial parasite of the infected cells produces a superoxide, H₂O₂, that enhances lipid peroxidation and activates host cell hexose monophosphate shunt. This stress is sustained during the digestion of host cell hemoglobin by the parasite [22]. It has been recently shown that de novo synthesis of GSH is not essential for the survival of the blood stages of the rodent malarial parasite, P. berghei ANKA [23]. Eventually, erythrocyte defense mechanisms, namely, SOD, catalase, GPx, nicotinamide adenine dinucleotide phosphate (NADPH), nicotinamide adenine dinucleotide, reduced glutathione (GSH) and glutathione reductase, are depleted in the host owing to the infection [24]. Therefore, efficient antioxidant and redox systems are required to protect the cells from the parasites’ oxidative damage caused by reactive oxygen species (ROS). Furthermore, intraerythrocytic malarial parasites also impose an oxidative stress on their host cells.

In our study, the liver, brain and serum of P. berghei ANKA-infected mice showed significantly higher GPx concentrations in comparison to those of the control group. GPx catalyzes the reduction of H₂O₂ by two molecules of GSH as part of the defense system against ROS. Consequently, GSH is converted into oxidized glutathione during the elimination of H₂O₂ and of lipid peroxides by the activity of GPx [25]. Considering the relative contributions of catalase and GPx to the elimination of H₂O₂, Cohen and Hochstein [26] stated that GPx is the major route for H₂O₂ breakdown under physiological conditions. However, at higher concentrations of H₂O₂, the action of catalase becomes increasingly important [26]. Nevertheless, Gaetani et al. [25] demonstrated that catalase and GPx are equally involved in the removal of H₂O₂ in human erythrocytes. However, failure of only one of the two mechanisms for disposing H₂O₂ may not be deleterious for the host [27]. When the erythrocyte is under peroxidative stress, the cell enhances its ability to destroy H₂O₂ through the NADPH-dependent glutathione reductase/GPx system [28]. During the blood stages of malarial parasites, the parasites endocytose large quantities of the surrounding erythrocyte cytoplasm and deliver them to a digestive food vacuole via vesicles [29]. If the uptake of GSH into the endocytic vesicles is followed by the transport into the parasite’s cytoplasm, the parasite might use the host GSH for its own GSH metabolism. In this case, despite the high GPx activity, the host cannot overcome the oxidative stress because of the depletion of the available GSH of the host cells.

When the increase in GPx activity is accompanied by an adequate amount of SOD, dismutation of ROS by SOD generates peroxides, which are further reduced to lipid peroxides by GPx [30]. SOD catalyzes the oxidation/reduction conversion of superoxide radicals to H₂O₂. GPx converts H₂O₂ into water [31]. Plasmodium berghei ANKA lacks endogenous SOD; instead, it appears to take up and concentrate SOD from its host erythrocytes. Thus the adopted host enzyme is localized inside the lysosomes [32]. Actually, P. berghei ANKA in mice was found to derive a substantial amount of SOD from the host cell cytoplasm. It is evident that plasmodia isolated from mouse red cells contain a mouse type of SOD [33]. Linares et al. [34] reported that the functional failure of the host’s antioxidant defense system at the level of SOD is associated with the reduced expression of the transcription factors of the redox status in the host brain. The authors also indicated that hsp70, which is a regulator of the SOD-1 and SOD-2 enzymes, is down-regulated during the course of CM [34]. Thus, in our study, reduced tissue and serum SOD concentrations might be related to the transcriptional inhibition of the SOD enzyme activity. Also, P. berghei ANKA might have taken up a large amount of the SOD enzyme from its host cells. Therefore, inadequate levels of SOD might lead to excessive superoxide radical availability.

Nevertheless, the significant increase in the levels of MDA in the serum and tissues (liver and brain) of our P. berghei ANKA-infected mice indicated an extensive lipid peroxidation due to inadequate host antioxidant capacity [35]. The higher levels of NOx in the liver, spleen and brain as well as in the serum of P. berghei ANKA-infected mice confirmed the systemic effect of NO.

Actually, the induction of inducible nitric oxide synthase (iNOS) and the subsequent biological actions of NO are complex. The net effect of NO depends on its available concentrations in body fluids, target cell type and interactions with ROS [36]. Therefore, the amount of serum NO availability most probably might be due to the formation of peroxynitrite and other reactive nitrogen species (RNS), nitrite-NO pathway, rather than to the iNOS expression rate of inflammatory cells. Regarding the confirmation of oxidative stress, actual plasma NO concentrations are more important indicators than iNOS expression [37]. In this study, the eventual higher NOx and MDA levels in the liver and brain tissues of P. berghei ANKA-infected mice may also be an indicator of multiple organ failure. Gramaglia et al. [38] indicated that exogenously given NO inactivates NO-scavenging free hemoglobin in the plasma and restores nitrite to the concentrations observed in uninfected mice; therefore they concluded that low rather than high NO bioavailability contributes to the genesis of experimental cerebral malaria.

Activation of macrophages and subsequent development of strong pro-inflammatory immune responses are key events in the pathogenesis of severe CM in mice [39]. Couper et al. [40] showed that parasitized red blood cell-derived microparticles as a major inducer of systemic inflammation during malarial infection can stimulate tumor necrosis factor (TNF)-α production by macrophages. Microparticles derived from P. berghei ANKA-infected TNF-α-/- and IFN-γ-/- mice promote the up-regulation of TNF-α production more than live infected red blood cells [40]. TNF-α is apparently required for P. berghei ANKA-induced NO up-regulation and subsequent pathology resulting in fatal CM. In the absence of TNF-α, endothelial intercellular adhesion molecule-1 expression as well as the systemic release of NO is reduced, and P. berghei ANKA infection fails to cause fatal cerebral malaria [41].

Cross-regulation between iNOS and IDO is important for the functions of the host cell and may also prevent the accumulation of potentially toxic metabolites in the Trp degradation pathway [42]. In fact, IDO induction converts Trp to N-formylkynurenine and results in Trp depletion. Consequently, Trp depletion causes the inhibition of parasite growth. Actually, the expression of iNOS or IDO appears to depend on the cell type or tissue. IDO activity is characterized best by the Kyn-Trp ratio, which correlates with concentrations of immune activation markers such as neopterin or tetrahydrobiopterin [43]. In contrast to the findings of Medana et al., our study we did not observe any alteration in serum Kyn-Trp ratio in comparison to the control group, nor did find any significant correlation between urinary biopterin level and the serum Kyn/Trp of P. berghei ANKA-infected mice [44]. In the mononuclear cells of infected mice, it was not neopterin but high concentration of biopterin that was found [45]. The correlation between amount of bound tetrahydrobiopterin and NOS activity clearly suggests that tetrahydrobiopterin is an essential cofactor of the enzyme. Basal tetrahydrobiopterin synthesis appears to be adequate to support iNOS activity [46]. However, when the tetrahydrobiopterin is oxidized, it leads to the increased formation of oxygen-derived radicals instead of NO by decoupled NOS [47]. Therefore, in our study, higher available NO concentration and its inhibitor effect on IDO activity should be taken into consideration in P. berghei ANKA-infected mice. The cofactor tetrahydrobiopterin is required for normal NOS function, but it is highly sensitive to oxidant stress. The decrease in iNOS activity induced by the superoxide is almost complete owing to tetrahydrobiopterin oxidation [48]. Although tetrahydrobiopterin is essential for NO production from iNOS, elevated tetrahydrobiopterin, or its oxidized product biopterin, above the basal level is not needed for excess NO synthesis [49]. Thus, our data demonstrated that oxidation of tetrahydrobiopterin is not the major mechanism of oxidative stress-induced iNOS inhibition and that iNOS induction is not associated with an IFN-γ-mediated mechanism. Mostly in murine fibroblasts, TNF-α induces guanosine triphosphate (GTP) cyclohydrolase I activity up to 30-fold, thus potentiating the biosynthesis of tetrahydrobiopterin [50]. Another difference to the human system is that murine macrophages already contain high GTP-cyclohydrolase I activities even in unstimulated conditions. This gives an enzymatic basis for the high intracellular tetrahydrobiopterin concentrations in murine macrophage cell lines [45, 51, 52]. Both exogenous and endogenous NO inhibit IDO activity and oxidative arginine and Trp metabolism in IFN γ-primed mononuclear phagocytes. However, induction of IDO activity was observed in murine macrophages when the synthesis of RNS was inhibited [53]. In our study, the urinary biopterin concentrations of mice 7 days after the inoculation of P. berghei ANKA were found to be 120% higher than those of the control animals, indicating a significant increase in tetrahydrobiopterin synthesis. The corresponding increase in serum NOx levels was 239% compared to the healthy mice. Excess NOx production may induce the formation of RNSs, which are harmful for the host.

Although the spleen is rich in macrophages and malarial infection is hematogenous in nature, splenic IDO activity was unchanged in non-CM mice and only minimally elevated in CM mice. This apparent resistance of the spleen to IDO induction may be attributed to the up-regulation of iNOS. Spleen iNOS is significantly increased in malarial infection [54]. The combination of IFN-γ and TNF-α synergizes to induce NOx synthesis in the macrophages [55]. The spleen is considered as the major site of parasite removal and splenic function is increased in acute malaria; the clearance thresholds for both mechanical filtration and Fc receptor-mediated parasitized erythrocyte clearance are lowered [56, 57]. The spleen normally removes residual host nuclear materials from erythrocytes, but how it recognizes damaged intraerythrocytic parasites is not known [58]. The spleen of non-CM patients has more phagocytes than that of CM patients. This observation reveals that the spleen plays a major role in malarial parasite clearance and is associated with host defense mechanisms against malaria [59]. The adhesion of parasitized red blood cells to spleen endothelial cells induces caspase activation, oxidative stress and apoptosis [60]. Superoxide anions can form peroxynitrites in association with NO. Parasitized red blood cell-induced apoptosis in endothelial cells is mediated through an oxidative stress pathway [61]. Under peroxidative stress, unchanged SOD level and significantly decreased GPx activity of the spleen in infected mice decrease the ability of the cell to destroy H₂O₂ through the NADPH-dependent glutathione reductase/GPx system [29]. Thus, in our study, probable diminution of GSH and related inappropriate amount of GPx could not provide a protective effect against the oxidative damage of H₂O₂.

NO inhibits IDO activity in the macrophages. Thus Sanni et al. [9] found significant increases in splenic Trp concentration above that of control mice in CM mice, as well as in non-CM mice, but plasma Trp concentrations were similar in all groups. However, we could not measure either the high IDO activity or the excessive Trp degradation products in the serum samples 7 days after the injection. Results of this study revealed that P. berghei ANKA parasitemia did not affect the serum Trp levels and IDO activity despite the high amount of biopterin on the seventh day. Our findings are in agreement with the previous observations of Nie et al. [62], who supported the notion that IFN-γ does not seem to play a major role in the control of parasite growth in the P. berghei ANKA rodent model of CM.

In conclusion, a large amount of NO released from several cellular sources seems to be ineffective against intracellular parasite clearance. Thus, P. berghei ANKA may cause multi-organ failure and death owing to host-derived severe oxidative stress and inadequate macrophage response on the seventh day. Further studies should focus on the role of NO in cell-mediated immunity and in the interaction with murine malaria.