TRPV1 Contributes to Cerebral Malaria Severity and Mortality by Regulating Brain Inflammation

Abstract

Transient receptor potential vanilloid 1 (TRPV1) is a Ca⁺²-permeable channel expressed on neuronal and nonneuronal cells, known as an oxidative stress sensor. It plays a protective role in bacterial infection, and recent findings indicate that this receptor modulates monocyte populations in mice with malaria; however, its role in cerebral malaria progression and outcome is unclear. By using TRPV1 wild-type (WT) and knockout (KO) mice, the importance of TRPV1 to this cerebral syndrome was investigated. Infection with Plasmodium berghei ANKA decreased TRPV1 expression in the brain. Mice lacking TRPV1 were protected against Plasmodium-induced mortality and morbidity, a response that was associated with less cerebral swelling, modulation of the brain expression of endothelial tight-junction markers (junctional adhesion molecule A and claudin-5), increased oxidative stress (via inhibition of catalase activity and increased levels of H₂O₂, nitrotyrosine, and carbonyl residues), and diminished production of cytokines. Plasmodium load was not significantly affected by TRPV1 ablation. Repeated subcutaneous administration of the selective TRPV1 antagonist SB366791 after malaria induction increased TRPV1 expression in the brain tissue and enhanced mouse survival. These data indicate that TRPV1 channels contribute to the development and outcome of cerebral malaria.

Figure 1

Brain TRPV1 mRNA expression and cerebral malaria progression. (a) TRPV1 mRNA expression in brain samples of noninfected and infected (at stage III/IV) TRPV1 wild-type (WT) mice. Disease progression (b) and stage (c); survival rates (d) and body temperature (e) recordings from TRPV1 WT and knockout (KO) mice infected with P. berghei ANKA. Disease progression, stage, and survival rates were registered over 14 days postinfection. Mouse body temperatures were evaluated at baseline and postmalaria induction (at stage III/IV or at day 14 for those that survived the observation period). Results represent the mean  ±  SEM of all mice per group, obtained from two-three independent experiments. n is indicated on each graph. Data were analysed by repeated-measures analysis of variance (ANOVA) followed by the Bonferroni test with FDR correction (panels b and c). Paired and unpaired t tests were used when appropriate (panels a and e). Survival curves were analysed by the nonparametric Mantel-Cox test (panel d). ^∗ p < 0.05 differs from noninfected WTs or baseline readings; ^# p < 0.05 differs from infected WT mice.

Figure 2

Blood and brain parasitaemia. (a) Blood parasitaemia and (b) brain 18S levels in TRPV1 wild-type (WT) and knockout (KO) mice infected with P. berghei ANKA. Blood parasitaemia data was collected over 14 days postinfection; brain samples were collected at stage III/IV or at day 14 for those that survived the observation period. Results represent the mean  ±  SEM of all mice per group, obtained from two-three independent experiments. n is indicated on each graph. Data were analysed by repeated-measures analysis of variance (ANOVA) followed by the Bonferroni test with FDR correction (panel a). Unpaired t test was used when appropriate (panel b).

Figure 3

Brain swelling and expression of blood brain barrier integrity markers. (a) Brain weight/body weight ratios and mRNA expression levels of claudin-5 (b) and JAM-A (c) in brain samples of TRPV1 wild-type (WT) and knockout (KO) mice infected with P. berghei ANKA. Brain samples were collected at stage III/IV or at day 14 for those that survived the observation period. Samples from noninfected mice were used as controls. Results represent the mean  ±  SEM of all mice per group, obtained from three independent experiments. n is indicated on each graph. Data were analysed by one-way analysis of variance (ANOVA) followed by the Bonferroni test with FDR correction. ^∗ p < 0.05 differs from noninfected WTs; ^# p < 0.05 differs from infected WT mice.

Figure 4

Levels of H₂O₂, protein nitrotyrosine, and carbonyl residues. (a) H₂O₂ concentrations, protein (b) nitrotyrosine and (c) carbonyl residues in brain samples obtained from TRPV1 wild-type (WT) and knockout (KO) mice infected or not with P. berghei ANKA. (d) H₂O₂ concentrations, protein (e) nitrotyrosine, and (f) carbonyl residues in plasma samples obtained from TRPV1 wild-type (WT) and knockout (KO) mice infected or not with P. berghei ANKA. Samples were collected at stage III/IV or at day 14 for those that survived the observation period. Results represent the mean  ±  SEM of all mice per group, obtained from three independent experiments. n is indicated on each graph. Data were analysed by one-way analysis of variance (ANOVA) followed by the Bonferroni test with FDR correction. ^∗ p < 0.05 differs from noninfected WTs; ^# p < 0.05 differs from infected WT mice.

Figure 5

Activity levels of antioxidant enzymes. (a) Superoxide dismutase (SOD), (b) catalase, (d) glutathione peroxidase (GPx), (e) glutathione reductase (GR), and (f) thioredoxin reductase (TrxR) activity levels in brain samples obtained from TRPV1 wild-type (WT) and knockout (KO) mice infected or not with P. berghei ANKA. (c) Activity levels of catalase in plasma samples of infected TRPV1 WT and KO mice. Samples were collected at stage III/IV or at day 14 for those that survived the observation period. Results represent the mean  ±  SEM of all mice per group, obtained from three independent experiments. n is indicated on each graph. Data were analysed by one-way analysis of variance (ANOVA) followed by the Bonferroni test with FDR correction (panels a, b, d, e, and f). Unpaired t test was used when appropriate (panel c). ^∗ p < 0.05 differs from noninfected WTs; ^# p < 0.05 differs from infected WT mice.

Figure 6

Brain and circulating levels of cytokines. Brain levels of (a) tumor necrosis α (TNFα), (b) interleukin-6 (IL-6), and (c) interferon γ (IFNγ) and plasma concentrations of (d) TNFα, (e) IL-6, and IFNγ (f) in TRPV1 wild-type (WT) and knockout (KO) mice infected with P. berghei ANKA. Samples were collected at stage III/IV or at day 14 for those that survived the observation period. Results represent the mean  ±  SEM of all mice per group, obtained from three independent experiments. n is indicated on each graph. Data were analysed by unpaired t test. ^# p < 0.05 differs from infected WT mice.

Figure 7

Brain and vascular changes in cerebral malaria in TRPV1 WT and KO mice. (a) Several alterations occur in the brain tissue and vasculature during cerebral malaria. Wild-type (WT) red blood cells (RBC) infected with Plasmodium berghei ANKA reach the brain vasculature and trigger the accumulation of leukocytes in the vascular space. As a result of this close interaction between infected RBC, leukocytes, and the endothelium, oxidative stress products (H₂O₂, nitrosylated and carbonylated proteins) and cytokines (TNFα, IL-6, and IFNγ) are detected in the circulation and in the brain tissue; H₂O₂ levels are a lot higher in the brain tissue in comparison with the circulation. Plasma extravazation is increased in the brain and this is associated with reduced mRNA expression of the tight-junction endothelial markers claudin-5 and JAM-A. These alterations may culminate with neuronal death, thus, contributing to the increased morbidity and mortality observed in WT mice following infection with P. berghei ANKA. (b) Infected mice lacking TRPV1 (TRPV1KO) present increased levels of H₂O₂ and nitrosylated and carbonylated proteins than WT animals at both brain tissue and circulation. TRPV1KOs also exhibit lower concentrations of plasma and brain cytokines, especially TNFα and IL-6, and less plasma extravazation than WT mice, a response that is accompanied by higher expression of claudin-5 and JAM-A in their brain vasculature. The inflammatory response profile observed in TRPV1KO mice may reflect in less neuronal damage, as these animals are protected from P. berghei ANKA-induced death and symptoms.