IL-33-mediated protection against experimental cerebral malaria is linked to induction of type 2 innate lymphoid cells, M2 macrophages and regulatory T cells

Abstract

Cerebral malaria (CM) is a complex parasitic disease caused by Plasmodium sp. Failure to establish an appropriate balance between pro- and anti-inflammatory immune responses is believed to contribute to the development of cerebral pathology. Using the blood-stage PbA (Plasmodium berghei ANKA) model of infection, we show here that administration of the pro-Th2 cytokine, IL-33, prevents the development of experimental cerebral malaria (ECM) in C57BL/6 mice and reduces the production of inflammatory mediators IFN-γ, IL-12 and TNF-α. IL-33 drives the expansion of type-2 innate lymphoid cells (ILC2) that produce Type-2 cytokines (IL-4, IL-5 and IL-13), leading to the polarization of the anti-inflammatory M2 macrophages, which in turn expand Foxp3 regulatory T cells (Tregs). PbA-infected mice adoptively transferred with ILC2 have elevated frequency of M2 and Tregs and are protected from ECM. Importantly, IL-33-treated mice deleted of Tregs (DEREG mice) are no longer able to resist ECM. Our data therefore provide evidence that IL-33 can prevent the development of ECM by orchestrating a protective immune response via ILC2, M2 macrophages and Tregs.

Figure 1. IL-33 protects mice from PbA-induced cerebral malaria.

C57BL/6 mice were infected i.v. with Plasmodium berghei ANKA (PbA) (10⁴ pRBCs) and injected i.p. daily with PBS or IL-33 (0.2 μg) from day 0. (A) Kaplan–Meier survival curves (pool of 3 experiments, n = 14–20 per group). (B) body weight loss (n = 5 per group); Statistical differences are shown for day 7. (C) Clinical score (n = 5 per group). The hatched area indicates ECM related scores. (D) Time-course analysis of parasitemia (n = 5 per group). (E) Parasite-derived bioluminescence. One representative mouse is shown at each time point. Radiance (P/Sec/cm²/Sr), color scale Min = 2×10⁶, Max = 5×10⁷. (F) Mean body bioluminescence (n = 5 per group). (G) Mice infected with PbA-Luc were sacrificed on day 7 and bioluminescence of the brain was recorded. (H) Mean brain bioluminescence per group (n = 6). (I) Representative H&E histopathology of brain vasculature from PBS- or IL-33-treated mice. Magnification ×400. (J) Severity of brain microvascular obstruction and local hemorrhage assessed from a whole-brain section (n = 6 per group). (K) Icam-1 mRNA expression (relative to Hprt1) in the brain determined by qPCR (n = 6 per group); Data are mean ± SEM. Representative data from 3 independent experiments were shown. ns, non significant, ***P<0.001.

Figure 2. IL-33 reduces pro-inflammatory cytokines and chemokines in PbA-infected mice.

C57BL/6 mice were not infected (NI) or infected with PbA and injected i.p. daily with PBS (open columns) or IL-33 (filled columns) from day 0 for 5 consecutive days. (A-H) Serum IFN-γ, IL-12, TNF-α, IL-5, IL-10, IP-10, KC and MCP-1 concentrations were determined by ELISA or Multiplex. Mean ± SEM (n = 5–7 per group). (I-J) Expression of Tbet and Gata3 mRNA (relative to Hprt1) in purified splenic CD4⁺ T cells was analysed by qPCR (n = 3 per group). Intracellular staining of Granzyme B (GrmB) in splenic CD8⁺ (K) and CD4⁺ (L) T cells on day 6. Data are mean ± SEM (n = 5–7 per group). ns, not significant, *P<0.05, **P<0.01, ***P<0.001 by two-tailed ANOVA.

Figure 3. IL-33 expands ILC2 which adoptively protect mice against ECM.

C57BL/6 mice were not infected (NI) or infected with PbA and treated daily with PBS or IL-33 from day 0. Splenic ILC2 were analysed by FACS on day 3. (A) Representative FACS showing % of ICOS⁺ST2⁺ cells gated on Lin⁻ CD45⁺ cells. Cumulative percentage (B) and number (C) of ILC2 per spleen are shown. (D) Percentage of Ki67⁺ ILC2. Data are mean ± SEM (n = 5 per group) and representative of 2 independent experiments. *P<0.05, **P<0.01, ***P<0.001 by two-tailed ANOVA. (E-F) Total spleen cells were restimulated ex vivo for 4 h with PMA/ionomycin and stained for intracellular cytokines. (E) Dot plots show expression of IL-4, IL-5 and IL-13 by ILC2 (gated on live Lin⁻CD45⁺ICOS⁺ST2⁺ cells) on day 3 post-infection. (F) Cumulative data shows mean ± SEM of cytokine production (n = 5 per group). Data are representative of 2 independent experiments. (G-K) ILC2 sorted from mice pre-treated with IL-33 were adoptively transferred (2×10⁶ cells, i.v., on day −1) into naïve C57BL/6 mice which were infected with PbA (10⁴ pRBCs, i.v., on day 0). The recipients were given 2 injections of IL-33 (0.2 μg, i.p.) 30 min and 24 h after cell transfer. Survival (G), body weight loss (H) and parasitemia (I) were assessed daily. ***P<0.001 between PbA and PbA+IL-33+ILC2 groups; ^#P<0.05 between PbA and PbA+IL-33 groups. (J) Representative H&E staining of brain sections on day 7, magnification ×400. (K) Vessel obstruction scores in the brain (same colour code as in F). Data are mean ± SEM (n = 5 per group). ***P<0.001 by two-tailed ANOVA.

Figure 4. IL-33 polarizes M2 macrophages in PbA-infected mice.

C57BL/6 mice were not infected (NI) or infected i.v. with PbA and treated daily for 5 days with PBS or IL-33 and spleen cells were harvested at indicated time points and analysed. (A) Representative dotplots showing percentage of CD11b⁺F4/80⁺ cells among total splenocytes. (B) Percentages and (C) numbers of CD11b⁺F4/80⁺CD11c⁻ macrophages in the spleen. (D) Mean Fluorescence Intensity (MFI) of CD206, CD86, MHC-II and CD40 on macrophages (n = 5 per group). (E) Expression (relative to Hprt1) of Arg-1, Ym1, Fizz1, Nos2 and Hmox-1 mRNA in the spleen on day 5. Data are mean ± SEM, representative of 2 independent experiments. ns, not significant, *P<0.05, **P<0.01, ***P<0.001 by two-tailed ANOVA.

Figure 5. ILC2 promote M2 macrophage polarization.

(A) BMDM from C57BL/6 mice were cultured in the lower chamber of a 24-transwell plate in complete medium alone (M0) or supplemented with IL-4 (M2). In some experiments ILC2, sorted from naïve WT mice pre-treated with IL-33, were added to the upper chamber. After 48 h, BMDM were collected and assayed for the expression of M2 markers by qPCR (relative to Hprt1). (B) ST2-deficient BMDM were co-cultured in transwell plates as above with WT ILC2 in the presence of IL-33 alone or in combination with IL-7. After 48 h, BMDM were collected and assayed for the expression of M2 markers by qPCR (relative to Hprt1). Type 2 cytokines in the supernatants of ILC2 cultured in the presence of IL-33 or IL-33 + IL-7 were determined by ELISA (C), or by FACS (D, E). Data are mean ± SEM (n = 3 per group), representative of two independent experiments, *P<0.05, **P<0.01, ***P<0.001 by two-tailed ANOVA. (F) ILC2 sorted from mice pre-treated with IL-33 were adoptively transferred (2×10⁶ cells, i.v., on day −1) into naïve C57BL/6 mice which were infected with PbA (10⁴ pRBCs, i.v., on day 0). The recipients, were given 2 injections of IL-33 (0.2 μg, i.p.) 30 min and 24 h after cell transfer. Expression of Arg-1, Ym1 and Fizz1 mRNA in the spleen was measured by qPCR (relative to Hprt1) on day 7. Data are mean ± SEM (n = 5 per group) *P<0.05, **P<0.01, ***P<0.001 by two-tailed ANOVA.

Figure 6. IL-33, ILC2 and M2 induce Tregs.

(A-C) C57BL/6 mice were infected i.v. with PbA and treated daily with PBS or IL-33 from day 0. (A) Relative expression (% of Hprt1) of Foxp3⁺ mRNA in purified CD4⁺ cells from the spleen. (B) percentage of Foxp3⁺ cells gated on CD4⁺ cells and number of Foxp3⁺CD4⁺ cells in the spleen. Data are mean ± SEM (n = 5 per group), representative of at least 2 independent experiments. *P<0.05, **P<0.01, ***P<0.001 by two-tailed ANOVA (C-D) ILC2 sorted from mice pre-treated with IL-33 were adoptively transferred (4×10⁶ cells, i.v., on day −1) into naïve C57BL/6 mice which were infected with PbA (10⁴ pRBCs, i.v., on day 0). The recipients were given 2 injections of IL-33 (0.2 μg, i.p.) 30 min and 24 h after cell transfer. Spleen cells were harvested on day 7 and analysed by FACS for Foxp3 expression gated on CD4⁺ cells. Data are mean ± SEM (n = 5 per group). *P<0.05, **P<0.01, ***P<0.001 by two-tailed ANOVA. (E, F) M2 induce Tregs in vitro. (E) CD4⁺CD25⁺ T cells from naïve C57BL/6 mice were co-cultured with or without BMDM-derived M2 for 2 days and analysed for Foxp3 expression by FACS gated on live CD4⁺ cells. (F) CD4⁺CD25⁻ T cells from naïve C57BL6 mice were cultured under iTreg polarizing conditions in the presence or absence of BMDM-derived M2 for 2 days and Foxp3 expression was analysed by FACS gated on live CD4⁺ cells. Data are mean ± SEM of 3 experiments. *P<0.05 by two-tailed ANOVA.

Figure 7. IL-33-mediated protection against ECM is Treg-dependent.

DEREG mice were infected with PbA (10⁵ pRBC, i.p.) and daily treated with PBS or IL-33 for 5 consecutive days. Where indicated, mice received diphteria toxin (DT, 1 μg, i.p.) every second day from day 1. (A) Treg depletion was assessed in blood leukocytes on day 5 of infection by FACS on GFP-Foxp3 gated on CD4⁺ T cells, representative of 5 mice. (B) Survival (Kaplan–Meier survival curves) and (C) clinical score (hatched area indicates ECM related scores) of DEREG mice treated with PBS or IL-33 with or without DT (n = 4–5 per group) are shown. (D-E) Percentage of IFN-γ⁺ and Granzyme B⁺ (GrmB) CD8⁺ T cells in the spleen on day 5 post-infection was determined by FACS. (F) Serum IFN-γ and IL-12 concentrations were determined by ELISA (day 5) (n = 5 per group). Data are mean ± SEM representative of 2 experiments, *P<0.05, **P<0.01, ***P<0.001 by two-tailed ANOVA.

Figure 8. Schematic representation of the pathways by which IL-33 attenuates ECM.

Arrows represent activation/promotion; dotted blunted arrows = proposed mechanism of suppression.