Impairment in natural killer cells editing of immature dendritic cells by infection with a virulent Trypanosoma cruzi population

Abstract

Early interactions between natural killer (NK) and dendritic cells (DC) shape the immune response at the frontier of innate and adaptive immunity. Activated NK cells participate in maturation or deletion of DCs that remain immature. We previously demonstrated that infection with a high virulence (HV) population of the protozoan parasite Trypanosoma cruzi downmodulates DC maturation and T-cell activation capacity. Here, we evaluated the role of NK cells in regulating the maturation level of DCs. Shortly after infection with HV T. cruzi, DCs in poor maturation status begin to accumulate in mouse spleen. Although infection induces NK cell cytotoxicity and cytokine production, NK cells from mice infected with HV T. cruzi exhibit reduced ability to lyse and fail to induce maturation of bone marrow-derived immature DCs (iDCs). NK-mediated lysis of iDCs is restored by in vitro blockade of the IL-10 receptor during NK-DC interaction or when NK cells are obtained from T. cruzi-infected IL-10 knockout mice. These results suggest that infection with a virulent T. cruzi strain alters NK cell-mediated regulation of the adaptive immune response induced by DCs. This regulatory circuit where IL-10 appears to participate might lead to parasite persistence but can also limit the induction of a vigorous tissue-damaging T-cell response.

Fig. 1

Outcome of murine infection with the HV and LV T. cruzi populations. C3H/HeN mice were infected with HV and LV strains. Parasitemia levels (a), parasite burden in the spleen measured by quantitative PCR (b) and survival (c) were registered at the indicated time points. Splenocytes from acutely infected HV or LV T. cruzi strains (7 d.p.i.) and control mice were cultured with concanavalin A (ConA; 5 µg/ml) or anti-CD3 (3 µg/ml) or medium alone (d). Means ± SEM. One representative experiment of 2 using 6-10 individual mice per group is shown. a, b Mann-Whitney test. c log-rank (Mantel-Cox) test. d ANOVA (Bonferroni's post hoc test). * p < 0.05 and ** p < 0.01. T = Thymidine.

Fig. 2

Accumulation of immature splenic DC subsets during acute in vivo HV T. cruzi infection. a Total number of spleen cells in C3H/HeN mice infected with HV or LV T. cruzi populations at the indicated time points. b Dot plot representation of spleen cell suspensions from a control mouse stained for CD11c and MHC class II (Ie-K) surface markers. The dotted line represents the limit between Ie-K-positive and -negative cells. CD11c^hi and CD11c^int splenic DC subsets are plotted in R1 and R2 regions, respectively. Frequency (c) and absolute cell number (d) of CD11c^hi and CD11c^int splenic DC subsets from infected mice. Surface expression of MHC II (Ie-K) by the splenic CD11c^hi DC subset (e) and by the CD11c^int DC (f) from infected mice. Means ± SEM (a, c, f) of values obtained from 3 independent experiments with 3-5 pooled (e, f) or independent (a, c, d) mice per group. Mann-Whitney test. a, e, f * p < 0.05 HV vs. LV. c, d * p < 0.05 HV CD11c^int vs. LV CD11c^int; ^# p < 0.05, ^## p < 0.01 HV CD11c^hi vs. LV CD11c^hi.

Fig. 3

Frequency, activation and cytokine production by NK cells following T. cruzi infection. C3H/HeN mice were infected with the HV and LV T. cruzi populations and sham infected. PBMC (blood) and spleen cell suspensions (spleen) obtained at different time points after infection. Frequency of DX5+/CD3- cells and expression of CD69 by DX5+/CD3- cells were evaluated in PMBC (a) and spleen cell suspensions (b). Ex vivo intracellular production of IFN-γ and IL-10 by DX5+/CD3- cells from PBMC fractions (c) and spleen cell suspensions (d) 18 h.p.i. Dot plots show results from 1 representative experiment of 3 each using a pool of 3-5 mice per group. Bar graphs show results obtained from 3 independent experiments with 3-5 pooled mice per group (means ± SEM). ANOVA (Bonferroni's post hoc test). * p < 0.05 sham vs. LV and HV, ^# p < 0.05 HV vs. sham and LV.

Fig. 4

NK cell-mediated lysis of YAC-1 cells and iDCs upon T. cruzi infection. a Lysis of YAC-1 cells by splenic NK cells from T. cruzi HV, LV or sham-infected (18 h.p.i.) and LPS (100 µg/mouse) inoculated C3H/HeN mice was evaluated. YAC-1 cells (5 × 10⁴) were cultured with freshly purified splenic NK cells at a NK/YAC-1 ratio of 5/1. Left panel shows CFSE-labeled YAC-1 cells and middle panel shows PI uptake by CFSE+ cells gated in R. Results from a representative experiment with 3-5 pooled mice per group. Right panel show results of 3 independent experiments with 3-5 pooled mice per group (means ± SEM). ANOVA (Bonferroni's post hoc test). * p < 0.05 sham vs. LV and HV. b iDCs (5 × 10⁴) were cultured with freshly purified spleen NK cells from HV, LV and sham-infected (18 h.p.i.) C3H/HeN mice. Left panel shows CFSE-labeled iDCs and middle panel shows PI uptake by CFSE+ cells gated in R. Results from a representative experiment with 3-5 pooled mice per group. Right panel shows results of 3 independent experiments with 3-5 pooled mice per group (means ± SEM). ANOVA (Bonferroni's post hoc test) * p < 0.05 sham vs. LV and HV.

Fig. 5

IL-10 mediates NK cell cytotoxicity on iDCs. Mice were infected with HV T. cruzi. a iDCs (5 × 10⁴ per well) were cultured with freshly purified spleen NK cells (NK/DC ratio: 2/1) from HV (■) and sham-infected (□) C3H/HeN male mice (18 h.p.i.) in the presence of an IL-10R blocking antibody (Ab; 40 µg/ml, clone 1B1.3A) or rat IgG isotype control (HPRN). Results are expressed as percent lysis from 2 independent experiments with 4-5 pooled mice per group (means ± SEM). ANOVA (Bonferroni's post hoc test). * p< 0.05 vs. sham isotype. b iDCs (5 × 10⁴/well) were cultured with freshly purified spleen NK cells (NK/DC ratio: 2/1) from IL-10KO mice of Balb/c background or wild-type (WT) mice (18 h.p.i.). Results are expressed as percent lysis from 2 independent experiments with 4-5 pooled mice per group (means ± SEM). ANOVA (Bonferroni's post hoc test). * p< 0.05 vs. HV WT.