Tip-DC development during parasitic infection is regulated by IL-10 and requires CCL2/CCR2, IFN-gamma and MyD88 signaling

Abstract

The development of classically activated monocytic cells (M1) is a prerequisite for effective elimination of parasites, including African trypanosomes. However, persistent activation of M1 that produce pathogenic molecules such as TNF and NO contributes to the development of trypanosome infection-associated tissue injury including liver cell necrosis in experimental mouse models. Aiming to identify mechanisms involved in regulation of M1 activity, we have recently documented that during Trypanosoma brucei infection, CD11b(+)Ly6C(+)CD11c(+) TNF and iNOS producing DCs (Tip-DCs) represent the major pathogenic M1 liver subpopulation. By using gene expression analyses, KO mice and cytokine neutralizing antibodies, we show here that the conversion of CD11b(+)Ly6C(+) monocytic cells to pathogenic Tip-DCs in the liver of T. brucei infected mice consists of a three-step process including (i) a CCR2-dependent but CCR5- and Mif-independent step crucial for emigration of CD11b(+)Ly6C(+) monocytic cells from the bone marrow but dispensable for their blood to liver migration; (ii) a differentiation step of liver CD11b(+)Ly6C(+) monocytic cells to immature inflammatory DCs (CD11c(+) but CD80/CD86/MHC-II(low)) which is IFN-gamma and MyD88 signaling independent; and (iii) a maturation step of inflammatory DCs to functional (CD80/CD86/MHC-II(high)) TNF and NO producing Tip-DCs which is IFN-gamma and MyD88 signaling dependent. Moreover, IL-10 could limit CCR2-mediated egression of CD11b(+)Ly6C(+) monocytic cells from the bone marrow by limiting Ccl2 expression by liver monocytic cells, as well as their differentiation and maturation to Tip-DCs in the liver, showing that IL-10 works at multiple levels to dampen Tip-DC mediated pathogenicity during T. brucei infection. A wide spectrum of liver diseases associates with alteration of monocyte recruitment, phenotype or function, which could be modulated by IL-10. Therefore, investigating the contribution of recruited monocytes to African trypanosome induced liver injury could potentially identify new targets to treat hepatic inflammation in general, and during parasite infection in particular.

Figure 1. CD11b⁺Ly6C⁺ monocytic cells are affected by CCR2 but not CCR5 or Mif signaling during T. brucei infection.

A) Inflammatory monocytic cells gated based on their expression of CD11b and Ly6C from non-parenchymal cells isolated from the liver of WT mice on day 6 of T. brucei infection were stained for CCR2, CCR5 or CD74 expression (filled grey curves compared to dotted line curves representing specific isotype controls) FACS profiles are representative of one of four animals tested in two independent experiments. B) Percentage and D) Number of CD11b⁺Ly6C⁺ monocytic cells in liver non-parenchymal, blood and bone marrow of CCR2 KO, CCR5 KO and Mif KO mice on day 6 post infection. Data are shown as mean ± SEM of three individual mice of one of three independent experiments performed. C) Non-parenchymal cells isolated from liver, blood and bone marrow of WT and CCR2 KO mice on day 6 post infection were assayed for co-expression of CD11b and Ly6C. Percentages of CD11b⁺Ly6C⁺ monocytic cells within the indicated gate are shown. FACS profiles are representative of one of nine animals tested in three independent experiments. *, significantly lower (p<0.01) and **, significantly higher (p<0.05) compared to WT mice.

Figure 2. CCR2 does not contribute to liver extravasation of CD11b⁺Ly6C⁺ monocytic cells during T. brucei infection.

WT and CCR2 KO CD11b⁺Ly6C⁺ monocytic cells were isolated from bone marrow of mice on day 6 of infection, labeled with PKH26 and CellVue kits, respectively, and injected at a 1-1 ratio in the blood of recipient WT mice on day 6 of T. brucei infection. Left panel shows percentages of gated, labeled WT and CCR2 KO cells before injection. 24 hours after transfer liver non-parenchymal were harvested from recipient mice and analysed for the presence of labeled cells. Percentages of gated, labeled WT and CCR2 KO cells are shown (right panel). FACS profiles are representative of one of six animals tested in two independent experiments.

Figure 3. CCR2, IL-10R, MyD88 and IFN-γ signaling differentially contribute to Tip-DC differentiation/maturation during T. brucei infection.

A) Liver non-parenchymal cells were isolated from WT, CCR2 KO, anti-IL-10R treated WT, MyD88 KO and IFN-γ KO mice on day 6 of infection. CD11b⁺Ly6C⁺ monocytic cells were gated (as indicated in figure 1C) and assayed for co-expression of CD11b and CD11c (top panels), TNF (middle panels) or iNOS (bottom panels). Percentages of CD11c⁺, TNF⁺ or iNOS⁺ cells (rectangular gate) within CD11b⁺Ly6C⁺ monocytic cells are indicated. Corresponding percentages of CD11c⁺, TNF⁺ or iNOS⁺ cells within the total number of living non-parenchymal cells recovered from the liver are indicated between brackets. Data are shown as mean ± SEM of three individual mice of one of three independent experiments performed. FACS profiles are representative of one of nine animals tested in three independent experiments performed. B) CD11b⁺Ly6C⁺ monocytic cells were isolated from bone marrow of WT and MyD88 KO or IFN-γR KO mice on day 6 of T. brucei infection, labeled with PKH26 (WT) and CellVue kits (KO), and injected at a 1-1 ratio in the blood of recipient WT mice on day 6 of T. brucei infection. Twenty four hours after transfer, liver non-parenchymal were harvested from recipient mice and analyzed for the presence of labeled cells. Percentages of gated, labeled WT and MyD88 or IFN-γR KO cells after transfer are shown (left panels). The expression of MHC-II molecules on the gated labeled cells is shown in the right panels. Data are shown as mean ± SEM of three individual mice of one of three independent experiments performed. FACS profiles are representative of one of four animals tested in three independent experiments. Note that due to the low numbers of transferred cells recovered from the liver, we were unable to assess iNOS or TNF expression on transferred cells using intracellular staining assays. *, significantly lower (p<0.05) and **, significantly higher (p<0.05) compared to WT mice.

Figure 4. Absence of CCR2 signaling reduces TNF, NO and IFN-γ production during T. brucei infection.

At day 6 post infection, A) production of TNF, NO₂ (as measure of NO) and IFN-γ by unstimulated liver non-parenchymal cells after 24 hours of in vitro culture and B) percentages of CD4⁺IFN-γ⁺ and CD8⁺IFN-γ⁺ cells within the liver non-parenchymal cell fraction were determined in WT (white bars) and CCR2 KO (grey bars) mice. Data are shown as mean ± SEM of three individual mice of one of three independent experiments performed. *, significantly (p<0.05) lower compared to WT mice.

Figure 5. Absence of CCR2 signaling reduces liver pathogenicity and prolongs survival of T. brucei infected mice.

A) Serum ALT levels were measured in WT and CCR2 KO mice on day 28 of infection (mean ± SEM of three individual mice of one of two independent experiments performed). *, significantly (p<0.05) lower compared to WT mice. B) Survival time of infected WT and CCR2 KO mice. Data are representative of one of two independent experiments performed. *, significantly longer (p<0.05) compared to WT mice. C) Microscopic examination (H&E staining; magnification ×100) of liver sections from WT and CCR2 KO mice on day 28 of infection (representative of 3 animals tested). In CCR2 KO mice hepatocyte necrosis is low, whereas in WT mice important hepatitis and fields of necrosis are observed (arrows).

Figure 6. Transfer of CD11b⁺Ly6C⁺ TIP-DC precursors increases pathogenicity during T. brucei infection.

CD11b⁺Ly6C⁺ monocytic cells were isolated from the bone marrow of WT or TNF KO mice on day 6 of T. brucei infection, labeled with CellVue and transferred intravenously into infected recipient CCR2 KO mice. Twenty four hours after transfer blood serum was analyzed for TNF concentration and ALT activity. Data are mean ± SEM of three individual mice of one of three independent experiments performed. Note that due to the low numbers of transferred cells recovered from the liver, we were unable to assess iNOS or TNF expression on transferred cells using intracellular staining assays. *, significantly higher (p<0.05) compared to CCR2 KO mice that did not receive CD11b⁺Ly6C⁺ monocytic cells.

Figure 7. Anti-IL-10R treatment increases CD11b⁺Ly6C⁺ monocytic cell percentages and serum CCL2 concentration during T. brucei infection.

WT mice were treated with neutralizing anti-IL-10R antibody or control antibody on days 7 and 9 of infection and euthanized on day 10. A) Percentages of CD11b⁺Ly6C⁺ monocytic cells in the non-parenchymal liver cell fraction, blood and bone marrow of WT (white bars) and CCR2 KO (grey bars) mice with or without anti-IL-10R treatment. B) CCL2 protein concentration in blood sera of WT T. brucei infected mice. Data are the mean ± SEM of three individual mice tested in one of three independent experiments performed. *, significantly (p<0.05) higher compared to control treated WT mice.

Figure 8. Model for migration, differentiation and maturation of CD11b⁺Ly6C⁺ monocytes to Tip-DCs during T. brucei infection.