Impaired innate immunity in mice deficient in interleukin-1 receptor-associated kinase 4 leads to defective type 1 T cell responses, B cell expansion, and enhanced susceptibility to infection with Toxoplasma gondii

Abstract

Interleukin-1 receptor (IL1R)-associated kinase 4 (IRAK4) is a member of the IRAK family and has an important role in inducing the production of inflammatory mediators. This kinase is downstream of MyD88, an adaptor protein essential for Toll-like receptor (TLR) function. We investigated the role of this kinase in IRAK4-deficient mice orally infected with the cystogenic ME49 strain of Toxoplasma gondii. IRAK4(-/-) mice displayed higher morbidity, tissue parasitism, and accelerated mortality than the control mice. The lymphoid follicles and germinal centers from infected IRAK4(-/-) mice were significantly smaller. We consistently found that IRAK4(-/-) mice showed a defect in splenic B cell activation and expansion as well as diminished production of gamma interferon (IFN-γ) by T lymphocytes. The myeloid compartment was also affected. Both the frequency and ability of dendritic cells (DCs) and monocytes/macrophages to produce IL-12 were significantly decreased, and resistance to infection with Toxoplasma was rescued by treating IRAK4(-/-) mice with recombinant IL-12 (rIL-12). Additionally, we report the association of IRAK4 haplotype-tagging single nucleotide polymorphisms (tag-SNPs) with congenital toxoplasmosis in infected individuals (rs1461567 and rs4251513, P < 0.023 and P < 0.045, respectively). Thus, signaling via IRAK4 is essential for the activation of innate immune cells, development of parasite-specific acquired immunity, and host resistance to infection with T. gondii.

Fig 1

IRAK4 mediates control of parasite replication and influences disease outcome during primary infection with Toxoplasma gondii. Groups of IRAK4^−/−, IRAK4^+/−, and 129/Ola mice were orally infected with 20 cysts of the ME49 strain. Mortality (A), morbidity (B), and weight loss (C) (different letters in parentheses indicate P of < 0.0001) were assessed for 30 days after infection. Numbers within the parentheses indicate the number of animals used in the experiment shown in panels A and C. In panel B, each dot at one time point corresponds to one animal. Morbidity was scored as follows: bright and active, 0; ruffled coat, 1, 10% weight loss, 2, hunched, tottering gait, staring, stiff coat, 3; reluctance to move and 20% weight loss, 4, and death, 5. Data are representative of two independent experiments that yielded similar results. (D) Tissue cysts in IRAK4^−/− (left panels), IRAK4^+/− (middle panels), and 129 (right panels) mice are indicated with black arrowheads. Black arrows indicate the absence or presence of perivascular inflammatory infiltrates in IRAK4^−/−, IRAK4^+/−, and 129 mice. The asterisk indicates an area of moderate necrosis in the brain tissue of IRAK4^−/− mice. (E) Cyst counts in histological sections of the brain from IRAK4^−/− and IRAK4^+/− mice infected with T. gondii (left) and quantification of parasite DNA in the livers and spleens of IRAK^−/− (middle) and IRAK^+/− (right) mice at 7 and 9 days after infection. The presented results are an average of data from eight mice per group, and the P values were calculated for a 95% confidence interval. The experiments were repeated three times with similar results.

Fig 2

Altered structures of follicles and germinal centers in the spleens of IRAK4^−/− mice infected with T. gondii. (A) Spleens from mice at 7 days postinfection were removed and stained with hematoxylin and eosin for histological evaluation. Black arrowheads indicate activated and nonactivated germinal centers in spleens from IRAK4^−/− (upper panel) and IRAK4^+/− (bottom panel) mice, respectively. Bar graphs show the quantification of follicles and germinal center areas in spleens from IRAK4^−/− and IRAK4^+/− mice. (B) Dot plot graphs (frequency) and bar graphs (absolute number) show B cells in proliferation (CD19⁺ Ki67⁺) from IRAK4^+/− and IRAK4^−/− mice (left and bottom right graphs). (C) The contour plots on the left and the graphs show, respectively, the frequency and total numbers of activated (i.e., CD86⁺ or CD69⁺) B lymphocytes (CD19⁺). The results from uninfected and infected mice are represented with white and black bars, respectively. The presented results are an average of data from four mice per group, and the P values were calculated for a 95% confidence interval. The experiments were repeated three (A) or two (B and C) times with similar results.

Fig 3

Impaired cytokine response in IRAK4^−/− mice infected with T. gondii. IRAK^−/− and IRAK^+/− mice were orally infected with the ME49 strain of T. gondii. (A) The levels of cytokines in serum were evaluated at 7 days postinfection by enzyme-linked immunosorbent assay (ELISA) and cytometric bead array (CBA). The P values were calculated for a 95% confidence interval. (B) Splenocytes from IRAK^−/− and IRAK^+/− mice were harvested at 7 days postinfection and cultured for 48 h without the addition of exogenous stimuli. Cytokine levels were measured in the supernatants by ELISA and CBA. The results of uninfected and infected mice are represented in white and black, respectively. The presented results are an average of data from eight mice per group, and the P values were calculated for a 95% confidence interval. The experiments were repeated three times with similar results.

Fig 4

IRAK4-dependent IL-12 and TNF-α production by macrophages from mice infected with T. gondii. Macrophages of IRAK4^+/− and IRAK4^−/− mice were analyzed at 7 days after infection with T. gondii. Representative dot plots of the frequency (A) and bar graphs (C, top graphs) showing the average number of different subsets of myeloid cells in spleens from IRAK4^+/− and IRAK4^−/− mice, based on the levels of F4/80, CD11b, and GR1 expression (B). The percentages of CD11b^hi F4/80^lo GR1^hi, CD11b^lo F4/80^hi GR1^int, and CD11b^int F4/80^lo GR1^lo cells producing IL-12 and TNF-α are shown in panel C (middle and bottom graphs, respectively). The results of uninfected and infected mice are represented by white and black bars, respectively. The presented results are an average of data from four mice per group, and the P values were calculated for a 95% confidence interval. The experiments were repeated three times with similar results. NS, not stimulated.

Fig 5

IRAK4-dependent production of IL-12 by dendritic cells from mice infected with T. gondii. Dendritic cells (DCs) were analyzed 7 days after infection with T. gondii. (A) Representative dot plots (top graphs) show the frequency of DCs (CD11⁺, MHC-II^hi) in spleens from IRAK4^+/− and IRAK4^−/− mice. Representative dot plots of the frequency of DCs producing IL-12 or TNF-α are shown in the middle and bottom graphs. Bar graphs showing the average number of DCs (CD11⁺, MHC-II^hi) (top right graph) and the frequency of DCs producing IL-12 (middle right graph) or TNF-α (bottom-right graph) in spleens from IRAK4^+/− and IRAK4^−/− mice. (B) The expression (median fluorescence intensity [MFI]) of activation of the markers CD86 and CD40. (C) Survival curve of the deficient mice treated with recombinant IL-12 or not treated during 7 days of T. gondii infection. The results for uninfected and infected mice are represented by the white and black bars, respectively. The presented results are an average of data from four mice per group, and the P values were calculated for a 95% confidence interval. The experiments were repeated three (A and B) or two (C) times with similar results.

Fig 6

Impairment of the cell number and IFN-γ-producing T cells in IRAK4-deficient mice. (A) Representative dot plot on the left shows the frequency of CD4⁺ and CD8⁺ T cells, within the CD3⁺ lymphocytes. Bar graphs on the right show the absolute number of CD4⁺ T as well as CD8⁺ T lymphocytes from IRAK4^+/− and IRAK4^−/− mice infected or not infected with T. gondii. (B) Dot plots illustrating the IFN-γ-producing CD4⁺ and CD8⁺ T lymphocytes stimulated with STAg from IRAK4^−/− mice (lower panels) compared to IRAK4^+/− mice (upper panels) infected with T. gondii. Bar graphs show the frequency of IFN-γ-producing CD4⁺ and CD8⁺ T lymphocytes cultured with STAg. The presented results are an average of data from four mice per group, and the P values were calculated for a 95% confidence interval. The experiments were repeated three times with similar results.

Fig 7

Association of SNPs in the IRAK4 gene with human congenital toxoplasmosis. The upper diagram shows the positions of genotyped SNPs relative to the intron/exon structure of the gene. The lower diagram shows the LD plots generated in Haploview using IRAK4 gene data from our North American patient cohort. LD values (D′ × 100) between markers are indicated at the intersection of the 2 markers on the matrix. Where there is no value, D′ = 1 (i.e., 100). In outlining gene SNP association with susceptibility to congenital toxoplasmosis, D′ values between loci were calculated and displayed using Haploview. When there is high confidence in the value of D′ (logarithm of the odds score, >2), pink and red are used. If the confidence is lower (logarithm of the odds score, <2), blue shading (D′ = 1) or no shading (D′ < 1) is used.