Development of a unique epigenetic signature during in vivo Th17 differentiation

Abstract

Activated naive CD4(+) T cells are highly plastic cells that can differentiate into various T helper (Th) cell fates characterized by the expression of effector cytokines like IFN-γ (Th1), IL-4 (Th2) or IL-17A (Th17). Although previous studies have demonstrated that epigenetic mechanisms including DNA demethylation can stabilize effector cytokine expression, a comprehensive analysis of the changes in the DNA methylation pattern during differentiation of naive T cells into Th cell subsets is lacking. Hence, we here performed a genome-wide methylome analysis of ex vivo isolated naive CD4(+) T cells, Th1 and Th17 cells. We could demonstrate that naive CD4(+) T cells share more demethylated regions with Th17 cells when compared to Th1 cells, and that overall Th17 cells display the highest number of demethylated regions, findings which are in line with the previously reported plasticity of Th17 cells. We could identify seven regions located in Il17a, Zfp362, Ccr6, Acsbg1, Dpp4, Rora and Dclk1 showing pronounced demethylation selectively in ex vivo isolated Th17 cells when compared to other ex vivo isolated Th cell subsets and in vitro generated Th17 cells, suggesting that this unique epigenetic signature allows identifying and functionally characterizing in vivo generated Th17 cells.

Figure 1.

Frequency of IL-17A- and IFN-γ-producing cells among memory T cells from female aged Balb/c mice. (A) CD4⁺ enriched cells from lymph nodes and spleen of aged Balb/c mice were stimulated with PMA/ionomycin, labeled with the IL-17A- and IFN-γ-cytokine secretion kit, analyzed and sorted by flow cytometry. Rectangles in the plots indicate the sorting gates. Frequencies of IFN-γ⁻IL-17A⁺ Th17 cells, IFN-γ⁺IL-17A⁻ Th1 cells, IFN-γ⁻IL-17A⁻ Th-mem cells and IFN-γ⁻IL-17A⁻ naive T (Tnaive) cells are depicted. (B) Analysis of sorted naive T cells, Th-mem, Th1 and Th17 cells. Numbers indicate the frequency of cells within quadrants. Representative dot plots of 23 independently performed sorts are shown.

Figure 2.

Distribution of methylated regions in Th cell subsets. Methylated genomic DNA fragments from Th cell subsets were enriched by precipitation with the human MBD2 protein and subjected to next-generation sequencing. The quantity of sequences that were mapped to the same chromosomal location was translated into an activity value. This experiment was performed once. (A) The plot depicts the distribution of activity values (empirical probability density function) with a specified number of CpG motifs in Th17 cells. (B) Pre-defined probability thresholds for activity values were used to compare the methylation status in different Th cell subsets. The plot depicts the frequencies of uniquely methylated DNA regions out of all methylated regions for naive T cells, Th-mem, Th1 or Th17 cells.

Figure 3.

Shared and uniquely demethylated regions within Th cell subsets. (A) The two-way Venn diagrams display the distribution of demethylated regions between naive T cells (orange) and Th1 (red) or Th17 (green) cells. (B) The three-way Venn diagram compares the distribution of those regions between Th-mem (blue), Th1 (red) and Th17 (green) cells. Numbers indicate the uniquely or shared (gray) demethylated regions at a probability threshold of 25%.

Figure 4.

Differential methylation of control regions within Th cell subsets. (A) Locus organization and CpG motif distribution are shown for the loci Xist, Itga4, Il17a and Ifng. Pyramidal-arranged black bars display precipitated DNA-fragments that were combined into a DNA region. The maximum height of the overlapping fragments represents the MBD-seq activity. (B) Difference plots derived from high resolution melting analysis of regions described in (A). Analyzed CpG motifs are marked in black. Relative signal difference at each temperature was calculated and plotted by comparing differences between the melting curves of each subset. The experiment was performed twice. Representative plots are shown.

Figure 5.

Candidate regions Zfp362, Ccr6, Acsbg1, Dpp4, Rora and Dclk1 are differentially methylated. Locus organization and CpG motif distribution are shown for six candidate regions. The methylation status of marked CpG motifs (black) was studied by MS-HRM analyses. Relative signal difference at each temperature was calculated and plotted by comparing differences between the melting curves of each subset. Data from one out of two independently performed experiments are depicted.

Figure 6.

Detailed methylation analysis of Th17-specific epigenetic signature genes in ex vivo isolated and in vitro generated T cell subsets by pyrosequencing. Detailed analysis of the methylation status of the candidate DMRs was performed by pyrosequencing using genomic DNA from the indicated (A) ex vivo isolated and (B) in vitro generated T cell subsets. The methylation rates were translated into a color code from yellow (0%) via green (50%) up to blue (100%). Each rectangle represents the methylation of one CpG motif. Data from three independently performed cell sorts and cultures are depicted.

Figure 7.

Detailed methylation analysis of the Th17-specific epigenetic signature genes in major immune cell subsets. Detailed analysis of the methylation status of the candidate DMRs was performed by pyrosequencing using genomic DNA from the indicated immune cell subsets. The methylation rates were translated into a color code from yellow (0%) via green (50%) up to blue (100%). Each rectangle represents the methylation of one CpG motif. Data from two independently performed cell sorts are depicted.