TET2 Regulates Mast Cell Differentiation and Proliferation through Catalytic and Non-catalytic Activities

Abstract

Dioxygenases of the TET family impact genome functions by converting 5-methylcytosine (5mC) in DNA to 5-hydroxymethylcytosine (5hmC). Here, we identified TET2 as a crucial regulator of mast cell differentiation and proliferation. In the absence of TET2, mast cells showed disrupted gene expression and altered genome-wide 5hmC deposition, especially at enhancers and in the proximity of downregulated genes. Impaired differentiation of Tet2-ablated cells could be relieved or further exacerbated by modulating the activity of other TET family members, and mechanistically it could be linked to the dysregulated expression of C/EBP family transcription factors. Conversely, the marked increase in proliferation induced by the loss of TET2 could be rescued exclusively by re-expression of wild-type or catalytically inactive TET2. Our data indicate that, in the absence of TET2, mast cell differentiation is under the control of compensatory mechanisms mediated by other TET family members, while proliferation is strictly dependent on TET2 expression.

Keywords: DNA hydroxymethylation; TET; differentiation; epigenetics; mast cells; proliferation.

Figure 1. Altered Mast Cell Differentiation, Proliferation, and Effector Functions in the Absence of Tet2

(A) Bone marrow cells from Tet2^+/+, Tet2^−/−, and Tet2^+/– mice were differentiated to mast cells. Shown is one representative surface staining for mast cell markers (Kit and FcεRIα) and other myeloid markers (Mac1 and Ly6G) after 3 weeks of differentiation. (B) Percentage of Kit⁺ FcεRIα⁺ mast cells in bone marrow cultures of Tet2^+/+ and Tet2^−/− littermates, over 6 weeks of differentiation, is shown (n = 4; two-way ANOVA, ***p < 0.0003 and **p = 0.0084). (C) Same as (B), except that quantification for seven independent experiments at the third week of differentiation is shown. Each dot represents one biological sample (unpaired t test, two-tailed; mean ± SEM). (D) Mast cell proliferation was assessed by BrdU incorporation assay. Kit⁺ FcεRIα⁺ mast cells were either physically separated at least 1 day before the experiment or were gated at the analysis step. (E) Same as (D), except that the compiled result of 13 independent experiments is shown (paired t test, two tailed; mean ± SEM). (F) Cells were stimulated with IgE and antigen before intracellular cytokine staining. One representative staining for TNF-α expression is shown. (G) Same as (F), except that the compiled results for six (IL-6 and TNF-α) or seven (IL-13) independent experiments are shown (paired t test, two tailed;mean ± SEM). See also Figures S1 and S2.

Figure 2. RNA-Seq Analysis Identifies Dysregulated Genes and Functional Pathways in Tet2^−/− Mast Cells

(A) Total RNA from Tet2^+/+ and Tet2^−/− mast cells was used for RNA-seq. DEGs are shown in a hierarchically clustered heatmap. For hierarchical clustering, distance metric between rows was calculated according to the Pearson correlation on thelog2 of FPKM values and then normalized by Z score. Lowest Z score values are dark blue and highest values are bright red. (B and C) Enriched GO functional categories (left) and over-represented TF consensus DNA-binding sites (right) in up- (C) and downregulated (B) genes were identified using REVIGO and PSCAN, respectively. GO terms are visualized in semantic similarity-based scatterplots. Bubble color indicates the p value as follows: blue and green bubbles are GO terms with more significant p values compared to orange and red bubbles. Bubble size shows how much the GO term is represented in the GO database. Selected representative categories are shown. For PSCAN analyses, each over-represented TF subfamily is shown according to the obtained Welch p values (−log10). See also Figure S3.

Figure 3. VitC Treatment Restores Gene Expression in Tet2^−/− Mast Cells

(A) FPKM values are plotted for selected DEGs identified by RNA-seq (mean ± SEM). (B) The same genes as in (A) were validated by qRT-PCR. Each dot represents one independent biological sample (n = at least 4). Shown is the expression relative to the endogenous control Tbp and to one Tet2^+/+ sample. (C) After 2.5 weeks of differentiation, Tet2^+/+ and Tet2^−/− mast cells were either left untreated or were supplemented with 10 and 50 μg/ml VitC for 10 days. The DNA was extracted and 2-fold dilutions starting from 500 ng were spotted on a membrane. The 5hmC content was assessed using an anti-5hmC antibody. (D) Cells were treated as in (C) with medium supplemented with 50 μg/ml VitC for 10 days. Total RNA was then extracted and gene expression analyzed by qRT-PCR. Shown is the expression relative to Tbp. Expression of all genes also is related to one Tet2^+/+ sample except for Mpo that was mostly undetectable in Tet2^+/+ cells (n = at least 3, mean ± SEM). See also Figure S4.

Figure 4. VitC Treatment Restores Mast Cell Differentiation and Cytokine Production, but Does Not Normalize Proliferation

(A) Tet2^+/+ and Tet2^−/− mast cells were treated with 50μg/ml VitC for 10 days, after which proliferation was assessed by BrdU incorporation. One representative experiment is shown. (B) Same as (A), except that the results of seven experiments are shown (paired t test, two tailed; mean ± SEM). (C) Tet2^−/− mast cells were treated as in (A) prior to intracellular cytokine staining. One representative experiment for TNF-α is shown. (D)Same as (C), except that the results of five (IL-6 and TNF-α) or three (IL-13) experiments are shown. (E) Tet2^+/+ and Tet2^−/− bone marrow cells were differentiated to mast cells for 2 weeks, after which culture medium was further supplemented with 50 μg/ml VitCfor 8 days. The percentage of mast cells in culture was assessed by surface staining for Kit and FcεRIα. One representative experiment is shown. (F) Same as (E), except that the results of four experiments are shown. Statistical analysis was performed with two-way ANOVA.

Figure 5. Knockdown of TET3 Further Exacerbates the Differentiation Defect of Tet2^−/− Cells, but Does Not Affect Proliferation

(A) NSC-34 cells were transduced with lentiviruses expressing either an irrelevant control hairpin (shLuc) or two different shRNAs against murine TET3 (sh1 and sh2), alone or in combination. Levels of genomic 5hmC were measured by dot blot. The 2-fold dilution of DNA, starting from 100 ng, was spotted on the membrane. Values of relative quantification are indicated on the blot. (B) Tet2^−/− bone marrow cells were transduced with a mixture of shRNAs against TET3 as in (A) and the expression of Tet3 mRNA was measured by qRT-PCR. Shown is the result of two independent experiments each performed with two biological replicates. Data were normalized to the expression of β2 microglobulin (unpaired t test, two tailed; mean ± SEM). (C) Same as (B), except that proliferation of Tet2^−/− cells transduced with a control hairpin or shTet3 was measured by BrdU incorporation assay. One representative result is shown. (D) Same as (C), except that the results of five experiments are shown (paired t test, two tailed; mean ± SEM). (E) Tet2^−/− bone marrow precursors were transduced as in (B), and cell differentiation was followed overtime by surface staining for Kit and FcεRIα. Shown is one representative result after 2.5 weeks of differentiation. (F) Same as (E), except that the data of two different experiments are shown, each performed with two independent biological replicates. Statistical analysis wasperformed with two-way ANOVA.

Figure 6. Dysregulated Expression of C/EBP TFs Leads to Delayed Mast Cell Differentiation

(A) (Left) Expression of Cebpe was assessed in Tet2^+/+ and Tet2^−/− mast cells by qRT-PCR (n = 5, each dot represents one biological sample; unpaired t test, two tailed; mean ± SEM). (Right) Tet2^+/+ Lin–bone marrow hematopoietic precursors were transduced with a lentiviral vector to express C/EBPε. Expression of Cebpe was assessed by qRT-PCR. Endogenous control: Tbp. (B) Tet2^+/+ Lin^– hematopoietic precursors were transduced as in (A), and mast cell differentiation was assessed over time by surface staining for Kit and FcεRIα. One representative staining after 2 weeks of differentiation is shown on the left, while the graph on the right shows the results of four independent transductions (two-way ANOVA, ****p < 0.0001 and ***p = 0.0001). (C) Same as (B), except that the expression of selected mast cell proteases was measured by qRT-PCR (n = 4, mean ± SEM). (D) Same as (A) and (B), except that forced expression of C/EBP was performed on bone marrow cells of Tet2^−/− mice. One representative staining at 2 weeks of differentiation is shown on the left; the graph on the right shows the results of six transductions (two-way ANOVA, p values are, from left to right, ***p = 0.0005, ****p < 0.0001, and ***p = 0.0002). (E) Tet2^+/+ Lin⁻ hematopoietic precursors were transduced to express C/EBPα or a control vector. Expression of Cebpa was assessed by qRT-PCR. (F) Same as (E), with mast cell differentiation assessed by surface staining for Kit and FcεRIα. One representative staining after 1.5 weeks of differentiation is shown on the left; the graph in the middle shows one representative experiment over time, while the bar graph on the right shows the results of three different experiments after 1.5 weeks (paired t test, two tailed; mean ± SEM).

Figure 7. TET2 Reconstitution Reduces Mast Cell Proliferation, Independently of 5hmC Modification

(A) Differentially hydroxymethylated distal regions in Tet2^−/− mast cells. Hyper-hydroxymethylated peaks are depicted in red, hypo-hydroxymethylated in blue, and unchanged peaks in gray. (B) Genomic location of the differentially hydroxymethylated regions is shown. (C) Functional enrichment analysis of differentially hydroxymethylated regions was performed using GREAT. Selected categories are shown; see Table S8 for the full list of significant categories. (D) Distance of hyper- and hypo-hydroxymethylated regions from the nearest dysregulated (up- or downregulated) gene is shown. (E) Distance of H3K4me1 peaks, as assessed by ChIP-seq of Tet2^+/+ cells, from the summit of hypo-hydroxymethylated regions is shown. (F) Tet2^−/− mast cells were transduced with an empty vector or with vectors expressing WT or catalytically inactive TET2(H×D). Levels of 5hmC were measured by dot blot. One representative experiment is shown. (G) Proliferation of mast cells transduced as in (F) was assessed by BrdU incorporation assay. Each dot represents one independent experiment (paired t test, two tailed; mean ± SEM). See also Figure S5.