Iron-dependent epigenetic modulation promotes pathogenic T cell differentiation in lupus

Abstract

The trace element iron affects immune responses and vaccination, but knowledge of its role in autoimmune diseases is limited. Expansion of pathogenic T cells, especially T follicular helper (Tfh) cells, has great significance to systemic lupus erythematosus (SLE) pathogenesis. Here, we show an important role of iron in regulation of pathogenic T cell differentiation in SLE. We found that iron overload promoted Tfh cell expansion, proinflammatory cytokine secretion, and autoantibody production in lupus-prone mice. Mice treated with a high-iron diet exhibited an increased proportion of Tfh cell and antigen-specific GC response. Iron supplementation contributed to Tfh cell differentiation. In contrast, iron chelation inhibited Tfh cell differentiation. We demonstrated that the miR-21/BDH2 axis drove iron accumulation during Tfh cell differentiation and further promoted Fe2+-dependent TET enzyme activity and BCL6 gene demethylation. Thus, maintaining iron homeostasis might be critical for eliminating pathogenic Th cells and might help improve the management of patients with SLE.

Keywords: Autoimmunity; Epigenetics; Lupus; T cells.

Figure 1. Increased intracellular iron in lupus CD4⁺ T cells.

(A) Representative flow cytometry and quantification of Fe²⁺ in CD4⁺ T cells from healthy donors or patients with SLE (n = 58 for healthy donors, n = 61 for patients with SLE). (B) Quantification of Fe²⁺ in CD4⁺ T cells from healthy control (n = 58), inactive (SLEDAI ≤ 4, n = 38), and active (SLEDAI > 4, n = 23) patients with SLE. (C) qPCR of FTH in CD4⁺ T cells from healthy donors (n = 37) and patients with SLE (n = 34). (D) qPCR of FTH in CD4⁺ T cells from healthy donors (n = 26) and patients with SLE (n = 25). (E) Western blot of ferritin in CD4⁺ T cells from healthy donors (n = 5) and patients with SLE (n = 4). (F) Correlation between Fe²⁺ and Tfh cell percentage in SLE CD4⁺ T cells (n = 28). Data are shown as mean ± SEM. Data are representative of 2 independent experiments. *P < 0.05, ***P < 0.001, ****P < 0.0001 (unpaired 2-tailed Student’s t test for A, C, and D; 1-way ANOVA and Tukey’s multiple-comparisons test for B; and Pearson’s correlation for F).

Figure 2. HID contributes to pathogenic T cell differentiation in lupus mice.

3-week-old female MRL/lpr mice were fed with a normal iron diet (ND, 50 mg/kg, n = 8) or a high-iron diet (HID, 500 mg/kg, n = 8) for 20 weeks. (A–F) Representative flow cytometry and (A) quantification of CD4⁺Ki67⁺ cells, (B) CD4⁺CD44⁺CD62L^– effector memory (EM) cells, (C) CD4⁺CXCR5⁺PD-1⁺ Tfh cells, (D) B220⁺GL-7⁺FAS⁺ GC B cells, (E) CD4⁺CXCR5⁺PD-1⁺FOXP3⁺ Tfr cells, and (F) CD4⁺CD25⁺FOXP3⁺ Tregs in MRL/lpr mice fed with ND or HID. (G–J) Quantification of (G) CD4⁺IFN-γ⁺ cells, (H) CD4⁺IL-17A⁺ cells, (I) CD4⁺ IL-4⁺ cells, and (J) CD4⁺ IL-21⁺ cells in MRL/lpr mice fed with ND or HID. (K) Serum levels of anti-dsDNA IgG in MRL/lpr mice fed with ND or HID. (L) Urine protein of MRL/lpr mice fed with ND or HID. (M) Representative morphology (by H&E and PAS staining) and histological scoring of kidneys of MRL/lpr mice after 20 weeks of ND or HID treatment. Scale bar: 50 μm. Cells were isolated from dLNs and spleens of 23-week-old ND- and HID-treated mice. Data are shown as mean ± SEM. Data are representative of 2 independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (unpaired 2-tailed Student’s t test for A–K and unpaired 2-tailed Mann-Whitney U tests for L and M).

Figure 3. HID promotes exogenous antigen-induced GC response.

3-week-old female B6 mice were treated with a normal iron diet (ND, 50 mg/kg, n = 5) and a high-iron diet (HID, 500 mg/kg, n = 5) for 5 weeks and immunized with sheep red blood cells (SRBCs) by i.p. injection. Mice were sacrificed for analysis after 2 weeks of immunization. (A) Schematic diagram of the HID treatment and SRBC immunization. (B) Body weight change of mice treated with ND or HID. (C) The level of serum iron in mice treated with ND or HID. (D) mRNA expression of Fth and Tfrc in splenic CD4⁺ T cells of ND- or HID-treated mice. (E) Representative flow cytometry and quantification of CD4⁺CXCR5⁺PD-1⁺ Tfh cells and CD4⁺CXCR5⁺PD-1⁺Foxp3⁺ Tfr cells. (F) Representative flow cytometry and quantification of B220⁺GL-7⁺FAS⁺ GC B cells. (G) Representative flow cytometry and quantification of B220^–CD138⁺ plasma cells. (H–K) Quantification of the percentage and numbers of (H) CD4⁺IL-21⁺ cells, (I) CD4⁺IFN-γ⁺ cells, (J) CD4⁺IL-4⁺ cells, and (K) CD4⁺IL-17A⁺ cells. (L) Serum levels of anti-SRBC IgM and anti-SRBC IgG2a in ND- and HID-treated mice at day 7 and day 14 of SRBC immunization. (M) Representative histology and quantification of GCs in the spleen after 2 weeks of SRBC immunization. Blue, CD3; red, B220; green, PNA. Scar bar: 100 μM. Cells were isolated from the spleens of ND- and HID-treated mice immunized with SRBCs. Data are shown as mean ± SEM. Data are representative of 2 independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001 (unpaired 2-tailed Student’s t test for B–M).

Figure 4. Intracellular iron promotes Tfh cell differentiation in vitro.

(A) Quantification of intracellular iron in healthy donor naive CD4⁺ T cells cultured under Tfh cell–polarized conditions for different amounts of time (n = 3). (B) Healthy donor naive T cells were cultured under Tfh cell–polarized conditions in the presence of PBS control or iron dextran (20 μM), and the percentage and quantification of CD4⁺CXCR5⁺PD-1⁺ Tfh cells were determined by flow cytometry 3 days later (n = 5). (C) Healthy donor naive CD4⁺ T cells were cultured under Tfh cell–polarized conditions in the presence of 2,5-DHBA (10 μM and 20 μM). After 3 days of differentiation, the percentage and quantification of the CD4⁺CXCR5⁺PD-1⁺ Tfh percentage were determined (n = 3). (D and E) Healthy donor naive CD4⁺ T cells were cultured under Tfh cell–polarized conditions for 3 days and treated with DMSO or CPX (20 μM) for the last 4 hours. Representative flow cytometry and quantification of (D) viable cells and (E) CD4⁺CXCR5⁺PD-1⁺ Tfh cells are shown (n = 5). Data are shown as mean ± SEM. Data are representative of 2 independent experiments with 3–5 donors. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (1-way ANOVA with Tukey’s multiple-comparisons test for A and C, and paired 2-tailed Student’s t test for B, D, and E).

Figure 5. miR-21 contributes to Tfh cell differentiation in vitro.

(A) Healthy naive CD4⁺ T cells were cultured under Tfh cell–polarized conditions for 1, 3, and 5 days. Representative flow cytometry and quantification of CD4⁺CXCR5⁺PD-1⁺ Tfh cells are shown (n = 3). (B) qPCR of miR-21 during the differentiation process of Tfh cells in A. (C–E) Healthy naive CD4⁺ T cells were transfected with Agomir-NC or Agomir-21 and cultured under Tfh cell–polarized conditions for 3 days (n = 5). After 3 days of polarization, (C) the expression level of miR-21, (D) flow cytometry and quantification of CD4⁺CXCR5⁺PD-1⁺ Tfh cells, and (E) mRNA expression of CXCR5, PDCD1, IL21, and BCL6 were analyzed. (F–H) Healthy naive CD4⁺ T cells were transfected with Antagomir-NC or Antagomir-21 and cultured under Tfh cell–polarized conditions for 3 days (n = 5). After 3 days of polarization, (F) the expression of miR-21, (G) flow cytometry and quantification of CD4⁺CXCR5⁺PD-1⁺ Tfh cells, and (H) mRNA expression of CXCR5, PDCD1, IL21, and BCL6 were analyzed. (I) Representative flow cytometry and quantification of CD4⁺CXCR5⁺PD-1⁺ Tfh cells transfected with Agomir-NC, Agomir-21, and Agomir-21 plus 2,5-DHBA (n = 3). (J) Representative flow cytometry and quantification of CD4⁺CXCR5⁺PD-1⁺ Tfh cells transfected with Antagomir-NC, Antagomir-21, or Antagomir-21 plus iron dextran (n = 3). Data are shown as mean ± SEM. Data are representative of at least 2 independent experiments with 3–5 donors. *P < 0.05, **P < 0.01, ****P < 0.0001 (1-way ANOVA with Tukey’s multiple-comparisons test for A, B, I, and J and 2-tailed Student’s t test for C–H).

Figure 6. miR-21 promotes Tfh cell–mediated GC response.

8-week-old WT (n = 5) or miR-21 cKO mice (n = 5) were immunized with sheep red blood cells (SRBCs) for 7 days. After 7 days of SRBCs stimulation, mice were sacrificed for analysis. (A) Representative flow cytometry and quantification of CD4⁺ T cells. (B) Representative flow cytometry and quantification of CD4⁺CXCR5⁺PD-1⁺ Tfh cells. (C) Representative flow cytometry and quantification of B220⁺GL-7⁺FAS⁺ GC B cells. (D) Serum levels of anti-SRBC IgG isotypes after 7 days of SRBC immunization. (E) Representative histology of spleens at day 7 after SRBC immunization and quantification of GC number and GC area (dashed line). Blue, CD3; red, B220; green, PNA. Scale bar: 100 μM. For A–C, cells were isolated from the spleens of 8-week-old WT or miR-21 cKO mice after 7 days of SRBC immunization. Data are shown as mean ± SEM. Data are representative of 3 independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (unpaired 2-tailed Student’s t test for A–E).

Figure 7. miR-21 regulates Tfh cell differentiation in lupus CD4⁺ T cells.

(A–E) qPCR of (A) miR-21, (B) CXCR5, (C) PDCD1, (D) BCL6, and (E) IL21 in CD4⁺ T cells from healthy donors and patients with SLE (n = 18). (F and G) Correlation between miR-21 and (F) SLEDAI score and (G) CXCR5 mRNA in SLE CD4⁺ T cells (n = 18). (H) Representative flow cytometry and quantification of CD4⁺CXCR5⁺PD-1⁺ Tfh cells in CD4⁺ T cells isolated from peripheral blood from healthy donors (n = 18) and patients with SLE (n = 15). (I–K) SLE CD4⁺ T cells were transfected with Antagomir-NC or Anagomir-21 and stimulated by anti-CD3 and anti-CD28 for 2 days (n = 4). (I) miR-21 expression, (J) flow cytometry and quantification of CD4⁺CXCR5⁺PD-1⁺ Tfh cell subsets, and (K) mRNA levels of CXCR5, PDCD1, IL21, and BCL6 were analyzed. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (Mann-Whitney U test for A, C, D, E, and H and 2-tailed Student’s t test for B and I–K; Spearman’s correlation for F and G). Data are shown as the mean ± SEM. For I–K, data are representative of 3 independent experiments with 4 donors.

Figure 8. BDH2 is the target gene of miR-21 in regulation of Tfh cells.

(A–C) Healthy naive CD4⁺ T cells were transfected with siRNA-NC or siRNA-BDH2 and then cultured under Tfh cell–polarized conditions for 3 days (n = 5). After 3 days of stimulation, (A) qPCR and Western blot of BDH2, (B) flow cytometry and quantification of CD4⁺CXCR5⁺PD-1⁺ Tfh cells, and (C) qPCR of CXCR5, PDCD1, IL21, and BCL6 were analyzed. (D–F) Healthy naive CD4⁺ T cells were transfected with pCMV6-NC or pCMV6-BDH2 and then cultured under Tfh cell–polarized conditions for 3 days (n = 5). (D) qPCR and Western blot of BDH2, (E) flow cytometry and quantification of CD4⁺CXCR5⁺PD-1⁺ Tfh cells, and (F) qPCR of CXCR5, PDCD1, IL21, and BCL6 were analyzed. (G) Representative flow cytometry and quantification of induced Tfh cells in cells transfected with Agomir-NC, Agomir-21, and Agomir-21 plus pCMV6-BDH2 (n = 3). (H) Representative flow cytometry and quantification of CD4⁺CXCR5⁺PD-1⁺ Tfh cells in cells transfected with siRNA-NC, siRNA-BDH2, and siRNA-BDH2 plus 2,5-DHBA (n = 3). (I) Representative flow cytometry and quantification of CD4⁺CXCR5⁺PD-1⁺ Tfh cells in cells transfected with pCMV6-NC, pCMV6-BDH2, or pCMV6-BDH2 with iron dextran (n = 3). Data are shown as mean ± SEM. Data are representative of at least 2 independent experiments with 3–5 donors. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (2-tailed Student’s t test for A–F and 1-way ANOVA with Tukey’s multiple-comparisons test for G–I).

Figure 9. The miR-21/BDH2 axis promotes DNA hydroxymethylation of the BCL6 gene by regulating intracellular iron.

Healthy naive CD4⁺ T cells were isolated from peripheral blood from healthy donors and cultured under Tfh cell–polarized conditions for 3 days. After 3 days of stimulation, cells were collected for analysis. (A) Activity of TET enzymes, and (B) relative hydroxymethylation and (C) methylation levels in the promoter of Tfh cell–related genes BCL6, CXCR5, PDCD1, and IL21 in induced Tfh cells transfected with Agomir-NC or Agomir-21 (n = 3). (D) Activity of TET enzymes, and (E) relative hydroxymethylation and (F) methylation levels in the promoter of BCL6, CXCR5, PDCD1, and IL21 in induced human Tfh cells transfected with siRNA-NC or siRNA-BDH2 (n = 3). (G) Activity of TET enzymes, and (H) relative hydroxymethylation and (I) methylation levels in the promoter of BCL6, CXCR5, PDCD1, and IL21 in induced human Tfh cells transfected with Antagomir-NC or Antagomir-21 (n = 3). (J) Activity of TET enzymes, and (K) relative hydroxymethylation and (L) methylation levels in the promoter of BCL6, CXCR5, PDCD1, and IL21 in induced human Tfh cells transfected with pCMV6-NC or pCMV6-BDH2 (n = 3). Data are shown as mean ± SEM. Data are representative of 3 independent experiments with 3 donors. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (2-tailed Student’s t test for A–L).

Figure 10. Schematic illustration of the contribution of iron overload to the pathogenic T cell differentiation and pathogenesis of SLE.

In lupus CD4⁺ T cells, iron accumulation promotes Tfh cell differentiation and Tfh cell–mediated autoimmune responses, autoantibody production, as well as inflammatory cytokine secretion, driving disease progression of lupus. Mechanically, miR-21 represses BDH2 to induce iron accumulation in lupus CD4⁺ T cells by limiting the synthesis of siderophore 2,5-DHBA, which enhances Fe²⁺-dependent TET enzyme activity and promotes BCL6 promoter hydroxymethylation and transcription activation, leading to excessive Tfh cell differentiation in SLE. Together, iron overload is an important inducer of the autoimmune response in lupus, and maintaining iron homeostasis will provide a good way for therapy and management of SLE.