Elevated transferrin receptor impairs T cell metabolism and function in systemic lupus erythematosus

Abstract

T cells in systemic lupus erythematosus (SLE) exhibit multiple metabolic abnormalities. Excess iron can impair mitochondria and may contribute to SLE. To gain insights into this potential role of iron in SLE, we performed a CRISPR screen of iron handling genes on T cells. Transferrin receptor (CD71) was identified as differentially critical for T_H1 and inhibitory for induced regulatory T cells (iT_regs). Activated T cells induced CD71 and iron uptake, which was exaggerated in SLE-prone T cells. Cell surface CD71 was enhanced in SLE-prone T cells by increased endosomal recycling. Blocking CD71 reduced intracellular iron and mTORC1 signaling, which inhibited T_H1 and T_H17 cells yet enhanced iT_regs. In vivo treatment reduced kidney pathology and increased CD4 T cell production of IL-10 in SLE-prone mice. Disease severity correlated with CD71 expression on T_H17 cells from patients with SLE, and blocking CD71 in vitro enhanced IL-10 secretion. T cell iron uptake via CD71 thus contributes to T cell dysfunction and can be targeted to limit SLE-associated pathology.

Fig. 1.. Transferrin receptor is conditionally essential for effector T cells and iT_regs.

(A) Naïve CD4 T cells from Cas9-transgenic mice were stimulated and skewed into T_H1 (left) or iT_reg (right) cells. Cells were transduced with a retroviral library of gRNAs specific for genes in iron metabolism or nontargeting controls. MAGeCK analysis was used to determine gRNA depletion (blue) and enrichment (red) in T cell cultures over 7 days. Results are representative of two independent screens. (B) CD71 expression at 5 days after differentiation of activated T_H1 and iT_reg cultures. (C) Intracellular ferrous iron in T_H1 cells and iT_regs measured by BioTracker iron staining. (D) CD71 expression at 72 hours and 6 days after activation with a representative plot (left). (E) T cell pellets from 72 hours after activation were subjected to ICP-MS to determine total (2⁺ and 3⁺) intracellular iron concentrations. “Unstim” represents T cells that were left unstimulated, and “Media” is a control for iron content in the cell culture medium only. ppb, parts per billion. (F) Tfrc mRNA was quantified in activated T cells by qRT-PCR. (G) Irp2 protein levels were quantified by immunoblot in activated CD4 T cells and normalized to the actin loading control. (H) Whole-cell lysates from ex vivo purified CD4 T cells were examined for Hfe expression by immunoblot. (I) Activated CD4 T cells from control or SLE1.2.3 mice were treated with transferrin protein for 0, 10 min, 30 min, or 2 hours to stimulate CD71 endosomal trafficking. Confocal microscopy was used to quantify the percentage of cells that contained CD71⁺ recycling endosomes by Rab8 colocalization with CD71 (right). Each data point is from an individual biological replicate where data from three technical replicates were averaged. Representative images from ×20 magnification at the 2-hour time point are shown. Scale bars, 50 μm. (B to H) Student’s unpaired two-tailed t test. (I) Two-way ANOVA. ns, not significant.

Fig. 2.. CD71 blockade normalizes T cell activation and mTORC1 in SLE1.2.3 T cells.

(A) CD69 expression was determined by flow cytometry 24 hours after activation in T cell cultures. (B) CD44 expression was determined by flow cytometry 72 hours after activation. (C) T cell cultures were restimulated on day 5 after activation with PMA/ionomycin to quantify the percentage of IL-2⁺ CD4 T cells. (D) IL-2 mRNA was quantified on day 4 of activation by qRT-PCR. (E) Cell culture supernatants were collected on day 4 of activation, and IL-2 concentrations were determined by ELISA. Values were then normalized to viable cell counts. (F) T cell cultures on day 5 of stimulation were analyzed by intracellular flow cytometry for p-S6. Results are from two independent experiments. (G) Forward scatter (FSC) and side scatter (SSC) were measured in T cell cultures. (H) CD4 T cells were activated for 24 hours and then treated with antibodies or nM rapamycin (rapa) for 2 days. IL-7 cells were left unstimulated as a control group. CD71 was measured by flow cytometry, and (I) CellTrace Violet (CTV) staining was used to calculate the division index. (A, B, and D to I) One-way ANOVA with Sidak’s multiple comparisons test. (C) Paired ANOVA with Sidak’s multiple comparisons test. Results are from three experiments combined.

Fig. 3.. CD71 blockade restores metabolic and mitochondrial function in SLE1.2.3 T cells.

(A and B) MitoTracker Green (A) and MitoSOX Red (B) staining was measured by flow cytometry on day 5 after activation. (C) T cell cultures were processed for EM. A minimum of 100 mitochondria were analyzed in each sample group to quantify mitochondrial area (right). Scale bars, 1 μm (at 2700×) and 400 nm (at 6500×). (D) Extracellular flux analysis on day 5 after activation. Representative oxygen consumption rate (OCR) during a Mito Stress Test is shown. Cells were treated with dimethyl sulfoxide (DMSO) or CPX for 4 hours. (E) Representative Mito Stress Test for T cells activated with isotype control or anti-CD71. Maximal respiration (F) and proton leak (G) were quantified from Mito Stress Test (E). R/A, rotenone and antimycin A. (H) An ATP rate assay was performed with CD4 T cells as activated in (E) to quantify mitochondrial (left) versus glycolytic (right) ATP production. (A, B, and E to H) One-way ANOVA with Sidak’s multiple comparisons test. (C and D) Paired ANOVA.

Fig. 4.. CD71 blockade differentially affects T cell subsets.

(A) RNA sequencing results of T_H17 differentiated cultures from either healthy control or SLE1.2.3 mice. IL-17 signaling and T_H17 differentiation gene expression is shown in a heatmap by log₂ fold change (Log₂FC). Data are normalized to controls. (B) Iron metabolism genes in T_H17 cultures from control or SLE1.2.3 samples. P values on plots are the adjusted P values determined by DESeq2 analysis. (C) CD71 expression on day 3 of activation in T_H17 cells or iT_regs treated with isotype control, anti-CD71, CPX, or iron supplementation. (D) iT_reg and T_H1 cultures ± CD71 blockade were analyzed for Foxp3 expression on day 5 after activation. (E) iT_reg cultures were analyzed for IL-2 production by flow cytometry. (F) Naïve T cells were subjected to iT_reg differentiation for 3 days. Supernatants were analyzed for IL-10 by ELISA and normalized to cell number. (G) Naïve CD4 T cells isolated from SLE1.2.3 and control mice. Day 4 cells were restimulated as in (C) to determine the percentage of IL-10⁺ T cells. (H) c-MAF expression in Foxp3⁺ iT_regs. (I) Naïve T cells from Cas9 transgenic mice were differentiated into iT_reg cultures for 2 days and transduced with a gRNA targeting Tfrc or a nontargeting control (NTC). Transduced cells (Thy1.1⁺) were analyzed for Foxp3 expression (left) and p-S6 (right) by flow cytometry 5 days later. (C, D, and H) Two-way ANOVA with Dunnett’s multiple comparisons test. (G) One-way ANOVA with Sidak’s multiple comparisons test. (I) Paired ANOVA. (E and F) Paired Student’s t test. (I) Unpaired two-tailed Student’s t test. FPKM, fragments per kilobase of transcript per million mapped fragments.

Fig. 5.. CD71 blockade reduces autoimmunity and pathology in SLE1.2.3 mice.

(A) SLE1.2.3 mice or age-matched B6 controls were treated twice per week with an isotype control or anti-CD71 for 4 weeks, n = 4. (B) Anti-dsDNA antibodies were measured in serum before the treatment regimen. The dotted line represents the cutoff value for inclusion criteria. (C) End point serum levels of anti-dsDNA Ig antibodies. (D) Splenomegaly was measured by spleen–to–body mass ratio at study end point. (E) CD4 T cells were isolated from the spleen and LNs by negative selection, and cell counts were determined by automated cell counter. (F) CD4 T cells as described in (E) were stained for CD44 and examined by flow cytometry. (G) and (H) Pathological assessment of inflammation in the kidney (G) and liver (H) was scored semiquantitatively from H&E-stained tissue sections. Arrows indicate infiltrates of lymphocytes and plasma cells. Scale bars, 200 μm (large) and 20 μm (insets). (I) End point sera were measured for IL-10 concentrations. Assay limit of detection was 0.000531. (J) CD4 T cells from spleens and LNs were stimulated with PMA/ionomycin. IL-10⁺ CD4 T cells were quantified by flow cytometry. (K) The percentage of CD25⁺ Foxp3⁺ cells within total CD4 T cells was determined by flow cytometry at the study end point. All experiments were n = 3. One-way ANOVA with Sidak’s multiple comparisons test was used to determine significance of all plots.

Fig. 6.. CD71 expression on T cells is required to drive SLE.

Splenocytes from CD4⁻Cre⁻ Tfrc^fl/fl or Cre⁺ Tfrc^fl/fl mice were transferred into bm12 recipient mice to induce SLE. Sham mice received a control PBS injection with no cells. (A) Sera from days 3, 8, and 14 after transfer were measured for anti-dsDNA IgG antibodies. Red, Cre⁻ recipients; blue, Cre⁺ recipients; gray, sham recipients. (B) T_FH cell frequency as a percentage of total CD4 T cells in peripheral blood and GC B cell frequency as a percentage of CD19⁺ B cells in peripheral blood. (C) Spleen weights measured on day 14 study end point. (D) Day 14 splenocytes were analyzed for GC B cell frequency and plasma cell frequency as a percentage of CD19⁺ B cells. (E) Mean fluorescence intensity (MFI) of CD71 on CD4 T cells in the spleen. (F) Frequency of T_FH cells in the spleen and ICOS expression on all CD4 T cells in the spleen. (G) CD44 expression MFI on CD4 T cells in the spleen and MitoTracker Green staining. (H) CD71 MFI on CD4 T cells from the LNs. (I) T_FH frequency and ICOS expression within CD4 T cells from the LN. (J) BioTracker (2⁺) labile iron dye staining in CD4 T cells. (K) Lymphocytes were isolated from the kidneys on day 14. Percentage of lymphocyte layer that were CD4 T cells (left) and CD71 expression on CD4 T cells (right). (A) Two-tailed Student’s t test. (B to K) One-way ANOVA with Sidak’s multiple comparisons test.

Fig. 7.. CD71 expression correlates with cells in patients with SLE.

T_H17 (A) Naïve human T cells were activated for 5 days under nondifferentiating conditions (T_H0) or T_H17 conditions. Fe measured by ICP-MS and normalized to sulfur (S) for protein content. Paired Student’s t test. (B) PBMCs were isolated from healthy donors (Control) and patients with SLE (SLE). Percentage of T_H17 cells in CD4 T cell compartment compared with CD71 expression on bulk CD4 T cells. Controls, black; SLE, red. Pearson correlation was used to determine r value and statistical significance. (C) CD71 expression on bulk CD4T cells. (D) Percentage of CD4T cells defined as T_H17 cells. (E) CD71 expression on T_H17 cells. (F to H) Flow cytometry data from patients with SLEDAI scores of 3 or less compared with those with scores of 4 or higher. (I) Maximal respiration was determined from Mito Stress Test assay on days 4 to 5 after activation. (J) IL-10 concentrations in day 3 supernatants measured by ELISA. (C to FI) Student’s two-tailed t test. (I and J) Paired Student’s t test.