Treg cells limit IFN-γ production to control macrophage accrual and phenotype during skeletal muscle regeneration

Abstract

Skeletal muscle regeneration is a highly orchestrated process that depends on multiple immune-system cell types, notably macrophages (MFs) and Foxp3⁺CD4⁺ regulatory T (T_reg) cells. This study addressed how T_reg cells rein in MFs during regeneration of murine muscle after acute injury with cardiotoxin. We first delineated and characterized two subsets of MFs according to their expression of major histocompatibility complex class II (MHCII) molecules, i.e., their ability to present antigens. Then, we assessed the impact of T_reg cells on these MF subsets by punctually depleting Foxp3⁺ cells during the regenerative process. T_reg cells controlled both the accumulation and phenotype of the two types of MFs. Their absence after injury promoted IFN-γ production, primarily by NK and effector T cells, which ultimately resulted in MF dysregulation and increased inflammation and fibrosis, pointing to compromised muscle repair. Thus, we uncovered an IFN-γ-centered regulatory layer by which T_reg cells keep MFs in check and dampen inflammation during regeneration of skeletal muscle.

Keywords: Treg cells; interferon-γ; macrophages; muscle regeneration; muscle repair.

Fig. 1.

Two subsets of MFs in skeletal muscle. (A) Gating strategy for cytofluorimetric delineation of skeletal muscle MF subsets at steady state. Numbers refer to fraction of cells within the designated gate. (B) Summaries of cell numbers (Left) and fractions (Right) of the MHCII⁺ and MHCII⁻ subsets from 6- to 10-wk-old male C57BL/6J mice. Three independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001 by the unpaired t test. (C) Volcano plot of transcriptomic differences between the MHCII⁺ and MHCII⁻ MF subsets isolated from skeletal muscle at steady state. Averages of duplicate samples. (D) Pathway enrichment analysis (KEGG) of transcripts from ≥twofold differentially expressed genes. P values represent maximum EASE score determined according a modified Fisher Exact test (DAVID). Representative genes in these pathways are labeled in C. Cytofluorimetric dot plots (E) and quantification of cell number (F, Left) and frequency (Right) for MHCII⁺ vs. MHCII⁻ MF subsets during regeneration after acute injury with CTX. Results are from three individual experiments.

Fig. 2.

Changes in MF representation and phenotype in the absence of T_reg cells. (A, Left) Immunofluorescence imaging of MFs and Foxp3⁺ T_reg cells in a cross-section of TA muscle at day 7 after injury with CTX. Original magnification 400× and 4× zoom. White arrows: T_reg cells and MFs within touching proximity. (A, Right) Quantification of the distances between T_reg cells and the nearest MF on day 7 after injury. Measured manually from 20 separate field views from three mice. (B) Continuous T_reg cell depletion regimen. (C–F) Analysis of diverse parameters following the regimen depicted in B: fraction of Foxp3⁺ T_reg cells (C); number of CD45⁺ cells per gram tissue (D); representative dot plots (E, Left) and quantification of MHCII⁺ to MHCII⁻ MF subset ratio (Right); representative dot plots and summary quantification of fraction of EdU⁺MHCII⁺ MF cells after a 4-h pulse just before tissue collection (F). DTR⁺, Foxp3^DTR+ mice; DTR⁻, Foxp3^DTR− littermates. Numbers on dot plots refer to fraction of cells within the designated gates. Results are from three independent experiments. Statistics as per Fig. 1B. (G) Volcano plot comparing transcriptomes from MHCII⁺ MFs of T_reg-depleted (as per B) vs. nondepleted mice. (H) List of top pathways (KEGG) enriched in the sets of genes ≥twofold differentially expressed (P < 0.05). Representative genes in these pathways are indicated in G. Statistics are as per Fig. 1D. ***P < 0.001.

Fig. 3.

An enhanced IFN-γ response in muscle MHCII⁺ MFs in the absence of T_reg cells. (A) Volcano plot as in Fig. 2H, except the IFN-γ response up-signature, detailed description in Materials and Methods, is highlighted in red. (B) DT injection regimens for short-term (24-h) depletion of T_reg cells at early or late phases of regeneration after injury with CTX, followed by analysis at the indicated time-points. Fraction of Foxp3⁺CD4⁺ T_reg cells (C); number of CD45⁺ cells per gram of muscle (D); representative dot plots and summary quantification of MF subset ratios (E). Numbers on dot plots refer to fraction of cells within the designated gates. Data are from at least two independent experiments. Statistics are as in Fig. 1B. (F) Volcano plots of transcriptomes from MHCII⁺ MFs of DTR⁺ vs. DTR⁻ mice upon short-term depletion of T_reg cells beginning at day 1 or 7. Highlighted in red is the IFN-γ response up-signature, as in A. Averages of two to three replicates per group. P values according to the χ² test. ***P < 0.001.

Fig. 4.

Sources of muscle IFN-γ during muscle repair. (A–C) In T_reg-replete mice, quantification of the frequency (A), cell number per gram of muscle (B), and mean fluorescence intensity (MFI) (C) of IFN-γ synthesis by the designated cell types during regeneration after CTX-induced injury. Results are from two independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001.

Fig. 5.

Requirement for T_reg cells early after injury to limit IFN-γ production. (A) DT-treatment regimens. Male DTR⁺ or DTR⁻ littermates were 8–10 wk of age: fraction of Foxp3⁺CD4⁺ T_reg cells (B); number of CD45⁺ cells (C); and quantification of IFN-γ⁺ NK cells, CD4⁺ T_conv, and CD8⁺ T cells per gram of muscle (D). Data are from at least two independent experiments. Statistics are as per Fig. 1B. Cont., continuous. **P < 0.01; ***P < 0.001.

Fig. 6.

Contribution of MHCII⁺ MFs to type 1 inflammation during muscle repair. (A) Representative dot-plot (Left) and summary data for fraction of MHCII⁺ MFs (Right) for MF-MHCII^−/− mice and wild-type littermate controls. Numbers refer to fraction of cells within the gate. (B) Fraction and number of CD4⁺ T_conv and T_reg cells in injured muscle of wild-type and mutant littermates. (C) Fraction and number of IFN-γ⁺ NK, CD4⁺ T_conv, and CD8⁺ T cells at day 14 after injury in WT and mutant littermates. Data are from two independent experiments. Statistics are as per Fig. 1B. *P < 0.05; **P < 0.01; ***P < 0.001.

Fig. 7.

Partial complementation of T_reg cell loss by IFN-γ injection. (A) Strategy for evaluating the ability of IFN-γ to compensate for T_reg cell loss (DTR⁺) during muscle regeneration. rIFN-γ was i.v.-injected into DTR⁻ mice twice as indicated. (B) Representative dot plots (Left) and summary quantification of MF subset ratios (Right). Numbers indicate fraction of cells in the designated gate. (C) Representative histogram (Left) and quantification (Right) of PD-L1 expression in PD-L1⁺ total muscle MFs. (D) Quantification of CD45⁺ cells per gram of muscle. (E) Quantification of IFN-γ⁺ NK, CD4⁺ T_conv, and CD8⁺ T cells per gram of muscle. Results are from three independent experiments. (F) H&E histology: representative images (Left) and inflammation score (Right). Black arrows point to sites of inflammation. (Original magnification: 50×.) (Scale bars: 200 μm.) Scoring was assessed by two-independent, blinded scorers. (G) Gomori Trichrome histology. As per F except fibrosis was assessed. All statistics as per Fig. 1B. *P < 0.05; **P < 0.01.