Senescent skeletal muscle fibroadipogenic progenitors recruit and promote M2 polarization of macrophages

Abstract

Senescent cells compromise tissue structure and function in older organisms. We recently identified senescent fibroadipogenic progenitors (FAPs) with activated chemokine signaling pathways in the skeletal muscle of old mice, and hypothesized these cells may contribute to the age-associated accumulation of immune cells in skeletal muscle. In this study, through cell-cell communication analysis of skeletal muscle single-cell RNA-sequencing data, we identified unique interactions between senescent FAPs and macrophages, including those mediated by Ccl2 and Spp1. Using mouse primary FAPs in vitro, we verified increased expression of Ccl2 and Spp1 and secretion of their respective proteins in the context of both irradiation- and etoposide-induced senescence. Compared to non-senescent FAPs, the medium of senescent FAPs markedly increased the recruitment of macrophages in an in vitro migration assay, an effect that was mitigated by preincubation with antibodies to either CCL2 or osteopontin (encoded by Spp1). Further studies demonstrated that the secretome of senescent FAPs promotes polarization of macrophages toward an M2 subtype. These data suggest the unique secretome of senescent FAPs may compromise skeletal muscle homeostasis by recruiting and directing the behavior of macrophages.

Keywords: cellular senescence; fibroadipogenic progenitors (FAPs); macrophages; migration; polarization.

FIGURE 1

Senescence alters FAP autocrine and paracrine signaling in skeletal muscle. (a) Integration of scRNA‐seq data from non‐senescent FAPs, senescent FAPs, and macrophages. (b) CellChat analysis revealed an extensive intercellular communication network between non‐senescent and senescent FAPs and macrophages. (c) Dot plot showing significant outgoing (sender) and incoming (receiver) pathways of different cell subpopulations. (d,g) Significant interactions in CCL (d) and SPP1 (g) signaling pathways between senescent FAPs and macrophages. Heatmap showing the dominant senders and receivers in the network. (e,h) Contribution of different ligand and receptors in the CCL (e) and SPP1 (h) pathway. (f,i) Violin plot showing the gene expression of the contributing ligands and receptors in the CCL2 (f) and SPP1 (h) signaling pathway.

FIGURE 2

Features of an in vitro model of FAP senescence. (a) Representative SA‐β‐gal staining images of control, etoposide challenged, and irradiated FAPs. (b) Quantification of the SA‐β‐gal positive cells. (c) Gene expression of cell type, cyclin‐dependent kinase inhibitors, SASP components, and a proliferation marker quantified by qPCR. N = 3 per group, one‐way analysis of variance (ANOVA). Error bars represent SEM. *, **, ***,**** denote p < 0.05, 0.01, 0.001, and 0.0001, respectively.

FIGURE 3

Core features of the senescent FAP transcriptome in vivo are replicated in vitro. (a) PCA plot of the transcriptomes of the six samples analyzed by RNAseq. (b) Volcano plot showing the differently expressed genes of senescent compared to control FAPs, with a log fold change (LFC) exceeding 1 and a p‐value less than 0.01. (c) Enriched KEGG pathways in senescent compared to control FAPs using the DEGs. (d) GSEA enrichment plot of the cytokine–cytokine receptor interaction pathways, chemokine signaling pathway, cell cycle, and DNA replication pathways. (e) Normalized expression level of markers for in vivo senescent FAPs, a subpopulation of FAPs positive for p16 and senescence‐related gene sets as reported (Zhang et al., 2022), in control FAPs and etoposide‐induced senescent FAPs. (f) GSEA enrichment plot of etoposide‐induced senescent FAPs compared to the control FAPs. Normalized enrichment score (NES) and FDR q‐value was calculated in GSEA software.

FIGURE 4

Senescent FAPs actively recruit macrophages in vitro. (a) Compared to non‐senescent (control) cells, etoposide‐ and irradiation‐induced senescent FAPs robustly secrete diverse proteins into their medium. (b) Protein concentration of CCL2 and OPN in the conditioned medium of control FAPs and etoposide‐ and irradiation‐induced senescent FAPs. (c) Schematic of the macrophage migration experiment, using control and senescent FAPs. (d) Quantification of the macrophages recruited by blank (cell‐free), non‐senescent control (CM), or senescent (SM) conditioned medium without and with antibodies to OPN (+Ab‐OPN) and CCL2 (+Ab‐CCL2). N = 3 per group, one‐way analysis of variance (ANOVA). Error bars represent SEM. **, ***, and **** denote p < 0.01, 0.001, and 0.0001, respectively.

FIGURE 5

Senescent FAPs promote macrophage polarization toward M2. (a–c) Gene expression of M0 (a) M1 (b) and M2 (c) macrophage markers in blank (cell‐free) medium, or medium with non‐senescent (CM), or senescent (SM) conditioned medium. (d–f) Relative cell subtype abundance quantified with flow cytometry for M0 (d), M1 (e), and M2 (f) macrophages. N = 4 per group, one‐way analysis of variance (ANOVA). Error bars represent SEM. * and **** denote p < 0.05 and 0.0001, respectively.