C57BL/6 J male mice were purchased from CLEA Japan (Shizuoka, Japan) and Oriental Yeast (Tokyo, Japan). To induce muscle atrophy, bilateral hind limbs were immobilized for 2 weeks using a steel wire, as described previously [15]. The wire was replaced as needed during the 2-week immobilization period. In certain experiments, mice were injected intraperitoneally with recombinant mouse IL-33 (2 μg/100 μl PBS; BioLegend, San Diego, CA, USA) or soluble ST2 (sST2) (5 μg/100 μl PBS; 1004-MR-050; R&D Systems, Minneapolis, MN, USA) twice a week. The first injection to the hindlimb immobilization model was performed at the start of immobilization. B6N.Cg-Tg(Pdgfra-cre/ERT)467Dbe/J (018280) and B6.Cg-Gt(ROSA)26Sortm14(CAG−tdTomato)Hze/J (007914) transgenic mice were purchased from Jackson Laboratories (Bar Harbor, ME, USA). These two transgenic mice were mated to generate platelet-derived growth factor receptor α (PDGFRα) reporter mice, in which PDGFRα-positive cells, including FAPs, express the tdTomato fluorescent protein upon tamoxifen induction. These mice were used to visualize FAPs in muscle sections (Fig. 3C). To induce Cre-mediated recombination, mice were intraperitoneally injected with tamoxifen (100 mg/kg; Toronto Research Chemicals, Toronto, Canada) dissolved in corn oil three times every other day.

Skeletal muscles from the hind limbs were used for flow cytometric analysis. Single-cell isolation was performed as described previously [16]. In short, after removal of the fat tissues, vessels, nerves, and tendons, the muscles were minced with forceps and digested in Hanks’ balanced salt solution containing 0.2% collagenase type II (Worthington Biochemical, Lakewood, NJ, USA). The digested samples were filtered through 70- and 40-μm cell strainers to remove the debris. Red blood cells were removed using Red Blood Cell Lysis Buffer (Roche Diagnostics, Rotkreuz, Switzerland). The following fluorochrome-conjugated monoclonal antibodies were used for cell sorting: anti-CD31 (102,405, 1:50; BioLegend), anti-CD45 (103,107, 1:200; BioLegend), anti-PDGFRα (FAB1062P, 1:10; R&D Systems), anti-Sca1 (108,113, 1:80; BioLegend), anti-mouse ST2 (FAB10041A, 1:20; R&D Systems) and Brilliant Violet 421 streptavidin (405,226, 1:666; Biolegend). The biotinylated SM/C2.6 monoclonal antibody was generously provided by Dr. S. Fukada [17]. Flow cytometry was performed using the CytoFLEX S flow cytometer (Beckman Coulter, Brea, CA, USA). Because of the easier availability of young mice, more animals were used in the analysis for young mice than for adult and aged mice.

SCs (CD31−/CD45−/SM/C2.5+) and FAPs (CD31−/CD45−/Sca1+/PDGFRα+) were isolated using a MoFlo XDP cell sorter (Beckman Coulter). The isolated FAPs and SCs were cultured in high-glucose Dulbecco’s modified Eagle medium (DMEM) (Nacalai Tesque, Kyoto, Japan), 20% fetal bovine serum, and antibiotics in Matrigel (Corning, Corning, NY, USA)-coated dishes, as described previously [16]. To induce myogenic differentiation of SCs, the medium was replaced with DMEM supplemented with 2% horse serum and antibiotics.

FAPs isolated from mice 3 days and 2 weeks after hind-limb immobilization were used to evaluate transcriptional changes in FAPs after immobilization in young and aged mice. FAPs collected from untreated mice were used as baseline controls. Two mice were used at each time point (6 young and 6 aged mice). The total RNA was extracted from cells using the RNeasy Micro Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. cDNA was prepared using a SMART-Seq v4 Ultra Low Input RNA kit and sequenced using a NovaSeq 6000 sequencer (Illumina, San Diego, CA, USA). After filtering genes with low expression levels or levels with high variation from each RNA sequence dataset, we screened genes with > twofold change compared with baseline controls. To determine the 10 DEGs with the largest fold change, genes with a maximum transcript per million of less than 10 transcripts in both the immobilized muscle samples and the corresponding baseline samples were excluded. The sequence data are available in the DDBJ Sequence Read Archive (https://www.ddbj.nig.ac.jp/) under accession number DRA016863.

The tibialis anterior (TA) muscles from untreated and immobilized mice were harvested and minced. Total RNA was isolated using Sepasol-RNA I Super G reagent (Nacalai Tesque), according to the manufacturer’s instructions. Total RNA was reverse-transcribed using ReverTra Ace Reverse Transcriptase (Toyobo, Osaka, Japan). PCR and quantification were performed using the Thunderbird qPCR Mix (Toyobo) and a 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). Gene transcript levels were normalized to the expression levels of Actb transcripts. The following primers were used: Fbxo32 (forward: TCTCCAGACTCTCTACACATCC; reverse: GAATGGTCTCCATCCGATACAC), Trim63 (forward: TACGTTGGTGCGAAATGAAA; reverse: AATCGCCAGTCACACAATGA), Il-33 (forward: ACTATGAGTCTCCCTGTCCTG; reverse: ACGTCACCCCTTTGAAGC), Il1rl1 (forward: TCTGTGGAGTACTTTGTTCACC; reverse: TCTGCTATTCTGGATACTGCTTTC), and Actb (forward: CTGAACCCTAAGGCCAACCGTG; reverse: GGCATACAGGGACAGCACAGCC). Because of the easier availability of young mice, more animals were used in the analysis for young mice than for aged mice.

Cells were treated with mouse recombinant IL-33 (30 ng/ml) for 1 h, fixed with 4% paraformaldehyde, and permeabilized with 0.2% Triton X-100/PBS. Fixed cells were incubated with rabbit anti-NF-κB p65 (8242, 1:200; Cell Signaling Technology, Danvers, MA, USA) at 4 °C overnight, followed by secondary antibody staining with 4′,6-diamidino-2-phenylindole (DAPI). Images were captured using an Olympus FSX100 fluorescence microscope and Olympus FSX-BSW software (Olympus, Tokyo, Japan).

To evaluate the cross-sectional area (CSA) of the muscle fiber, muscle tissues were snap-frozen in isopentane cooled with liquid nitrogen. Cryosections with thickness of 10 μm were fixed in acetone at − 20 °C for 20 min. Sections were washed with PBS and incubated with Blocking One (Nacalai Tesque) for 1 h, followed by incubation with rat anti-laminin-α2 (L0663, 1:500; Sigma-Aldrich, St. Louis, MO) at 4 °C overnight. Bound antibodies were detected using Alexa Fluor 488-conjugated anti-rat IgG (A21208, 1:1000; Invitrogen, Waltham, MA, USA). Images of entire sections were captured using an Olympus FSX100 fluorescence microscope and Olympus FSX-BSW software. The CSA of the muscle fibers was measured using ImageJ (National Institutes of Health, Bethesda, MD, USA).

Muscle tissues obtained from PDGFRα reporter mice were used for immunostaining of IL-33. Muscles were fixed in 2% paraformaldehyde for 2 h and consecutively immersed in 10% and 20% sucrose solutions for 1 h and 30% sucrose solution overnight. Muscles were snap-frozen in isopentane that was cooled with liquid nitrogen. Cryosections with thickness of 10 μm were rehydrated with PBS, permeabilized with 0.2% Triton X-100 for 10 min, and treated with proteinase K (1 μg/ml; Nacalai Tesque) for 5 min at room temperature. Sections were incubated with Blocking One for 1 h and subsequently with primary antibodies at 4 °C overnight. Goat anti-IL-33 antibody (AF3626-SP, 1:100; R&D Systems) and rabbit anti-RFP antibody (ab62341, 1:250; Abcam, Cambridge, UK) were used for immunostaining. For immunostaining of IL-33 in aged mice, sections of TA from wild-type mice were used and rabbit anti-PDGFRα antibody (sc338, 1:50; Santa Cruz Biotechnology, TX, USA) were used as primary antibody. Images were captured using an LSM980 confocal microscope (Carl Zeiss, Oberkochen, Germany). Three arbitrary areas (500 μm2) were chosen per sample, and cell counts were determined using Image J. For Hematoxylin and Eosin (H&E) staining, sections were fixed in acetone at − 20 °C for 20 min, washed in PBS and then stained in hematoxylin for 10 min and eosin for 7 min. The muscle sections were dehydrated in gradually increasing concentration of ethanol/ water solutions and fixed in 100% xylene.

GraphPad Prism version 6.05 (GraphPad, La Jolla, CA, USA) was used for the statistical analyses. Student’s t-test (two-tailed and assuming equal variances) was used to calculate the P values. For comparison of more than two groups, one-way ANOVA was used, followed by Tukey’s post hoc testing. P < 0.05 was considered statistically significant. Data are presented as mean ± standard error of the mean unless otherwise noted.