Single-nucleus transcriptomic analysis reveals the regulatory circuitry of myofiber XBP1 during regenerative myogenesis

Abstract

Endoplasmic reticulum (ER) stress-induced unfolded protein response (UPR) is activated in skeletal muscle under multiple conditions. However, the role of the UPR in the regulation of muscle regeneration remains less understood. We demonstrate that gene expression of various markers of the UPR is induced in both myogenic and non-myogenic cells in regenerating muscle. Genetic ablation of X-box binding protein 1 (XBP1), a downstream target of the Inositol requiring enzyme 1α (IRE1α) arm of the UPR, in myofibers attenuates muscle regeneration in adult mice. Single nucleus RNA sequencing (snRNA-seq) analysis showed that deletion of XBP1 in myofibers perturbs proteolytic systems and mitochondrial function in myogenic cells. Trajectory analysis of snRNA-seq dataset showed that XBP1 regulates the abundance of satellite cells and the formation of new myofibers in regenerating muscle. In addition, ablation of XBP1 disrupts the composition of non-myogenic cells in injured muscle microenvironment. Collectively, our study suggests that myofiber XBP1 regulates muscle regeneration through both cell-autonomous and -non-autonomous mechanisms.

Keywords: Biochemistry; Cell biology; Genetics; Transcriptomics.

Graphical abstract

Figure 1

Gene expression of UPR molecules during muscle regeneration The scRNA-seq dataset (GSE143437) was analyzed using R software (v4.2.2). Dot plot showing the changes in gene expression of ER stress/UPR molecules in different cell types and at different time points during muscle regeneration.

Figure 2

Single nucleus RNA-sequencing (snRNA-seq) analysis identifies different cell types in regenerating muscle (A) TA muscle of Xbp1^fl/fl and Xbp1^mKO mice was injured using intramuscular injection of 1.2% BaCl₂ solution. Schematics presented here show TA muscle histological analysis or isolation of nuclei followed by performed by snRNA-seq. (B) Representative photomicrographs of H&E-stained transverse sections of 5d-injured TA muscle of Xbp1^fl/fl and Xbp1^mKO mice. Scale bar, 50 μm. (C) Average myofiber cross sectional area (CSA) and (D) proportion of myofibers containing two or more centrally located nuclei. n = 3 mice in each group. Data are presented as mean ± SEM. ∗p ≤ 0.05, values significantly different from corresponding muscle of Xbp1^fl/fl mice analyzed by unpaired Student’s t test. Pre-processed 10X Genomics sequencing data of 5d-injured muscle of Xbp1^fl/fl and Xbp1^mKO was analyzed on R software for the presence of nuclei of different cell types. (E) Dot plot representing the proportion of cells and average expression of known genes associated with distinct cell types. (F) UMAP plot representing manually annotated clusters for cell type identity. (G) Validation of cellular identities by differentially expressed genes (DEG) followed by pathway enrichment analysis. Representative heatmap showing top 10 enriched genes per cluster (left panel) and enriched terms and cell type identities for corresponding clusters (right panel).

Figure 3

XBP1 regulates proteolytic pathways and mitochondrial OXPHOS levels in regenerating muscle (A) The integrated Seurat object of the injured muscles of Xbp1^fl/fl and Xbp1^mKO mice were classified based on myogenic and non-myogenic nuclei. (B) Proportion of nuclei in different clusters of myogenic cells in 5d-injured TA muscle of Xbp1^fl/fl and Xbp1^mKO mice. (C) Feature plot showing gene expression of Xbp1 in all muscle nuclei or mature myofiber (Myo) cluster (right panel) of Xbp1^fl/fl and Xbp1^mKO mice. (D) Differentially expressed genes (DEG) in Myo clusters were identified and used for pathway enrichment analysis using Metascape. Bar graphs show enriched biological processes and pathways associated with upregulated (red) and downregulated (blue) genes in Xbp1^mKO compared to Xbp1^fl/fl mice. (E) Protein-protein interaction plots of the upregulated genes associated with respiratory electron chain/ATP synthesis pathway. (F) Violin plot showing downregulation of gene expression of protein ubiquitination-related molecules and (G) Feature plots showing downregulation of gene expression of autophagy-related molecules in Xbp1^mKO mice compared to Xbp1^fl/fl mice. (H) Primary myoblast cultures were incubated in differentiation medium for 48 h followed by transfection with control or XBP1 siRNA. Myotubes were collected after 24 h of transfection and cell lysates were used for immunoblotting. Immunoblots presented show protein levels of MAFbx (Fbxo32), MuRF1 (Trim63), Beclin1, LC3B, OXPHOS complexes, sXBP1, and unrelated protein GAPDH in myotubes transfected with control or XBP1 siRNA. (I) Quantification of protein levels of MAFbx, MuRF1, Beclin1, LC3bI and II, sXBP1 and OXPHOS complexes CII, III and V. n = 3 biological replicates. Data are presented as mean ± SEM. ∗p ≤ 0.05; values significantly different from cultures transfected with control siRNA analyzed by unpaired Student’s t test.

Figure 4

Myofiber XBP1 regulates formation of new myofibers during muscle regeneration (A) Pathway enrichment analysis of differentially expressed genes in clusters of eMyHC⁺ regenerating myonuclei (Regmyo1 and Regmyo2) of Xbp1^fl/fl and Xbp1^mKO mice. Violin plots show gene expression of the (B) upregulated molecules associated with respiratory electron transport system and (C) downregulated molecules involved in muscle differentiation and structure development. (D) Enrichment analysis in TRRUST database showing transcriptional regulators of the upregulated and downregulated genes in Xbp1^mKO mice compared to Xbp1^fl/fl mice. (E) Transverse sections of 5d-injured TA muscle of Xbp1^fl/fl and Xbp1^mKO mice were immunostained for eMyHC and laminin protein. Nuclei were counterstained by DAPI. Representative photomicrographs demonstrating eMyHC⁺ regenerating myofibers in 5d-injured TA muscle. Scale bar, 50 μm. Quantitative analysis of (F) eMyHC⁺ myofibers per laminin⁺ myofibers, (G) number of eMyHC⁺ myofibers per field, and (H) average cross-sectional area of eMyHC⁺ laminin⁺ myofibers. n = 3 mice in each group. Data are presented as mean ± SEM and analyzed by unpaired Student’s t test. ∗p ≤ 0.05; values significantly different from injured TA muscle of Xbp1^fl/fl mice.

Figure 5

XBP1 regulates the myogenesis trajectory and alters the transcriptomic profiles of muscle progenitor cells during muscle regeneration (A) Trajectory path of myonuclei along the pseudotime axis resembling the myogenic lineage was plotted using the Monocle2 package. UMAP plots show trajectories of myonuclei in injured TA muscles of Xbp1^fl/fl and Xbp1^mKO mice. (B) Representative images of uninjured and 21d-injured TA muscle sections of Xbp1^fl/fl and Xbp1^mKO mice after immunostaining for Pax7 and laminin protein. DAPI was used to identify nuclei. Scale bar, 50 μm. (C) Quantification of number of Pax7⁺ cells per unit area in uninjured and 21d-injured muscle of Xbp1^fl/fl and Xbp1^mKO mice. (D) Immunoblots, and (E) quantification of levels of Pax7, MyoD and Myogenin protein in uninjured and 5d-injured TA muscle of Xbp1^fl/fl and Xbp1^mKO mice. n = 3–4 mice in each group. Data are presented as mean ± SEM. ∗p ≤ 0.05, values significantly different from corresponding uninjured muscle of Xbp1^fl/fl and Xbp1^mKO mice, and #p ≤ 0.05, values significantly different from 5d- or 21d-injured muscle of Xbp1^fl/fl mice analyzed by two-way ANOVA, followed by Tukey’s multiple comparison test. (F) Heatmaps showing upregulated and downregulated gene sets along the pseudotime axis.

Figure 6

Myofiber XBP1 regulates transcriptomic profiles of satellite cells in regenerating muscle Nuclei of satellite cell clusters (MuSCs1 and 2) in the snRNA-Seq analysis were used to identify differentially expressed genes followed by pathway enrichment analysis in Xbp1^mKO mice compared to Xbp1^fl/fl mice. (A–D) Bar graph showing enriched pathways associated with upregulated (red) and downregulated (blue) genes in MuSCs1 and MuSCs2 clusters respectively. Yellow boxes indicate pathways used for assessment of gene expression. Average gene expression of some of the (B–E) downregulated and (C–F) upregulated molecules involved in the enriched pathways.

Figure 7

Myofiber XBP1 regulates distinct pathways in satellite cells during muscle regeneration (A) Pathway enrichment analysis of differentially expressed genes across the combined cluster of satellite cell nuclei (MuSCs1 and MuSCs2). (B) Ridge plot showing gene expression of mitochondrial respiration associated molecules in satellite cells of Xbp1^fl/fl and Xbp1^mKO mice. (C) Enrichment of transcriptional regulators of upregulated and downregulated molecules in combined and individual clusters of satellite cell nuclei performed using TRRUST database. Violin plots for some of the deregulated genes (D) regulated by Myod1/Mef2c; and associated with (E) TGFβ/SMAD signaling pathway, (F) autophagy-lysosome pathway, and (G) Notch signaling pathway in nuclei of MuSCs1 and 2 clusters.

Figure 8

XBP1 in myofibers regulates the abundance of non-myogenic cells during regenerative myogenesis (A) Proportion of non-muscle cell nuclei in injured TA muscle of Xbp1^fl/fl and Xbp1^mKO mice. Differential gene expression analysis was performed in nuclei of fibro-adipogenic progenitor cells (FAPs), endothelial cells (Endo), and macrophages (Macro) clusters. Pathway enrichment analysis of deregulated genes in (B) FAPs, (C) Endo, and (D) Macro clusters of Xbp1^mKO mice compared to Xbp1^fl/fl mice. (E) Violin plot showing gene expression levels of molecules related to regulation of endocytosis. (F) Heatmap showing average gene expression of cell surface markers and cytokines secreted by macrophages.