Deciphering the age-dependent changes of pulmonary fibroblasts in mice by single-cell transcriptomics

Abstract

Background and objectives: The heterogeneity of pulmonary fibroblasts, a critical aspect of both murine and human models under physiological and pathological conditions, is well-documented. Yet, consensus remains elusive on the subtypes, lineage, biological attributes, signal transduction pathways, and plasticity of these fibroblasts. This ambiguity significantly impedes our understanding of the fibrotic processes that transpire in lung tissue during aging. This study aims to elucidate the transcriptional profiles, differentiation pathways, and potential roles of fibroblasts within aging pulmonary tissue. Methods: We employed single-cell transcriptomic sequencing via the 10x Genomics platform. The downstream data were processed and analyzed using R packages, including Seurat. Trajectory and stemness of differentiation analyses were conducted using the Monocle2 and CytoTRACE R packages, respectively. Cell interactions were deciphered using the CellChat R package, and the formation of collagen and muscle fibers was identified through Masson and Van Geison staining techniques. Results: Our analysis captured a total of 22,826 cells, leading to the identification of fibroblasts and various immune cells. We observed a shift in fibroblasts from lipogenic and immune-competent to fibrotic and myofibroblast-like phenotype during the aging process. In the aged stage, fibroblasts exhibited a diminished capacity to express chemokines for immune cells. Experimental validation confirmed an increase of collagen and muscle fiber in the aged compared to young lung tissues. Furthermore, we showed that TGFβ treatment induced a fibrotic, immunodeficient and lipodystrophic transcriptional phenotype in young pulmonary fibroblasts. Conclusion: We present a comprehensive single-cell transcriptomic landscape of lung tissue from aging mice at various stages, revealing the differentiation trajectory of fibroblasts during aging. Our findings underscore the pivotal role of fibroblasts in the regulation of immune cells, and provide insights into why age increases the risk of pulmonary fibrosis.

Keywords: aging; cell interaction; lung fibrosis; myofibroblasts; pulmonary fibroblasts; single-cell transcriptomics.

FIGURE 1

Commencing with the classification of cells within aging mouse lung tissue via scRNA-seq. (A). Schematic diagram of mouse subjected to aging lung tissue. P10, postnatal day 10; 2 M, 2 months years old; 10 M, 10 months years old; 18 M, 18 months years old. (B). tSNE projection of all 22,826 sequenced mouse cells, showing the partitioning of 15 cell types by scRNA-seq. (C). Violin plot showing special marker genes for each cell type in the mouse lung dataset. (D). Bar chart for proportions of the 15 major cell types in lung tissues.

FIGURE 2

Transcriptional characteristics associated with the aging process of lung fibroblasts. (A). Volcano plot for aging differentially expressed genes were changed in 18 M group compared to the 2 M group for major cell types by Seurat analysis. Red, upregulated (LogFC > 0.5, adjusted p-value < 0.01); blue, downregulated group (LogFC < −0.5, adjusted p-value < 0.01); (B). Representative aging specific gene modules and pathways enriched in aging differentially expressed genes based on GO-biological process functional enrichment analysis.

FIGURE 3

Tissue characteristics of aging fibroblasts. (A). Heatmap showing the variable genes in aging mouse lung pFBs by z-score normalization analysis. Color scale of red to blue indicates z-score. (B). Lipid stain/Masson stain/Van Geison stain showing tissue characteristics changes in 18 M group compared to the 2 M group of lung tissue. Scale bars, 200 μm and 400 μm. (C). tSNE plot for BLM model pFB reclusters by scRNAseq analysis. (D). Feature plot showing on tSNE projection of pFB reclusters. (E). Heatmap showing BLM scRNAseq pFBs reclusters and aging pFBs reclusters relevance by pearson correlation analysis.

FIGURE 4

The differentiation potential of fibroblasts in lung tissue during aging. (A). Bar chart for aging pFBs stemness of differentiation by CytoTRACE analysis. (B). Pseudotime density showing the differentiation of fibroblasts from normal fibroblasts to myofibroblasts by Monocle2 analysis. (C). Complex trajectory plot showing the branch of aging pFBs. (D). Pseudotime trajectories showing cell states in all group. (E). Pseudotime trajectories gradients showing differentiation direction of pFBs. (F). Pseudotime trajectories showing cell states splite by each group. (G). Col1a1, Col1a2, Col3a1, and Col5a1 expression on cell pseudotime trajectory.

FIGURE 5

Anticipating the communication dynamics between fibroblasts and immune cells at various stages of aging. (A). Heatmap for significant regulons for each pFBs by SCENIC analyze. (B). Bar chart showing significant signaling pathways were ranked based on their differences in overall information flow within the inferred networks between aging lung tissue by cellchat analysis. (C). Circle and hierarchy plot showing total, 2 M, 18 M cell-cell communication in CCL signaling pathway (left, middle, right). (D). Circle and hierarchy plot showing total, 2 M, 18 M cell-cell communication in CXCL signaling pathway (left, middle, right).

FIGURE 6

Activation of TGFβ pathway induces the transcriptional signature of aging in cultured pulmonary fibroblasts. (A). Violin plots showing the expression of Il1r1 and Tlr2 in various pulmonary cell clusters as shown. (B). Violin plots showing changes of genes related to lipogenesis (Ly6a, Apoe), immune (Il1r1, Cxcl1, Cxcl12, Ccl2, and C1ra), and fibrosis (Col1a1, Eln) in pulmonary fibroblasts from 2-month and 18-month mice. (C–N). Primary pulmonary FBs were treated with recombinant mouse TGFβ at a concentration of 3 ng/mL for 2 days and then subjected to qRT-PCR of indicated genes (n = 3/group). All error bars indicate mean ± SEM; *p < 0.05, **p < 0.01, ***p < 0.001.