Age-associated Senescent - T Cell Signaling Promotes Type 3 Immunity that Inhibits the Biomaterial Regenerative Response

Abstract

Aging is associated with immunological changes that compromise response to infections and vaccines, exacerbate inflammatory diseases and can potentially mitigate tissue repair. Even so, age-related changes to the immune response to tissue damage and regenerative medicine therapies remain unknown. Here, it is characterized how aging induces changes in immunological signatures that inhibit tissue repair and therapeutic response to a clinical regenerative biological scaffold derived from extracellular matrix. Signatures of inflammation and interleukin (IL)-17 signaling increased with injury and treatment both locally and regionally in aged animals, and computational analysis uncovered age-associated senescent-T cell communication that promotes type 3 immunity in T cells. Local inhibition of type 3 immune activation using IL17-neutralizing antibodies improves healing and restores therapeutic response to the regenerative biomaterial, promoting muscle repair in older animals. These results provide insights into tissue immune dysregulation that occurs with aging that can be targeted to rejuvenate repair.

Keywords: aging; biomaterials; senescence; tissue engineering.

Figure 1.

Aging shifts the immune response to ECM biomaterials from pro-regenerative to pro-inflammatory in muscles. a) Schematic illustration of experimental design including no injury control, volumetric muscle loss injury (VML) treated with saline and VML treated with ECM in young (6 wk) and old (72 wk) mice. b) Quantification of type 1, type 2 or type 3 immune response-related genes in muscle one week after injury or ECM treatment (n = 3 all young; n = 3 old ECM; n = 4 old No Injury, old Saline) using PCR. c) Quantification of immune cells and expression of IL17A in γδ T cells in muscle one week after treatments as determined by spectral flow cytometry (n = 5 for immune cell quantification in wild type mice; n = 4 for IL17A expression in IL17A-GFP mice). IL17A expression is presented as mean fluorescence intensity (MFI). d) Transverse section of the quadricep muscle 1 week after injury or ECM stained with H&E. The black arrow indicates the ectopic adipogenesis region. Two-way ANOVA with Sidak’s multiple comparisons test within the treatment group (b,c) or unpaired two-tailed t-test (c, IL17A expression). For all bar graphs, data are mean ± s.d.

Figure 2.

Aging impairs immune-stromal communication in muscle, and aged fibroblasts exhibit a distinct fibrosis gene signature. a) UMAP overview of cell clusters identified using scRNA-seq dataset on muscle from young and old animals 1week after injury or treatment (3 mice pooled for each condition). Clusters were self-assembled into three signaling modules enriched in fibroblast, antigen-processing, and immune-tissue clusters. b) Gene set enrichment analysis of old mice-specific transcription factors (TF) in muscle. Adjusted p values (log 10) of significant GO terms are shown, c) Age-specific global signaling network in muscles. Three distinct modules are labelled on the basis of enrichment of receptors and transcription factors by each cluster. d) Chord plot (left) and heatmap (right) of predicted cluster-cluster signaling for selected clusters determined by Domino for young or old muscle. Pairwise interactions are shown between ligand and receptor genes expressed by fibroblast, myeloid/macrophage, endothelial/pericyte, and T cell clusters. The values shown are the summed z-scored expression values for ligands (in the ligand-cluster) targeting receptors predicted to be activated in the receptor-cluster. Higher values indicate increased expression of ligands predicted to be active for a given receptor cluster. The width of the chord shows the strength of the interaction, e) NMF-CoGAPS analysis of fibroblast populations in muscle. Region of cells expressing high levels of the gene sets are circled, p values in NMF-CoGAPS were determined using Mann-Whitney U test and adjusted with false discovery rate correction for multiple testing. To evaluate age-associated changes irrespective of the treatments (injury or ECM), we have combined 3 treatment groups (No injury, Saline, ECM) within the respective age groups, and labelled as Young or Old (b-e). Data was generated with Drop-seq (n = 3 per group, pooled). Major findings were further identified and validated using the 10X Genomics platform (n = 2–3 per group, hashed; Figures S29 and S30, Supporting Information).

Figure 3.

Aging induces a Th17-associated immune skewing, and injury and ECM treatment promote local and systemic type 3 immune response. a) Volcano plot of genes expressed in aged lymph node normalized to those in young lymph node (top left) assessed by Nanostring. Gene pathway scoring for type 17 helper T cell (Th17) pathways (top right) or differentially expressed genes for NF-κb/TNFα or Th17-associated pathways (bottom) based on Nanostring analysis are shown (n = 6). b) Multiparametric flow cytometry quantification of IFNγ⁺ or IL17A⁺ CD4 (n = 9) or γδ T cells (n = 6) in the lymph nodes from young or aged animals without injury or treatment. c) Multiparametric flow cytometry quantification of IFNγ⁺ or IL17A⁺ CD4 or γδ T cells in the blood from young or aged animals without injury or treatment (n = 4). d) Quantification of Th17 genes in lymph node one week after injury or ECM treatment using PCR (n = 3 young No injury, old Saline; n = 4 Young Saline, old No injury, old ECM; n = 5 young ECM). e) Quantification of Th17-associated genes in the lymph nodes using PCR. A schematic of the downstream production of matrix metalloproteinase (MMP) and chemokine from Th17 is shown (top left; n = 3 young No injury, old Saline; n = 4 young Saline, old ECM; n = 5 young ECM, old No injury). f) Representative images (top) and quantification (bottom) of flow cytometry data comparing IL17A⁺ γδ⁺ or CD4⁺ T cells between young and old animals in the lymph node one week after injury or ECM treatment (n = 3 old No injury, old Saline; n = 4 all young, old ECM). Unpaired two-tailed t-test (b,c), two-way ANOVA with Sidak’s multiple comparisons test within the treatment group (d-f). For all bar graphs, data are mean ± s.e.m (b) or s.d. (c-f).

Figure 4.

Age-associated signaling communication between senescent stromal cells and immune cells induces type 3 immune skewing in old animals. a) Senescence scores for cells shown on UMAP (top) and violin plot grouped by cluster and age (bottom). CD45⁻ stromal clusters are marked with dotted line, and a cluster with the highest senescence signature is marked red. (b-c) Chord plot (left) and heatmap (right) of predicted cluster-cluster signaling for selected clusters in young (b) or old (c) muscle. Pairwise interactions are shown between ligands of CD45⁻ clusters, fibroblasts and endothelial/pericyte, and receptors of all other clusters. The values shown are the summed z-scored expression values for ligands (in the ligand-cluster) targeting receptors predicted to be activated in the receptor-cluster. Higher values indicate increased expression of ligands predicted to be active for a given receptor cluster. The width of the chord shows the strength of the interaction. d, Identification of ligand-receptor-transcription factor (TF) signaling network between senescent fibroblast (Cart-like fibroblast) and T cells in old animals (left), and its UMAP representation (right). SCENIC is used to estimate TF modules and activation scores. Receptor expressions are correlated with TF activation scores with exclusion of receptors present in the TF modules. Public receptor-ligand databases are used to identify ligands activating receptors. e) Illustration of coculture platform designed to study senescent fibroblast-T cell communication in vitro (top) and multiparametric flow quantification of CD4 T cells after coculture (bottom; n = 4). Two-way ANOVA with Tukey’s multiple comparisons test in (e). To evaluate age-associated signatures irrespective of the treatments (injury or ECM), we combined 3 treatment groups (No injury, Saline, ECM) within the respective age groups, labeled as Young or Old (a-d). Data was generated with Drop-seq (a-d). These findings were further identified and validated using the 10X Genomics platform (n = 2–3 per group, hashed; Figures S29 and S30, Supporting Information). For all bar graphs, data are mean ± s.d.

Figure 5.

Aging is associated with increased Th17 effector T cells with a unique secretome profile. a) Quantification and representative plots of flow cytometry analysis on CD4 T cells isolated from lymph nodes or spleens of young and old animals. Naïve phenotype (CD4⁺CD44^-CD62L^-), effector phenotype (CD4⁺CD44⁺CD62L⁻) and Th17 cells (CD4⁺CD44⁺CD62L⁻RORγt⁺Tbet⁻) are shown (n = 3). b) Schematic illustration of naïveT cell isolation and Th17 differentiation in vitro (left) and quantification of flow cytometry analysis on the undifferentiated and differentiated CD4 T cells (right; n = 3). c) Representative images (left) and quantification (right) of the proteome profiler performed on cell culture supernatant from Th17-differentiated CD4 T cells from young and old mice (n = 3). Protein molecules with significant differences in pixel densities compared to young animals are labeled and quantified using iBright Analysis Software. Unpaired two-tailed t-test (a,c), two-way ANOVA with Sidak’s multiple comparisons test (b). For bar graphs, data are mean ± s.d. (a,b) or s.e.m. (c).

Figure 6.

Local IL17 suppression rejuvenates the type 2 immune response to injury and ECM to restore tissue repair and reduce fibrosis in old animals, a) Schematic illustration of experimental design (left) and quantification of flow cytometry data for IL4⁺ CD4T cells and eosinophils in muscle 3 weeks after injury (right; n = 6 No injury; n = 8 Isotype; n = 3 αIL17A, αIL17F). b) Experimental schematics (left), and representative images flow cytometry showing IL4⁺ CD45 or CD4 T cells (middle) and quantification of IL4⁺ cell populations in muscle six weeks after injury and ECM treatment (right; n = 4). c) Quantification of genes associated with fibrosis or adipogenesis in muscle using PCR (top left; n = 3), and transverse section of the quadricep muscle six weeks after injury stained with Masson’s Trichrome, d) Immunofluorescence images of the quadricep muscle six weeks after injury stained with dystrophin (top) or laminin (bottom). Quantification of muscle fibers with central nuclei are shown (right; n = 3). Nuclei were stained with DAPI (represented in yellow). One-way ANOVA with Tukey’s multiple comparisons test (a-d). For all bar graphs, data are mean ± s.e.m (a) or s.d. (b-d).