The Interfascicular Matrix of Energy Storing Tendons Houses Heterogenous Cell Populations Disproportionately Affected by Aging

Abstract

Energy storing tendons such as the human Achilles and equine superficial digital flexor tendon (SDFT) are prone to injury, with incidence increasing with aging, peaking in the 5^th decade of life in the human Achilles tendon. The interfascicular matrix (IFM), which binds tendon fascicles, plays a key role in energy storing tendon mechanics, and aging alterations to the IFM negatively impact tendon function. While the mechanical role of the IFM in tendon function is well-established, the biological role of IFM-resident cell populations remains to be elucidated. Therefore, the aim of this study was to identify IFM-resident cell populations and establish how these populations are affected by aging. Cells from young and old SDFTs were subjected to single cell RNA-sequencing, and immunolabelling for markers of each resulting population used to localise cell clusters. Eleven cell clusters were identified, including tenocytes, endothelial cells, mural cells, and immune cells. One tenocyte cluster localised to the fascicular matrix, whereas nine clusters localised to the IFM. Interfascicular tenocytes and mural cells were preferentially affected by aging, with differential expression of genes related to senescence, dysregulated proteostasis and inflammation. This is the first study to establish heterogeneity in IFM cell populations, and to identify age-related alterations specific to IFM-localised cells.

Figure 1.

11 cell clusters are present in equine tendon. (A) Uniform Manifold Approximation and Projection (UMAP) dimensionality reduction demonstrates the presence of 11 clusters, based on differential gene expression, namely FM tenocytes “FM”, IFM tenocytes “IFM”, mixed tenocytes “MixT”, mural cells “MuC1” and “MuC2”, vascular endothelial cells “EC-V”, lymphatic endothelial cells “EC-L”, T cells “TC”, macrophages “Mφ”, neutrophils “Neu”, and mast cells “MC”. Each cluster consists of cells originating from young and old donors (n=4/age group). (B) Dot plot showing genes used to differentiate and identify the clusters. Scale indicates average expression and ranges from grey = 0 to blue = 2, dot size indicates the percentage of cells expressing the gene. (C) Immunolabelling of longitudinal tendon sections reveal that all cells within fascicles, and some within the IFM compartment are positive for LOX, and cells within the IFM compartment show presence of TNXB, the endothelial cell marker PECAM1 validating the presence of both endothelial clusters EC-V and EC-L, the mural cell markers MYH11 and RGS5 for both mural cell clusters MuC1 and MuC2, and the immune cell marker CD74 for all immune cell clusters, TC, Mφ, Neu, MC (negative control in Supplementary FiFigure S3). Scale bar 75 µm. (D) Schematic of tendon demonstrating location of the different cell populations identified.

Figure 2.

IFM tenocytes and mural cells are preferentially affected by aging (n=4/age group). (A) The percentage of cells in the majority of clusters was predominantly unaffected by aging, except for an increase in the proportion of cells in mural cell cluster 2 (MuC2; p=0.049, unpaired t-test) and a decrease in the proportion of macrophages (Mφ; p=0.039, unpaired t-test). Significance is indicated by*. (B) The number of differentially expressed genes (DEGs) with aging varied between clusters, with the highest number of DEGs in MuC2 and IFM clusters (Wilcoxon Rank Sum test, log2FC threshold 0.25, adj. p < 0.05). Data are plotted on a log₁₀ scale. (C) Aging-related DEGs (Aging Atlas [26]) between young and old tendon cells were primarily associated with senescence and the senescence-associated secretory pathway (SASP). (D) Heatmap demonstrating IFM tenocyte and MuC2 clusters had the greatest number of aging related DEGs and in particular DEGs associated with senescence and SASP. Scale indicates number of genes and ranges from yellow = 0, to red = 10. (E) Venn diagrams showing the number of top 25 markers in each cluster that are maintained with aging. An age-dependent loss of the top genes characterising each cluster is observed for the IFM tenocyte, MixT and MuC2 clusters. (F) Heatmap showing IFM tenocyte and MuC2 clusters had most DE inflammation-related genes (GOTerm: inflammatory response, GO:0006954) with aging. Scale indicates log2FC and ranges from blue = -2.5, to white = 0, to red = 1.5.

Figure 3.

Sub-clustering reveals the presence of 5 tenocyte subclusters (n=4/age group). (A) Tenocyte subclusters were identified as FM_A, FM_B, IFM_A, IFM_B, and IFM_C, based on their marker expression. (B) Sankey diagram showing the provenance of each subcluster cell in relation to the original clusters. (C) Dot plot showing the top 5 differentially expressed markers and associated functions in each subcluster (Wilcoxon Rank Sum test, log2FC threshold 0.25, adj. p < 0.05). Scale indicates average expression and ranges from grey = -1 to blue = >1, dot size indicates the percentage of cells expressing the gene. (D) Heatmap showing average expression of the top 50 DE matrisomal genes across tenocyte subclusters in each tenocyte subcluster, with matrisome category indicated (Wilcoxon Rank Sum test, log2FC threshold 0.25, adj. p < 0.05). Scale indicates average expression and ranges from blue = 0, to white = 1, to red = 3. (E) Dot plot of established tenocyte lineage genes and their expression across the tenocytes subclusters. MKX and THBS4 are predominantly expressed in FM subclusters whilst COL1A1 and COL3A1 in IFM subclusters. Scale indicates average expression and ranges from grey = -1 to blue = >1, dot size indicates the percentage of cells expressing the gene.

Figure 4.

Tenocyte aging is predominantly observed in IFM subclusters (n=4/age group). (A) The percentage of cells in the subclusters was not statistically significantly affected by aging, despite an apparent decrease in IFM_C cell number. (B) UMAP of tenocyte subclusters in young and old tendons; the distribution of cells in subcluster IFM_A changes with aging. (C) The number of DEGs with aging varied between subclusters, with the highest number of DEGs in the IFM_A clusters (Wilcoxon Rank Sum test, log2FC threshold 0.25, adj. p < 0.05). Data are plotted on a log₁₀ scale. (D) Heatmap showing DE of selected core matrisome genes with aging in each tenocyte subcluster (Wilcoxon Rank Sum test, log2FC threshold 0.25, adj. p < 0.05). Scale indicates log2FC and ranges from blue = -3, to white = 0, to red = 2. (E) UMAP of tenocyte subclusters with prediction of differentiation state in the young and old tenocytes and (F) raincloud plot of tenocyte subclusters in young samples by order of least differentiation (CytoTRACE). Subcluster IFM_A has the largest number of least differentiated cells in young samples (raincloud plot and UMAP), which are located particularly in the elongated tip at the bottom of the subcluster. Whereas in old samples, this elongated tip of the IFM_A subcluster is absent and a reduction in the number of least differentiated cells is noted. Scale ranges from blue = more differentiated to red = least differentiated through green, yellow and orange. (G) UMAP and violin split plot of POSTN and TPPP3 expression in tenocyte subclusters in young and old tendons. POSTN and TPPP3 expression, which has been associated with a progenitor phenotype in tendon cells, is observed in young samples in the elongated tip at the bottom of subcluster IFM_A and it is significantly decreased with aging. Color indicates expression and ranges from grey = 0 to blue = 3.

Figure 5.

FM and IFM tenocyte clusters become the primary sources of outgoing signalling in aging (n=4/age group). (A) Heatmap of differential secreted signalling interaction strength among tendon clusters following aging, along with the heatmaps of secreted signalling interaction strength among clusters in the young and old tendons. With aging, the FM and IFM tenocyte clusters showed the largest change in outgoing secreted signalling, increasing their signalling interaction strength and becoming the primary sources of outgoing signalling in old tendon cells. Scales indicate relative values and interaction strength and range from blue to red and from white to red, respectively. (B) Secreted signalling pathways enriched in young (red) and old (blue) tendon. (C) Outgoing and incoming signalling patterns for each secreted signalling pathway and cluster in young and old tendon. Scales indicate relative strength and range from white to green and from white to blue. Significant interactions are identified using a permutation test, p < 0.05.