Il-6 signaling exacerbates hallmarks of chronic tendon disease by stimulating reparative fibroblasts

Abstract

Tendinopathies are debilitating diseases currently increasing in prevalence and associated costs. There is a need to deepen our understanding of the underlying cell signaling pathways to unlock effective treatments. In this work, we screen cell signaling pathways in human tendinopathies and find positively enriched IL-6/JAK/STAT signaling alongside signatures of cell populations typically activated by IL-6 in other tissues. In human tendinopathic tendons, we also confirm the strong presence and co-localization of IL-6, IL-6R, and CD90, an established marker of reparative fibroblasts. To dissect the underlying causalities, we combine IL-6 knock-out mice with an explant-based assembloid model of tendon damage to successfully connect IL-6 signaling to reparative fibroblast activation and recruitment. Vice versa, we show that these reparative fibroblasts promote the development of tendinopathy hallmarks in the damaged explant upon IL-6 activation. We conclude that IL-6 activates tendon fibroblast populations which then initiate and deteriorate tendinopathy hallmarks.

Keywords: assembloid; fibroblast; human; interleukin-6; mouse; regenerative medicine; scleraxis; stem cells; tendinopathy; tendon.

Figure 1.. Transcriptome analysis of up- and downregulated genes and pathways in tendinopathic compared to normal control human tendons (non-sheathed).

(A) Volcano plot of differentially expressed genes (DEGs) comparing tendinopathic to normal human tendons. Genes colored in red have a log2 (fold change)>1, a p-value<0.05, and are therefore considered to be significantly increased in tendinopathic tendons. Genes colored in blue have a log2 (fold change)<–1, a p-value<0.05, and are therefore considered to be significantly decreased in tendinopathic tendons. The log2 and p-value thresholds are represented by the dashed lines. Annotated genes are part of the IL-6 cytokine superfamily, the IL-6 signaling cascade, or involved in matrix turnover. (B) Unsupervised hierarchical clustering of expression values from members of the IL-6 cytokine superfamily, their receptors, and parts of the IL-6 signaling cascade (biological replicates: N=18 normal, N=18 tendinopathic). Genes are clustered by color with positive (red) or negative (blue) row-scaled z-scores. (C) Dotplot showing significantly enriched gene sets (p-value<0.05) as determined by gene set enrichment analysis (GSEA) based on the MSigDB human hallmark gene sets. The color of the circles represents their p-value, the size the number of enriched genes (count), and the position on the x-axis the enrichment score as well as its direction. (D) GSEA plot for the IL-6/JAK/STAT3 signaling hallmark contained in the MSigDB human hallmark gene sets. The green line traces the running enrichment score on the y-axis while going down the rank of genes listed on the x-axis, the black lines standing in blue and red bars indicate the locations of the genes related to the pathway in the ranked list, and the gray histogram shows the running list score across the ranks. (E) Dotplot showing the top 10 gene ontology (GO) gene sets for biological processes most significantly enriched by overlapping DEG sets. The color of the circles represents their adjusted p-value (false discovery rate [FDR]), the size the number of enriched genes (count), and the position on the x-axis the number of enriched genes in ratio to the total number of genes annotated to the gene set (gene ratio). (F) Dotplot showing significantly enriched fibroblast signature gene sets (p-value<0.05) as determined by GSEA based on the MSigDB human cell-type signature gene sets. The color of the circles represents their p-value, the size the number of enriched genes (count), and the position on the x-axis the enrichment score as well as its direction.

Figure 1—figure supplement 1.. Principal component analysis (PCA) plots of the human tendon microarray data.

(A) Principal components 1 and 2 for the full dataset with tendinopathic (red) and normal (blue) tendons. Tendons surrounded by a sheath in vivo are delineated by a dashed border. (B) Principal components 1 and 2 for the same dataset after excluding sheathed tendons as delineated in A.

Figure 1—figure supplement 2.. Detailed transcriptome analysis of up- and downregulated genes and pathways in normal and human tendinopathic tendons (non-sheathed).

(A) Dotplot showing the top 20 GO gene sets for biological processes significantly enriched by transcripts increased in human tendinopathic tendons. (B) Dotplot showing all the GO gene sets for biological processes enriched by transcripts decreased in human tendinopathic tendons. In all the dotplots, the color of the circles represents their adjusted p-value, the size the number of enriched genes (count), and the position on the x-axis the number of enriched genes in ratio to the total number of genes annotated to the gene set (gene ratio).

Figure 1—figure supplement 3.. Detailed transcriptome analysis of up- and downregulated genes and pathways in normal and human tendinopathic tendons (non-sheathed).

(A) Dotplot showing all the GO gene sets for molecular functions enriched by transcripts increased in human tendinopathic tendons. (B) Dotplot showing all the GO gene sets for molecular functions enriched by transcripts decreased in human tendinopathic tendons. In all the dotplots, the color of the circles represents their adjusted p-value, the size the number of enriched genes (count), and the position on the x-axis the number of enriched genes in ratio to the total number of genes annotated to the gene set (gene ratio).

Figure 2.. Distribution of IL-6, CD90, and IL-6R in normal control and tendinopathic human tendons (non-sheathed).

(A) Illustrative depiction of the origins of the tendons used in this experiment. Normal control tendon tissues were taken from semitendinosus and gracilis tendons leftover from anterior cruciate ligament (ACL) reconstruction surgery. Tendinopathic tissues were taken from painful shoulders during surgery (Figure 2—figure supplement 1). (B) Representative fluorescence microscopy images of normal control (left) and tendinopathic tendons (right) stained with DAPI (blue) and an IL-6 antibody (red). Boxplots depict the quantified co-localization of DAPI and IL-6 (IL-6⁺ cells) calculated as percentage of the total number of cells. Biological replicates: N=8. (C) Representative fluorescence microscopy images of normal control (left) and tendinopathic tendons (right) stained with DAPI (blue) and a CD90 antibody (green). Boxplots depict the quantified co-localization of DAPI and CD90 (CD90⁺ cells) calculated as percentage of the total number of cells. Biological replicates: N=7. (D) Representative fluorescence microscopy images of normal control (left) and tendinopathic tendons (right) stained with DAPI (blue) and an IL-6R antibody (magenta). Boxplots depict the quantified co-localization of DAPI and IL-6R (IL-6R⁺ cells) calculated as percentage of the total number of cells. Biological replicates: N=7. In all boxplots, each datapoint was calculated from eight representative fluorescence microscopy images taken from the same sample. The colored datapoint matches the presented fluorescence microscopy image. The upper and lower hinges correspond to the first and third quartile (25th and 75th percentile), the middle one to the median, the whiskers extend from the hinges no further than 1.5 times the interquartile range, and the points beyond the whiskers are treated as outliers. Results of the statistical analysis are indicated as follows: ^n.s.p ≥ = 0.05, *p<0.05, **p<0.01. The applied statistical test was the Student’s t-test.

Figure 2—figure supplement 1.. Metadata of the human patient-derived tissues analyzed with fluorescence microscopy.

Tendinopathic tissues were gathered from a total of 10 patients with painful biceps tendons. The photographic images depict the tendons from which the sections were collected. Normal control tissues were gathered from the leftover semitendinosus or gracilis tendons of a total of eight patients undergoing anterior cruciate ligament (ACL) reconstruction surgery. No images were taken from the normal control tendons.

Figure 3.. Co-localization of IL-6, IL-6R, CD90, and CD68 in tendinopathic human tendons (non-sheathed).

(A) Illustrative depiction of the origin of the tendon used in this experiment (painful biceps tendons, Figure 2—figure supplement 1). (B) Representative fluorescence microscopy images of tendinopathic tendons stained with DAPI (blue), an IL-6 antibody (red), and either a CD90 antibody (left image, green) or a CD68 antibody (right image, green). Boxplots depict the quantified co-localization of DAPI, IL-6, and either CD90 (biological replicates: N=7) or CD68 (biological replicates: N=8) calculated as percentage of the number of DAPI⁺IL-6⁺ cells. (C) Representative fluorescence microscopy images of tendinopathic tendons stained with DAPI (blue), an IL-6R antibody (magenta), and either a CD90 antibody (left image, green) or a CD68 antibody (right image, green). Boxplots depict the quantified co-localization of DAPI, IL-6, and either CD90 or CD68 calculated as percentage of the number of DAPI⁺IL-6R⁺ cells (biological replicates: N=7). In all boxplots, each datapoint was calculated from eight representative fluorescence microscopy images taken from the same sample. The colored datapoint matches the presented fluorescence microscopy image. The upper and lower hinges correspond to the first and third quartile (25th and 75th percentile), the middle one to the median, the whiskers extend from the hinges no further than 1.5 times the interquartile range, and the points beyond the whiskers are treated as outliers. Results of the statistical analysis are indicated as follows: *p<0.05, **p<0.01. The applied statistical test was the Mann-Whitney-Wilcoxon test.

Figure 4.. Concept behind the in vitro hybrid explant // hydrogel assembloid system.

(A) Abstract representation of the in vivo load-bearing tendon core subunits (light blue/white) surrounded by the extrinsic compartment (white) containing, i.e., extrinsic fibroblasts (light brown). (B) Sources of the in vitro model system components with the IL-6 knock-out core (KO core) in violet, the IL-6 wildtype core (WT core) in light blue, the IL-6 wildtype fibroblasts in light brown, and the ScxGFP fibroblasts in green. Core explants were clamped, and the fibroblasts embedded in a (liquid) collagen solution before crosslinking the mixture into a hydrogel around the clamped core explants in various combinations. (C) Photographic and light microscopic images of the in vitro assembloid model system. Lid of a 15 ml Falcon tube (Ø: 17 mm) used for scale.

Figure 4—figure supplement 1.. Characterization of cell populations derived from Achilles tendons and tail tendon fascicles of ScxGFP mice (morphology and surface markers).

(A) Representative light microscopy images depicting tail tendon fascicle-derived fibroblasts and Achilles tendon-derived fibroblasts cultured on standard tissue culture plastic before being passaged for the first time (P1) or the second time (P2). The Achilles tendon-derived fibroblasts were embedded in the collagen hydrogels at P2. (B) Presence of a selection of cell surface markers on cell populations isolated from tail tendon fascicles or Achilles tendons immediately after the digestion (left) and after passaging them twice (P2, right). Using flow cytometry, the following markers were analyzed after excluding doublets and dead cells: Scx, CD45, CD34, CD31, and CD136.

Figure 4—figure supplement 2.. Characterization of cell populations derived from Achilles tendons and tail tendon fascicles of ScxGFP mice (gene transcripts).

Presence of a selection of gene transcripts in cell populations isolated from tail tendon fascicles or Achilles tendons after passaging them twice. Using RT-qPCR, the following gene transcripts were measured: Spp1, Scx, Mkx, Col1a1, Dpt, Tnmd, Ctgf, Acta2, and Pdgfra. The dots indicate the relative expression values (dCt) of the transcripts normalized to that of Eif4a2, an established housekeeping gene, for a different animal each (biological replicates: N=3). The depicted gene transcripts and their typical expression levels in different cell populations were previously reported in mouse Achilles tendons analyzed with single-cell RNA-seq (De Micheli et al., 2020).

Figure 4—figure supplement 3.. Identification of cellular IL-6 and IL-6R sources in mouse tendon assembloids using flow cytometry and analyzing the supernatant.

(A) Detection of IL-6 protein level differences in the supernatants obtained from IL-6 wildtype core explant surrounded by a collagen hydrogel (WT core // cell-free, light blue), IL-6 knock-out core explants surrounded by a collagen hydrogel (KO core // cell-free, violet), and extrinsic, IL-6 wildtype fibroblasts seeded into a collagen hydrogel (fibroblasts [in hydrogel], light brown) after 7 days in culture (biological replicates: N=6). The upper and lower bounding boxes correspond to the first and third quartile (25th and 75th percentile) and the middle bar to the median. Whiskers extend from the upper/lower hinge to the largest/smallest value no further than 1.5 the interquartile range. Results of the statistical analysis are indicated as follows: **p<0.01, ^n.s.p>=0.05. The applied statistical test was the Wilcoxon rank sum. (B) On the left side: Flow cytometric analysis of digested wildtype core explants (WT core // cell-free) surrounded by a collagen hydrogel after 7 days in culture. The different colors indicate different biological replicates (N=3). On the right side: Flow cytometric analysis of extrinsic fibroblast cultured on tissue culture plastic before being passaged for the second time (P2, biological replicates: N=1). Assessed markers include the hematopoietic lineage marker CD45 and the IL-6R.

Figure 5.. Transcript changes in hydrogel-embedded fibroblasts seeded around an IL-6 knock-out (KO) core explant compared to those seeded around a wildtype (WT) core.

(A) Illustration of the assembloid combinations compared here (KO core // fibroblasts vs. WT core // fibroblasts), the assessed timepoint (d7), and the analyzed compartment (extrinsic fibroblasts only). (B) RNA-seq volcano plot of differentially expressed genes (DEGs). Genes colored in red have a log2 (fold change)>0.5, a p-value<0.05, and are considered to be significantly increased in the extrinsic compartment of KO core // fibroblast assembloids. Genes colored in blue have a log2 (fold change)<–0.5, a p-value<0.05, and are considered to be significantly increased in the extrinsic compartment of WT core // fibroblast assembloids. The log2 and p-value thresholds are represented by the dashed lines. (C) Unsupervised hierarchical clustering of the top 50 DEGs. Genes are clustered by color with positive (red) or negative (blue) row-scaled z-scores. Columns represent individual samples (biological replicates: N=6). (D) Dotplots depicting a selection of gene ontology (GO) annotations significantly enriched (adjusted p-value<0.05) by the DEGs. The selection was biased by GO biological process annotations enriched in the human dataset (Figure 1E). The color of the circles represents their adjusted p-value, the size the number of enriched genes (count), and the x-axis the number of enriched genes in ratio to the total number of genes annotated to the gene set (gene ratio).

Figure 5—figure supplement 1.. Detailed transcriptome analysis of genes up- and downregulated in hydrogel-embedded fibroblast seeded around an IL-6 knock-out (KO) core explant compared to those seeded around a wildtype (WT) core.

(A) Illustration of the assembloid combinations compared here (KO core // fibroblasts vs. WT core // fibroblasts), the assessed timepoint (d7), and the analyzed compartment (extrinsic fibroblasts only). (B) Dotplot showing significantly enriched gene sets (p-value<0.05) as determined by gene set enrichment analysis (GSEA) based on the MSigDB mouse hallmark gene sets. The +/- signs indicate the direction of the enrichment in the extrinsic fibroblasts around a KO core compared to those around a WT core. (C) Dotplot showing the top 20 gene ontology (GO) gene sets for biological processes significantly enriched by transcripts increased in fibroblasts seeded around a KO core. (D) Dotplot showing the top 20 GO gene sets for biological processes significantly enriched by transcripts decreased in fibroblasts seeded around a KO core. In all the dotplots, the color of the circles represents their p-value, their size the number of enriched genes (count), and their position on the x-axis the number of enriched genes in ratio to the total number of genes annotated to the gene set (gene ratio).

Figure 6.. Cell proliferation around and ScxGFP fibroblast recruitment to core explants.

(A) Illustrative depictions and representative fluorescence microscopy images of wildtype (WT) core explants surrounded by fibroblast populations from ScxGFP mice cultured with or without tocilizumab (10 µg/ml), and IL-6 knock-out (KO) explants cultured with or without recombinant IL-6 (25 ng/ml) for 7 days. All cells are colored in blue (NucBlue), ScxGFP fibroblasts in green (GFP), and dead cells in red (EthD). (B, C, D) Boxplots depicting the total number of cells, the number of ScxGFP cells, and the ratio between core-resident and extrinsic ScxGFP cells normalized to the WT median. Each datapoint was calculated from three representative fluorescence microscopy images taken from the same sample. The red datapoint matches the presented fluorescence microscopy image. The upper and lower hinges correspond to the first and third quartile (25th and 75th percentile), the middle one to the median, the whiskers extend from the hinges no further than 1.5 times the interquartile range, and the points beyond the whiskers are treated as outliers. (E) Lineplot depicting the cumulative percentage of ScxGFP cells depending on their distance from the center line of the core explant. The points and the line represent the mean cumulative percentages and the error bands the standard error of the mean (sem). The dashed line indicates locations inside the core area (biological replicates: N=12). Results of the statistical analysis are indicated as follows: *p<0.05, **p<0.01. The applied statistical test was the Mann-Whitney-Wilcoxon test, and lower hinges correspond to the first and third quartile (25th and 75th percentile), the middle one to the median, the whiskers extend from the hinges no further than 1.5 times the interquartile range, and the points beyond the whiskers are treated as outliers. (E) Lineplot depicting the cumulative percentage of ScxGFP cells depending on their distance from the center line of the core explant. The points and the line represent the mean cumulative percentages and the error bands the standard error of the mean (sem). The dashed line indicates locations inside the core area (biological replicates: N=12). Results of the statistical analysis are indicated as follows: *p<0.05, **p<0.01. The applied statistical test was the Mann-Whitney-Wilcoxon test.

Figure 7.. Cell proliferation around and Scx⁺ fibroblasts recruitment to acutely damaged mouse Achilles tendons 14 days after injury.

(A) Illustrative depiction of the experimental setup and the time schedule. (B) Representative fluorescence microscopy images from all four mice assessed showing longitudinal mouse hindleg sections from IL-6 wildtype, ScxGFP (IL-6 WT x ScxGFP) Achilles tendons (AT) that underwent tenotomy (left), the contralateral untreated control (middle), as well as sections from IL-6 knock-out (IL-6 KO x ScxGFP) AT that underwent tenotomy (right). In addition to the signal provided by the ScxGFP cells (green), NucBlue was used to identify all cell nuclei (blue), and 5-ethynyl-2-deoxyuridine (EdU) was used to identify proliferating cells (magenta). The dashed circles indicate the remaining AT stump close to the calcaneus. The healing neo-tendon tissue surrounding the calcaneal AT stump bridges the gap to the AT stump connected to the calf muscles further down (not shown). In both compartments, ScxGFP cell distribution was highly variable across acutely damaged samples, with no observable trends or statistically detectable differences between the conditions (biological replicates: N=4).

Figure 7—figure supplement 1.. Cell proliferation around and Scx⁺ fibroblast recruitment to damaged mouse Achilles tendons.

(A) Illustrative depiction of the experimental setup and the time schedule. (B) Representative fluorescence microscopy images of mouse hindleg cross-sections from wildtype (WT) Achilles tendons that underwent tenotomy (left), the contralateral untreated control (middle), as well as cross-sections from IL-6 knock-out (KO) Achilles tendons that underwent tenotomy (right). (C) Total number of cells stained with NucBlue. (D) Number of Scx⁺ cells. (E) Ratio between core-resident and extrinsic Scx⁺ cells depicted on a logarithmic y-axis (biological replicates: N=7). The upper and lower hinges correspond to the first and third quartile (25th and 75th percentile), the middle one to the median, the whiskers extend from the hinges no further than 1.5 times the interquartile range, and the points beyond the whiskers are treated as outliers. (F) Lineplot depicting the cumulative percentage of Scx⁺ cells depending on their distance from the Achilles tendon center. The points and the line represent the mean cumulative percentages and their error bands the standard error of the mean (sem). The dashed line indicates locations inside the Achilles tendon stump. Results of the statistical analysis are indicated as follows: *p<0.05. The applied statistical test was the Mann-Whitney-Wilcoxon test.

Figure 7—figure supplement 2.. CD146⁺ and TPPP3⁺ fibroblast recruitment to damaged mouse Achilles tendons.

(A) Representative fluorescence microscopy images of mouse hindleg sections from wildtype (WT) and IL-6 knock-out (KO) Achilles tendons that underwent unilateral tenotomy. (B) Boxplot reporting the number of CD146⁺ cells normalized to the WT median. (C) Boxplot reporting the number of TPPP3⁺ cells normalized to the WT median. In these boxplots, the upper and lower hinges correspond to the first and third quartile (25th and 75th percentile), the middle one to the median, the whiskers extend from the hinges no further than 1.5 times the interquartile range, and the points beyond the whiskers are treated as outliers. (D, E) Lineplots depicting the cumulative percentages of CD146⁺ and TPPP3⁺ cells depending on their distance from the Achilles tendon center. The points and the line represent the mean cumulative percentages and the error bands their standard error of the mean (sem). The dashed line indicates locations inside the Achilles tendon stump (biological replicates: N=4). The applied statistical test was the non-parametric Wilcoxon rank sum test and no significant differences were detected.

Figure 8.. Transcript analysis of differentially regulated genes and pathways in wildtype (WT) core explants surrounded by a hydrogel seeded with fibroblasts compared to a WT core surrounded by a cell-free hydrogel.

(A) Illustration depicting the assembloid combinations compared here (WT core // fibroblasts vs. WT core // cell-free), the assessed timepoint (d7), and the analyzed compartment (core only). (B) RNA-seq volcano plot of differentially expressed genes (DEGs). Genes colored in red have a log2 (fold change)>0.5, an adjusted p-value<0.05, and are considered to be significantly increased in the core of WT core // fibroblast assembloids. Genes colored in blue have a log2 (fold change)<–0.5, an adjusted p-value<0.05, and are considered to be significantly increased in the core of WT core // cell-free assembloids. The log2 and p-value thresholds are represented by the dashed lines. (C) Unsupervised hierarchical clustering of the top 50 DEGs. Genes are clustered by color with positive (red) or negative (blue) row-scaled z-scores. Columns represent individual samples (biological replicates: N=4). (D) Dotplots depicting a selection of gene ontology (GO) annotations significantly enriched (adjusted p-value<0.05) by the DEGs. The selection was biased by GO biological processes and gene set enrichment analysis (GSEA) hallmark annotations enriched in the human dataset (Figure 1C and E). The color of the circles represents their adjusted p-value, the size the number of enriched genes (count), and the x-axis the number of enriched genes in ratio to the total number of genes annotated to the gene set (gene ratio).

Figure 8—figure supplement 1.. Detailed transcriptome analysis of genes up- and downregulated in wildtype (WT) core explants surrounded by a hydrogel seeded with extrinsic fibroblasts compared to a WT core surrounded by a cell-free hydrogel.

(A) Illustration of the assembloid combinations compared here (WT core // fibroblasts vs. WT core // cell-free), the assessed timepoint (d7), and the analyzed compartment (core only). (B) Dotplot depicting the top 6 significantly enriched gene sets as determined by gene set enrichment analysis (GSEA) based on the MSigDB mouse cell-type signature gene sets. The color of the circles represents their p-value, the size the number of enriched genes (count), the position on the x-axis the number of enriched genes in ratio to the total number of genes annotated to the gene set (gene ratio), and the +/- signs the direction of the enrichment in the WT core surrounded by fibroblasts compared to the WT core surrounded by a cell-free hydrogel. (C) Detailed annotation of the enrichment map plot clustering the top 30 biological processes significantly enriched by DEG sets. The color of the circles represents their adjusted p-values, the size represents the number of enriched genes (count), and the gray lines connect GO annotations that share the same gene subsets.

Figure 8—figure supplement 2.. Overlap between differentially expressed transcripts in in vitro assembloids and differentially expressed transcripts between Achilles tendon fibroblasts from the extrinsic compartment and the tendon core in vivo.

(A) Illustrative depiction of the comparisons whose overlap was investigated here: The core explants of wildtype (WT) core // fibroblast vs. the WT core // cell-free assembloids (in vitro, red) and the extrinsic (paratenon-derived) vs. the core (tendon proper-derived) fibroblasts (in vivo, blue). (B) Venn diagram depicting the number and the overlap (violet) of differentially expressed genes (DEGs) as well as the top 7 gene ontology (GO) gene sets for biological processes significantly enriched by the overlapping DEGs. (C) Venn diagram depicting the number and the overlap (violet) of significantly enriched GO annotations for biological processes. (D) Dotplot depicting the top 30 biological processes significantly enriched by DEGs in tendon fibroblasts derived from the extrinsic compartment (paratenon-derived) compared to those derived from the core compartment (tendon proper-derived) colored in a blue to black gradient. The plot is augmented by the data of matched biological processes also significantly enriched in the core of WT core // fibroblast compared to that of the WT core // cell-free assembloids colored in a red to black gradient. The color gradient of the circles represents their adjusted p-values, the size the number of enriched genes (count), and the x-axis the number of enriched genes in ratio to the total number of genes annotated to the gene set (gene ratio).

Figure 9.. Transcript analysis of differentially regulated genes and pathways in IL-6 knock-out (KO) core explants surrounded by a hydrogel seeded with fibroblasts compared to a wildtype (WT) core surrounded by fibroblasts.

(A) Illustration depicting the assembloid combinations compared here (KO core // fibroblasts vs. WT core // fibroblasts), the assessed timepoint (d7), and the analyzed compartment (core only). (B) RNA-seq volcano plot of differentially expressed genes (DEGs). Genes colored in red have a log2 (fold change)>0.5, a p-value<0.05, and are considered to be significantly increased in the core of KO core // fibroblast assembloids. Genes colored in blue have a log2 (fold change)<–0.5, a p-value<0.05, and are considered to be significantly increased in the core of WT core // fibroblast assembloids. The log2 and p-value thresholds are represented by the dashed lines. (C) Unsupervised hierarchical clustering of the top 50 DEGs. Genes are clustered by color with positive (red) or negative (blue) row-scaled z-scores. Columns represent individual samples (biological replicates: N=4). (D) Dotplots depicting a selection of gene ontology (GO) annotations significantly enriched (adjusted p-value<0.05) by the DEGs in both the WT core // cell-free vs. WT core // fibroblast assembloid comparison (red to black gradient) and the KO core // fibroblast vs. WT core // fibroblast assembloid comparison (light blue to black gradient). The selection was biased by enriched GO biological process and gene set enrichment analysis (GSEA) hallmark annotations in the human dataset (Figure 1C and E). The color gradient of the circles represents their adjusted p-value, the size the number of enriched genes (count), and the x-axis the number of enriched genes in ratio to the total number of genes annotated to the gene set (gene ratio). (E) Venn diagram depicting the number and the overlap (violet) of significantly enriched GO annotations for biological processes between the WT core // cell-free vs. WT core // fibroblast assembloid comparison (red) and the KO core // fibroblast vs. WT core // fibroblast assembloid comparison (blue). (F) Linear elastic moduli of devitalized (Devital.), IL-6 knock-out (KO), and wildtype (WT) core explants surrounded by hydrogel-embedded fibroblast populations at day 21 normalized to day 0 (biological replicates: N=8). The data are displayed as barplots with mean ± standard error of the mean (sem). The applied statistical test was the Mann-Whitney-Wilcoxon test and yielded no significant differences.

Figure 9—figure supplement 1.. Detailed transcriptome analysis of genes up- and downregulated in knock-out (KO) core explants surrounded by a hydrogel seeded with fibroblasts compared to a wildtype (WT) core surrounded by a hydrogel seeded with fibroblasts.

(A) Illustration depicting the assembloid combinations compared here (KO core // fibroblasts vs. WT core // fibroblasts), the assessed timepoint (d7), and the analyzed compartment (core only). (B) Dotplot depicting the top 6 significantly enriched gene sets as determined by gene set enrichment analysis (GSEA) based on the MSigDB mouse cell-type signature gene sets. The +/- signs indicate the direction of the enrichment in the KO core surrounded by fibroblasts compared to the WT core surrounded by fibroblasts. (C) Dotplot showing the top 20 gene ontology (GO) gene sets for biological processes significantly enriched by transcripts increased in the core of KO core // fibroblast assembloids. (D) Dotplot showing the top 20 GO gene sets for biological processes significantly enriched by transcripts decreased in the core of KO core // fibroblast assembloids. (E) Dotplot showing the top 20 GO gene sets for molecular functions significantly enriched by transcripts increased in the core of KO core // fibroblast assembloids. (F) Dotplot showing the top 20 GO gene sets for molecular functions significantly enriched by transcripts decreased in the core of KO core // fibroblast assembloids. In all the dotplots, the color of the circles represents their p-value, the size the number of enriched genes (count), and the position on the x-axis the number of enriched genes in ratio to the total number of genes annotated to the gene set (gene ratio).