Transcriptome-Optimized Hydrogel Design of a Stem Cell Niche for Enhanced Tendon Regeneration

Abstract

Bioactive hydrogels have emerged as promising artificial niches for enhancing stem cell-mediated tendon repair. However, a substantial knowledge gap remains regarding the optimal combination of niche features for targeted cellular responses, which often leads to lengthy development cycles and uncontrolled healing outcomes. To address this critical gap, an innovative, data-driven materiomics strategy is developed. This approach is based on in-house RNA-seq data that integrates bioinformatics and mathematical modeling, which is a significant departure from traditional trial-and-error methods. It aims to provide both mechanistic insights and quantitative assessments and predictions of the tenogenic effects of adipose-derived stem cells induced by systematically modulated features of a tendon-mimetic hydrogel (TenoGel). The knowledge generated has enabled a rational approach for TenoGel design, addressing key considerations, such as tendon extracellular matrix concentration, uniaxial tensile loading, and in vitro pre-conditioning duration. Remarkably, our optimized TenoGel demonstrated robust tenogenesis in vitro and facilitated tendon regeneration while preventing undesired ectopic ossification in a rat tendon injury model. These findings shed light on the importance of tailoring hydrogel features for efficient tendon repair. They also highlight the tremendous potential of the innovative materiomics strategy as a powerful predictive and assessment tool in biomaterial development for regenerative medicine.

Keywords: extracellular matrix; hydrogel; materiomics; stem cell therapy; tendon regeneration.

Scheme 1

Study objectives and overview. A) Study objectives: The limited understanding of how material niche properties influence cellular responses has resulted in lengthy development cycles and unpredictable and uncontrolled healing outcomes. To address this challenge, “materiomics” has emerged as a powerful, data‐driven tool that can be utilized throughout the entire biomaterials development pipeline, which overcomes the limitations of traditional empirical methods and enables the rational engineering of biomaterials with enhanced therapeutic efficacy. B) Study overview: Our materiomics strategy is based on our in‐house RNA‐seq data and consists of two major components. First, we use bioinformatics to provide a mechanistic investigation into the cellular responses to different hydrogel features. Second, we develop mathematical modeling to quantitatively assess and predict the tenogenic effects of the hydrogel design. The knowledge generated from the materiomics assessment will ultimately lead to a highly effective, targeted approach for TenoGel design and improve its efficacy for stem cell‐mediated tendon regeneration.

Figure 1

Ex vivo characterization of GelMA/OA hydrogel. A) IPN‐structured hybrid hydrogel: the schematic illustrates the triple network formed in GelMA/OA hydrogel, consisting of Ca²⁺ cross‐linked OA, covalent cross‐links in photocured GelMA, and the imine bond formed between GelMA and OA based on Schiff base reaction. Digital images demonstrated the highly elastic property of the ADA/GelMA. B) FTIR: FTIR spectra of GelMA/OA hydrogel and its individual components, indicated the formation of imine‐bonds ≈1690–1640 cm⁻¹. C) Tensile test: representative stress‐strain curves and tensile tests showed the robust tensile attributes of the GelMA/OA hydrogel (freshly prepared and after PBS incubation at indicated time points), which were superior to those of GelMA hydrogel. n = 5; mean ± SD; *, p < 0.05; **, p < 0.01; ***, p < 0.001. D) Degradation kinetics: the degradation curve and representative SEM images after 1 day and 56 days of incubation in PBS, demonstrated the slow degradation rate and the maintenance of the homogeneous, porous structure of the GelMA/OA hydrogel. n = 5; mean ± SD. E) Cell viability test: representative LIVE/DEAD images and MTS assay of encapsulated hASCs cultured 6‐ and 10‐days showing cells have a high rate of viability and proliferation in GelMA/OA hydrogel and exhibiting elongated cell bodies. Green, calcein‐labeled live cells; red, EthD‐1‐labeled dead cells. n = 3 biological replicates; mean ± SD; **, p < 0.01.

Figure 2

Characterization of tECM and uniaxial tensile loading as two key features within TenoGel. A) Study overview: schematic of TenoGel setup and parameters of tECM concentration and loading regimens. B) tECM release kinetics: the release curve showed the controlled release of tECM from TenoGel, with ≈52% of the proteins released after 14 days of incubation in PBS. n = 5; mean ± SD. C) FEA: the representative model demonstrated the stress distribution of cell‐hydrogel constructs subjected to uniaxial tensile loading (8% strain) using a bioreactor. The simulated stress distribution within the hydrogel demonstrated a highly uniform mechanical field in the central region, with the exception of the areas near the clamps. Meanwhile, the cells located in the edge region (ROI 1) and middle region (ROI 2) experienced distinct loading effects (surface color map: principal stress intensity; red arrows: principal stress direction). To ensure consistency in evaluation, samples from the central region of the hydrogel were selected for subsequent analyses. D) LIVE/DEAD assay: representative LIVE/DEAD images showed high viability of hASCs in various TenoGel on days 6, 10, and 21. n = 3 biological replicates; Left panel: green, live cells; right panel: red, dead cells. E) F‐actin/Nuclear staining: representative images and quantitative analysis of cytoskeletal structure and nucleus shape illustrated the alignment and elongation of hASCs under uniaxial tensile loading in TenoGel (red, F‐actin; blue, nuclei). n = 3 biological replicates; mean ± SEM; *, p < 0.05; **, p < 0.01; ***, p < 0.001. F) Microscale mechanical characterization of TenoGel by nanoindentation. Indentation maps were generated from 100 points within a 200 µm × 200 µm area, demonstrating that all TenoGel constructs exhibited similar Young's modulus at day 21. These findings indicate that the incorporation of tECM and the application of uniaxial loading had no significant impact on the overall mechanical properties of TenoGel during the culture period. n = 3 biological replicates; mean ± SD; not significant (NS), p > 0.05.

Figure 3

Transcriptional analysis of hASC lineage differentiation in response to different TenoGel designs. A) Study overview: the schematic illustrates the experimental setup and TenoGel design considerations. n = 3 biological replicates. B) PCA: The PCA plot and the presentation of the top 10 genes demonstrated that the divergence in transcriptional profiles among different TenoGel groups was primarily driven by tendon‐related markers. C) DEGs: the pairwise comparison of transcriptional profiles across different TenoGel designs revealed that the largest numbers of DEGs were observed between TenoGel (tECM+Loading) and the Gel (Ctrl) group. The size of the circles represented the number of DEGs, as indicated in the legend. D) Top DEGs: the DEG heatmap displayed the top 20 upregulated and top 20 downregulated DEGs in the comparison of TenoGel (Loading)/(tECM)/(tECM+Loading) versus Gel (Ctrl) at respective time point, along with their representative marker genes. E) Lineage differentiation: the heatmap and quantitative analysis (ssGSEA) of genes associated with tendon and bone lineages suggested that TenoGel groups, with the tECM and w/o loading, exhibited enhanced tenogenesis over time. However, TenoGel (loading) also induced osteogenesis with the culturing. n = 3 biological replicates; mean ± SEM; *, p < 0.05; **, p < 0.01; ***, p < 0.001. F) GSEA (TISSUE (Tenocyte)): the enrichment plots revealed that pre‐conditioning of TenoGel (tECM+Loading) for 10 days achieved a significantly higher NES associated with tendon‐specific signature genes compared to pre‐conditioning for 6‐ and 21‐ days. Presented NES are significant (p < 0.05, FDR < 0.25). G) GSEA (cell senescence‐related pathways): Negative NES values indicate enrichment pre‐conditioning of TenoGel (tECM+Loading) for 21 days induced the activation of cell senescence pathways compared to pre‐conditioning for 10 days. Presented NES are significant (p < 0.05, FDR < 0.25).

Figure 4

Transcriptional insights of induced signaling mechanisms in response to different TenoGel designs. A) Study overview: the schematic illustrates the experimental setup and TenoGel design considerations. B) REVIGO analysis clusters all identified GO terms of biological processes into broader categories among TenoGel (tECM+Loading) compared to Gel (Ctrl) at day 10. Additionally, the heatmap illustrated the expression levels of related genes associated with selected GO terms in hASCs. C) KEGG: the significantly enriched pathways were identified among different TenoGels (Loading)/(tECM)/(tECM+Loading) compared to the Gel (Ctrl) group on days 6, 10, and 21. D) PPI: A PPI network was constructed using upregulated genes associated with PI3K/Akt and Wnt signaling pathways. Additionally, the heatmap illustrated the expression levels of related genes in each TenoGel group and Gel (Ctrl) group at day 10. E) qPCR and WB analysis of hASCs cultured in different TenoGel groups on day 10. n = 3 biological replicates; mean ± SEM; *, p < 0.05; **, p < 0.01. F) Time‐series analysis: the expression profiles of genes associated with key pathways in different TenoGel groups over time were presented. G) Potential mechanism: the schematic illustrated the potential involvement of growth factors/PI3K/Akt, integrin/PI3K/Akt, and Wnt signaling pathways in hASC tendon differentiation and related protein synthesis in response to different TenoGel designs.

Figure 5

Mathematical modeling and evaluations of the effects of TenoGel design parameters on hASC transcriptome. A) Study overview: the schematic illustrates the experimental setup and key design considerations. B) Lineage scoring for assessing hASC tenogenic and osteogenic differentiation: A total of 81 genes associated with stem cell tenogenic (teno) and osteogenic (osteo) differentiation were selected based on the literature. The Heatmap visualization of teno/osteo scores, derived from the transcriptomic data of hASCs in each TenoGel group, was presented. C) Model performance evaluation: The R² and MSE of the testing datasets were used to evaluate the performance of the RF and LR models. The evaluation was conducted using three‐fold cross‐validation based on the training and validation datasets. The scatter plots showed the predicted and observed differentiation scores from the first split of each model. mean ± SEM. D,E) LR models: 4D regression plot and multiple LR models were established to compare the effects of relative importance of TenoGel parameters on hASC tenogenic and osteogenic differentiation over time (panel D: teno score; panel E: osteo score). mean ± SD; *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Figure 6

In vitro analysis of optimized TenoGel design. A) Study overview: schematic illustrated the experimental setup and in vitro validation. n = 3 biological replicates. B) qPCR: the expression levels of established tendon marker genes in hASCs cultured under different TenoGel groups on day 10 are presented. mean ± SEM; *, p < 0.05; **, p < 0.01. IF staining: representative confocal images are shown, depicting the staining intensity of TNC and TNMD in hASCs cultured in different TenoGel groups on days 6, 10, and 21 (yellow, TNC; green, TNMD; blue, nuclei). These results demonstrated that the presence of tECM supplementation and stimulation of uniaxial tensile loading in TenoGel synergistically promoted hASC tenogenic differentiation. C) ARS staining and semi‐quantitative analysis: Representative histologic sections of hASCs cultured in various TenoGel groups with osteogenic medium on day 21. The red color indicates positively stained areas of calcium nodes, reflecting mineral accumulation. These findings suggest that loading alone can lead to enhanced hASC mineralization, while tECM did not show strong mineralization effects, whether used alone or in combination with loading. mean ± SEM; *, p < 0.05. D) SAβ‐Gal staining and semi‐quantitative analysis: Representative microscopic images of hASCs in TenoGel (tECM+Loading) after 6, 10, and 21 days of pre‐conditioning. Cells stained with blue indicate positive staining for SAβ‐Gal. These findings indicate that longer in vitro pre‐conditioning duration induces more pronounced cell senescence. mean ± SEM; *, p < 0.05. ***, p < 0.001.

Figure 7

Characterization of optimized TenoGel with rASCs for tendon repair in a rat tendon defect model. A) Study overview: schematic showing the experimental setup. The experimental groups included: 1) “Defect only”: the tendon defects were created without hydrogel implantation; 2) “Gel only”: cell‐free GelMA/OA hydrogels were implanted onto the tendon defects; 3) “rASCs‐Gel (Ctrl)”: rASCs seeded in GelMA/OA hydrogel in static culture for 10 days and implanted onto the tendon defects; and 4) “rASCs‐TenoGel (tECM+Loading)”: rASCs seeded in tECM containing‐GelMA/OA hydrogel in uniaxial tensile loading culture for 10 days and implanted onto the tendon defects. The uninjured tendons on the contralateral side were designated as the intact control (intact ctrl) for comparative analysis. B) Cell tracing: biodistribution and semi‐quantitative analysis of DiR‐labeled rASCs at designed time points after transplantation revealed that both the Gel (Ctrl) and TenoGel (tECM+Loading) supported the long‐term retention and viability of rASCs post‐implantation. n = 4 rats per group. C) Macroscopic observation: digital images and measurements of the length and CSA for repaired tendons at 8 weeks post‐surgery are presented (blue arrow, defect site). n⩾10 rats per group; mean ± SEM; *, p < 0.05; **, p < 0.01; ***, p < 0.001. D) Biomechanical assessment: Representative load‐displacement curves and mechanical parameters of tendons from different groups at 8 weeks post‐surgery are presented. n⩾7 rats per group; min. to max.; *, p < 0.05; ***, p < 0.001. These results suggested that TenoGel (tECM+Loading) promoted rASCs‐mediated tendon repair, as evidenced by improvements in macroscopic appearance and enhanced mechanical properties in a rat tendon defect model.

Figure 8

Histological evaluation of preconditioned, rASCs‐TenoGel (tECM+Loading) for tendon repair in rat patellar tendon defect model. A,B) Histological assessment: representative histological images (H&E, Picrosirius Red (PSR), Alcian blue, Safranin O and Von Kossa staining) and semi‐quantification analyses of tendons from different groups at week 4 and week 8 are presented. The rASCs‐TenoGel (tECM+Loading) group exhibited improved cellular alignment compared to other injured groups at both 4 and 8 weeks, as observed in H&E staining. Additionally, polarized PSR indicated that the rASCs‐TenoGel group exhibited more pronounced birefringence than the other injured groups, with a higher amount of orange‐to‐red (thicker) fibers at both 4 and 8 weeks. Furthermore, Alcian blue, Safranin O, and Von Kossa staining revealed the rASCs‐TenoGel (tECM+Loading) group exhibited less sGAG and calcium deposition compared to the other injured groups at 8 weeks, similar to the intact tendon tissues. n⩾3 rats per group; mean ± SEM; *, p < 0.05. Overall, the histological analyses revealed that preconditioned rASCs‐TenoGel (tECM+Loading) facilitated robust tendon healing at 8 weeks post‐surgery compared to the other injured groups.