Preserving Chondrocyte Phenotype: Impact of in vitro Expansion on Chromatin Remodeling, Cytoskeletal Changes, and Mechanosensitivity

Introduction: Repairing articular cartilage poses a significant challenge due to the tissue's limited intrinsic healing capacity. Autologous chondrocyte implantation (ACI) holds great promise as a surgical intervention for treating cartilage defects [1]. However, a significant challenge in ACI is the dedifferentiation of chondrocytes during in vitro expansion, leading to the loss of their phenotype. This process involves the downregulation of canonical markers defining the cartilage phenotype, limiting the therapeutic efficacy of ACI [2]. Hence, understanding the mechanisms underlying phenotype loss during in vitro expansion is crucial. Furthermore, leveraging this knowledge is essential for preserving the original phenotype of chondrocytes during expansion, thereby augmenting the therapeutic potential of ACI for cartilage repair. To address this, our study aims to explore the impact of in vitro monolayer expansion on the nano-scale chromatin organization, epigenetic and transcriptomic profiles, and cellular mechanosensitivity of human chondrocytes. These findings provide valuable insights for optimizing cell-based cartilage regeneration therapies.

Materials and

Methods: Human articular chondrocytes from three donors (Donor A: 46-year-old male; Donor B: 55-year-old male; Donor C: 52-year-old male, provided by LifeLink Foundation) were isolated and expanded on tissue culture plastic up to passage six (P0 to P6). Nano-scale chromatin organization was assessed using super-resolution STORM imaging (ONI) of histone-H2B (H2B) and analyzed using the Nanoimager software (ONI) [3]. The chromatin condensation parameter was determined by normalizing the Voronoi density of H2B to the mean Voronoi polygon density in a nucleus [3, 4]. To investigate alterations in histone modification status during in vitro expansion, immunoblotting was performed for acetylation of H3K9 (H3K9ac, epigenetic activation), tri-methylation of H3K4 (H3K4me3, epigenetic activation), tri-methylation of H3K27 (H3K27me3, epigenetic suppression), tri-methylation of H3K9 (H3K9me3, epigenetic suppression), along with histone H3 (H3). Bulk RNA sequencing was conducted and analyzed using DESeq2 and clusterProfiler R packages. To explore changes in mechanosensitivity in chondrocytes during expansion, P0 and P6 cells were seeded onto glass surfaces or polyacrylamide hydrogels with various stiffnesses and cultured in basal media with or without Y-27632 (Y27), a ROCK inhibitor. Immunofluorescence data of YAP nuclear to cytoplasmic (N/C) ratio and cell morphology (Phalloidin) were obtained using a widefield microscope (Leica). Statistical analyses were performed using Student’s t-test or ANOVA with Tukey’s post hoc testing.

Results, Conclusions, and Discussions: The Likelihood Ratio Test (LRT) from RNA-Seq data revealed significant gene expression changes during chondrocyte dedifferentiation from passage 0 (P0) to passage 6 (P6) (Fig. 1A). Cluster 1 includes 548 continuously downregulated genes, associated with ECM remodeling and cartilage development (Fig. 1B), while Cluster 2 includes 400 continuously upregulated genes, related to cytoskeleton organization and dynamics (Fig. 1C). This suggests a shift away from the specialized chondrogenic phenotype. STORM analysis showed a decreased H2B density in chondrocyte nuclei (Fig. 2A, B), indicating that chromatin becomes less condensed during in vitro expansion. This chromatin remodeling coincided with significant decreases in both activation and suppression histone modification markers (Fig. 2C, D), suggesting intricate epigenetic reprogramming during dedifferentiation, likely contributing to the loss of the chondrogenic phenotype. When chondrocytes were seeded onto soft (5kPa) hydrogels, they exhibited a smaller cell area and higher circularity compared to those on stiff (30kPa) substrates (Fig. 3A, B). Dedifferentiated P6 chondrocytes displayed a larger cell area and more elongated morphology than primary P0 chondrocytes (Fig. 3A, B). Additionally, the assessment of YAP nuclear localization levels revealed significantly greater nuclear YAP localization in P6 cells on both soft and stiff substrates compared to P0 cells (Fig. 3B), indicating altered mechanosensitivity with in vitro expansion. Treatment with Y27 resulted in a significant decrease in cell area and an increase in circularity of P6 cells (Fig. 3C) to levels similar to those observed in P0 cells, suggesting that modulating cellular contractility could be a potential strategy for preserving the chondrocyte phenotype. These findings demonstrate that chondrocyte dedifferentiation during in vitro expansion involves significant chromatin remodeling, epigenetic reprogramming, and cytoskeleton reorganization. Changes in cellular morphology and YAP localization indicate altered mechanosensitivity in dedifferentiated cells. Modulating cellular contractility with ROCK inhibitors, alongside epigenetic modifiers, may offer a strategy to maintain chondrocyte phenotype, enhancing the efficacy of ACI for cartilage repair. Future studies will focus on the roles of specific histone modifications in regulating chondrogenic gene expression and the mechanisms underlying altered mechanosensitivity to improve therapeutic outcomes in cartilage regeneration.

Acknowledgements (Optional): This work is supported by NIH R01 AR079224 and NSF CMMI 1548571.