Scientists Switch Off a Key Mechanosensor in Tendon Cells and Map What Breaks

Tendons are among the most mechanically abused tissues in the human body. Every step, grip, and sprint stretches the tenocytes that live inside them, and those cells must translate physical force into biochemical instructions that keep the tissue strong. A new study published in Frontiers in Bioengineering and Biotechnology has now taken a direct swing at one of the central questions in tendon biology: what happens to the genetic program of a tenocyte when its primary mechanical sensor, the protein YAP, is deleted while the cell is being stretched? The answer, drawn from a conditional knockout mouse model combined with RNA sequencing, reveals a surprisingly broad transcriptional upheaval that touches lipid metabolism, immune signaling, and a set of candidate drug targets that could one day inform how tendon injuries are treated.

The research team, led by investigators at the Hospital for Special Surgery in New York in collaboration with colleagues at Tianjin First Central Hospital and the Performance Medicine Institute, focused on Yes-associated protein, or YAP, a transcriptional regulator that sits at the heart of the Hippo signaling pathway. YAP is famous in mechanobiology because its location within the cell is dictated by physical forces: when a cell is stretched, YAP moves into the nucleus and switches on genes that help the cell adapt to its mechanical environment. Although this behavior has been documented in many cell types, no previous study had systematically deleted YAP in mature tenocytes and then interrogated the resulting transcriptomic landscape under controlled mechanical loading. That gap mattered, because tenocytes in a living tendon are never at rest; they are continuously exposed to tensile strain, and any gene whose regulation by YAP becomes critical under those conditions is a gene worth knowing about.

To answer the question, the researchers built a genetically engineered mouse strain carrying a tamoxifen-activated Cre recombinase at the Rosa26 locus crossed with mice whose YAP gene was flanked by loxP sites. Tenocytes were harvested from the tail tendons of three-month-old male mice and grown in culture. Treating the cells with 4-hydroxytamoxifen excised the critical exons of YAP, and Western blotting confirmed that the protein had been effectively eliminated. The cells were then embedded in collagen gels in Flexcell Tissue Train plates and subjected to a cyclic tensile loading regimen of 4 percent strain at 1 hertz for 24 hours, a protocol designed to mimic the physiological stretching that tenocytes experience in a working tendon. After loading, RNA was extracted and sequenced on an Illumina NovaSeq X Plus platform, with reads aligned to the mouse genome and analyzed using the edgeR package to identify differentially expressed genes.

The headline result was stark. Comparing YAP-deficient tenocytes with controls under identical mechanical loading, the team identified 54 differentially expressed genes, of which only six were upregulated and 48 were downregulated. The overwhelming downward shift suggests that YAP acts predominantly as a transcriptional activator in mechanically stimulated tenocytes, and that its removal silences a large slice of the load-responsive program. Among the most strongly upregulated genes were Gzmc and Gm28151, while the most strongly downregulated included Car3, which encodes carbonic anhydrase 3, and Rgs2, a regulator of G-protein signaling. Both Car3 and Rgs2 have been linked to YAP signaling in other biological systems, which the authors note lends partial external consistency to their findings.

When the researchers ran the 54 genes through Gene Ontology and KEGG pathway enrichment analyses, a clear thematic signature emerged: lipid metabolism. The genes were significantly enriched in biological processes such as fat cell differentiation and fatty acid metabolism, in cellular components including lipid droplets and lipoprotein particles, and in molecular functions such as heparin and glycosaminoglycan binding. At the pathway level, the PPAR signaling pathway stood out, alongside regulation of lipolysis in adipocytes and the AMPK signaling pathway. Gene set enrichment analysis reinforced the metabolic theme, showing that YAP knockout cells were significantly enriched for hallmark gene sets covering adipogenesis, cholesterol homeostasis, and epithelial-mesenchymal transition, while control cells showed enrichment of E2F target genes associated with proliferation. The authors are careful to frame these observations as correlative: the experiments were performed in an in vitro culture system without injury or degeneration, so the findings generate hypotheses rather than establish mechanisms of tendon repair.

The lipid connection is biologically intriguing rather than incidental. Lipids serve as an energy source for tenocytes and help organize cell membranes, influencing the fluidity and lipid raft dynamics that modulate mechanosensitive channels and integrin-mediated force transmission. Lipid-derived mediators also participate in extracellular matrix remodeling by regulating collagen synthesis and the activity of matrix-degrading enzymes. PPARγ, the central regulator of adipogenic differentiation, has been shown to interact functionally with YAP in other cell types, and the new data raise the possibility that this crosstalk operates in tendon cells as well. Whether the metabolic transcriptional program observed here is a cause or a consequence of YAP deficiency, and whether it helps maintain the tenocyte phenotype under load, remains an open question that the authors say will require lipidomic analyses and targeted perturbation of PPAR signaling in future work.

To distill the 54-gene list down to the most consequential players, the team deployed two complementary machine learning approaches. LASSO regression identified seven candidate genes, and support vector machine recursive feature elimination identified 39; their intersection yielded seven hub genes: Car3, Cxcl13, Hvcn1, Ltc4s, Plin4, Rasd1, and Rgs2. Expression analysis confirmed significant differences between knockout and control groups for all seven, and correlation analysis revealed strong positive relationships among them, hinting that they operate as a coordinated module. The authors emphasize that these selections were derived from a limited number of biological replicates, three per group, which is the standard minimum for bulk RNA sequencing, and that validation in larger cohorts is essential before any general conclusions can be drawn.

One of the more provocative analyses involved CIBERSORTx, a computational method that estimates immune cell proportions from bulk gene expression data. Applied here as an exploratory check rather than a quantitative measurement, the analysis suggested significant differences between knockout and control samples in the signatures of regulatory T cells and CD4-positive naive T cells, and all seven hub genes showed positive correlations with the Treg signature. The authors are explicit that cultured tenocytes contain no immune cells and that these scores are indirect inferences about the immune landscape that might arise from tenocyte-immune crosstalk in a living tendon. Even so, the result dovetails with a growing appreciation that tendon healing depends on precisely orchestrated inflammatory responses, and it suggests that YAP-dependent paracrine signaling from tenocytes could help shape the recruitment and polarization of immune cells during repair.

The study also ventured into translational territory. Screening the Drug-Gene Interaction Database revealed that four of the hub genes, LTC4S, CA3, CXCL13, and RGS2, are associated with known drugs or small molecules, and twelve compounds emerged from the network analysis. Molecular docking simulations ranked candidate ligands against the LTC4S protein, with Compound 39 showing the strongest predicted binding energy at minus 7.78 kilocalories per mole, followed by TK05 at minus 6.24. The researchers stress that these are purely computational predictions; docking scores indicate geometric and energetic compatibility, not functional binding or therapeutic efficacy, and experimental validation would be required before any clinical relevance could be considered.

The authors are refreshingly candid about the limitations of their work. There is no in vivo validation, the differentially expressed genes have not yet been confirmed by qRT-PCR or Western blotting, the sample size is small, and the cultured tenocytes lack the three-dimensional extracellular matrix architecture, vasculature, and immune environment of native tendon tissue. The immune findings in particular remain indirect predictions rather than measurements. Still, the study delivers something the field has lacked: a systematic, mechanically contextualized map of what YAP does in mature tenocytes at the transcriptomic level. The data, deposited in the Gene Expression Omnibus under accession GSE346125, provide a hypothesis-generating resource for tenocyte mechanobiology. Future experiments using conditional YAP knockout mice in tendon injury models, combined with functional validation of genes such as Car3 and Rgs2 and targeted manipulation of PPAR and lipid signaling, will determine whether the molecular circuitry uncovered here matters for real tendon healing, and whether the immune and pharmacological angles it opens can be turned into therapies for the millions of people sidelined by tendon injuries each year.

Subject of Research: YAP-dependent mechanotransduction and transcriptional responses in tenocytes under cyclic tensile loading

Article Title: YAP-dependent transcriptional responses in tenocytes under cyclic tensile loading: a conditional knockout RNA-sequencing study

Article References: Xing, X., Piacentini, A., Suzuki, Y., Sanchez, L. A., Mendias, C., & Rodeo, S. A. (2026). YAP-dependent transcriptional responses in tenocytes under cyclic tensile loading: a conditional knockout RNA-sequencing study. Frontiers in Bioengineering and Biotechnology, 14, Article 1901935. https://doi.org/10.3389/fbioe.2026.1901935

Image Credits: AI Generated

DOI: 10.3389/fbioe.2026.1901935

Keywords: YAP, tenocytes, mechanotransduction, tendon injury, conditional knockout, RNA sequencing, PPAR signaling, lipid metabolism, CIBERSORTx, hub genes, molecular docking, tendon repair