Ferroptosis is regulated by matrix mechanics

Introduction: : In solid tumors, the extracellular matrix (ECM), the network of proteins surrounding cells, is higher in stiffness and biochemically distinct in its composition compared to normal ECM. This is the result of aberrant ECM dynamics, which include altered expression of matrix remodeling enzymes and a disrupted balance between ECM synthesis and degradation. Abnormal ECM in cancer affects oncogenic signaling and gene expression, promotes cancer progression and metastasis, and contributes to therapeutic resistance. Ultimately, these alterations in the tumor microenvironment lend to more aggressive cancer phenotypes and poorer patient prognosis. Thus, there is a significant unmet need to address therapy-resistant stiff tumors by taking advantage of their mechanical properties. Interestingly, ferroptosis, a recently recognized form of cell death via lipid peroxidation, has emerged as a promising therapeutic option for several cancers. YAP/TAZ, the transcriptional co-activators of the Hippo pathway, has been shown to activate in response to ECM stiffness and separately shown to increase susceptibility to ferroptosis. Here, we report that ferroptosis is regulated by ECM stiffness through the YAP/TAZ expression program and can be used to treat stiff chemoresistant tumors.

Materials and

Methods: : To model ECM stiffness, we use 2D polyacrylamide (PA) hydrogels. Soft (0.5 kPa) and stiff (25 kPa) PA gels are generated by combining 40% acrylamide solution, 2% bis-acrylamide solution, Acrylic-NHS, ammonium persulfate, and TEMED followed by coating with rat-tail collagen I. We have chosen 3 cell lines whose sensitivity to ferroptosis has been shown previously, HT1080 fibrosarcoma cells, MDA-MB-231 breast cancer cells, and CAOV3 ovarian cancer cells. After cells have been seeded on soft or stiff hydrogels for 24 hrs, we treat with one of two ferroptosis-inducing drugs, erastin or FIN56. Following 6-16 hr treatments, we measure cell viability by CellTiterGlo, assess lipid peroxidation via flow cytometry, and perform a rescue assay via co-treatment with ferrostatin-1. To confirm matrix-mediated ferroptosis sensitivity is regulated by YAP/TAZ, we perform inhibition by siRNA, shRNA, or drug inhibition by verteporfin. Following co-treatment with erastin, we analyze by CellTiterGlo and assess lipid peroxidation via flow cytometry. To further confirm the mechanism, we have performed bulk RNA-seq analysis to identify YAP/TAZ target genes. To establish stiff tumors in vivo we combine MDA-MB-231 and an injectable hydrogel solution in the rear flank of female NOD/SCID mice. After tumors reach a size of 50-150 mm^3, we treat for 2 weeks with one of five conditions: (1) PBS, (2) JKE-1674 (ferroptosis-inducer), (3) carboplatin (chemotherapy), (4) JKE-1674 and carboplatin, and (5) JKE-1674 and liproxstatin-1 (ferroptosis inhibitor). We remove the tumors for analysis after 2 weeks including tumor weight, histology, and atomic force microscopy.

Results, Conclusions, and Discussions:: For all three cell lines, we have found that culture on a stiff-matrix leads to decreased cell viability after treatment with erastin compared to those cultured on a soft matrix (Figure 1A). Furthermore, cells cultured on a stiff matrix show increased lipid peroxidation (Figure 1B). To confirm this is ferroptosis specific, a rescue assay was performed where co-treatment with ferrostatin-1 prevents ferroptotic cell death in cells cultured on stiff hydrogels (Figure 1C). Inhibition of YAP/TAZ by siYAP1, shYAP1, or the drug verteporfin increased cell viability and decreased lipid peroxidation (Figure 1D-F). RNA-seq shows differential expression of several genes between erastin treated cells cultured on a soft or stiff matrix (Figure 1G). Furthermore, we have identified several gene hits which may be the YAP/TAZ target genes responsible for our observed phenotype confirmed by RT-PCR (Figure 1H). Lastly, in vivo, we confirmed a stiff tumor phenotype by increased stiffness measured by AFM and increased ECM deposition by Masson’s trichrome staining (Figure 1I-J). Crucially, we observed decreased tumor volume and tumor weight in JKE-1674 treated tumors compared to PBS, carboplatin, or JKE-1674 and liproxstatin-1 (Figure 1K-L). In conclusion, we have identified the mechanical regulation of ferroptosis by ECM stiffness, modulated by the YAP/TAZ expression program. Furthermore, we have begun identifying the YAP/TAZ target genes with future experiments focused on siRNA inhibition studies. Lastly, we have shown that ferroptosis can be used to target stiff chemoresistant tumors. Overall, our results have begun to elucidate the mechanical regulation of ferroptosis in cancer cells. This knowledge has the potential to address an unmet need to target therapeutically resistant stiff tumors and ultimately improve patient prognosis.

Acknowledgements and/or References (Optional)::