Pesic, M. | bioRxiv Science

bioRxiv2023

Stabilisation of β-Catenin-WNT signalling by USP10 in APC-truncated colorectal cancer drives cancer stemness and enables super-competitor signalling

The contribution of deubiquitylating enzymes to {beta}-Catenin stabilisation in intestinal stem cells and colorectal cancer (CRC) is poorly understood. Here, we report the deubiquitylase USP10 as an APC-truncation- specific enhancer of {beta}-Catenin stability, potentiating WNT signalling in CRC and cancer stem cells. Mechanistically, interaction studies in various CRC cell lines and in vitro binding studies, together with computational modelling, revealed that USP10 binding to {beta}-Catenin is mediated via the unstructured N-terminus of USP10 and requires the absence of full-length APC. Notably, loss of USP10 in CRISPR engineered intestinal organoids reduces tumorigenic properties of CRC and blocks the super competitor-signalling of APC-mutated CRC. Furthermore, reduction of USP10 induces the expression of differentiation genes, and opposes the APC-truncated phenotype in an intestinal hyperplasia model of D.melanogaster. Taken together, our findings reveal USP10s role in intestinal tumourigenesis by stabilising {beta}-Catenin, leading to aberrant WNT signalling, enhancing cancer cell stemness and implicate the DUB USP10 as a cancer specific therapeutic vulnerability in Apc truncated CRC.

cancer biology↗

bioRxiv2022

On optimal temozolomide scheduling for slowly growing gliomas

BackgroundTemozolomide (TMZ) is an oral alkylating agent active against gliomas with a favorable toxicity profile. It is part of the standard of care in the management of glioblastoma, and is commonly used in low-grade gliomas. In-silico mathematical models can potentially be used to personalize treatments and to accelerate the discovery of optimal drug delivery schemes. MethodsAgent-based mathematical models fed with either mouse or patient data were developed for the in-silico studies. The experimental test beds used to confirm the results were: mouse glioma models obtained by retroviral expression of EGFR wt or EGFR vIII in primary progenitors from p16/p19 ko mice and grown in vitro and in vivo in orthotopic allografts, and human glioblastoma U251 cells immobilized in alginate microfibers. The patient data used to parametrize the model were obtained from the TCGA/TCIA databases and the TOG clinical study. ResultsSlow growth virtual murine gliomas benefited from increasing TMZ dose separation in silico. In line with the simulation results, improved survival, reduced toxicity, lower expression of resistance factors and reduction of the tumor mesenchymal component were observed in experimental models subject to long-cycle treatment, particularly in slowly-growing tumors. Tissue analysis after long-cycle TMZ treatments revealed epigenetically-driven changes in tumor phenotype, which could explain the reduction in glioma growth speed. In-silico trials provided support for methods of implementation in human patients. ConclusionsIn-silico simulations, and in-vitro and in-vivo studies show that TMZ administration schedules with increased time between doses may reduce toxicity, delay the appearance of resistances and lead to survival benefits mediated by changes in the tumor phenotype in gliomas. IMPORTANCE OF THE STUDYIn-vivo evidence is provided of improvements in survival, resistance, and toxicity from TMZ schemes with long rest periods between doses in slowly-growing GBM mouse models. The results match hypotheses generated in silico using a mathematical model incorporating the main biological features and fed with real patient data. An epigenetically-driven change in tumor phenotype was also revealed experimentally, which could explain the reduction in glioma growth speed under the long cycle scheme. To determine the extent to which our results hold for human patients, large sets of simulations were performed on virtual patients. These in-silico trials suggest different ways to bring the benefits observed in experimental models into clinical practice.

cancer biology↗