Biology subjects

Radvanyi, F.

Publications and source records attributed to Radvanyi, F..

11 recordsLinked to original sources

bioRxiv2025

STAG2 cohesin cooperates with DREAM to maintain quiescence and suppress tumourigenesis in the urothelium

The maintenance of quiescence is essential for tissue homeostasis. STAG2 is one of the few genes mutated in the normal urothelium of organ donors, with mutant cells undergoing positive selection 1. STAG2 is also a major tumour suppressor gene 2-4 and its inactivation is an early event in bladder carcinogenesis 1,3. However, how STAG2, a cohesin component, regulates urothelial homeostasis remains largely unknown. Here, we demonstrate that Stag2 inactivation in normal murine urothelial cells interferes with differentiation programs, triggers transient cell cycle entry, and primes cells for clonal expansion under stress. Moreover, STAG2 loss enhances tumor formation in urothelial cells expressing mutant FGFR3 - the key oncogene in bladder cancer 5. We reveal that STAG2 cooperates with the DREAM transcriptional complex, a master regulator of quiescence 6,7, by binding to shared genomic sites, including cell cycle control genes. STAG2 loss alters DREAM target expression, complex composition, and chromatin distribution, and leads to rewiring of chromatin interactions involving DREAM binding motifs in genes critical for cell cycle entry. Our findings provide compelling evidence that STAG2 loss disrupts in 3D genome organization through a novel mechanism involving the DREAM complex, thereby impairing homeostatic quiescence and increasing oncogenic sensitivity.

cancer biology↗

bioRxiv2024

A divide-and-conquer approach based on deep learning for long RNA secondary structure prediction: focus on pseudoknots identification

The accurate prediction of RNA secondary structure, and pseudoknots in particular, is of great importance in understanding the functions of RNAs since they give insights into their folding in three-dimensional space. However, existing approaches often face computational challenges or lack precision when dealing with long RNA sequences and/or pseudoknots. To address this, we propose a divide-and-conquer method based on deep learning, called DivideFold, for predicting the secondary structures including pseudoknots of long RNAs. Our approach is able to scale to long RNAs by recursively partitioning sequences into smaller fragments until they can be managed by an existing model able to predict RNA secondary structure including pseudoknots. We show that our approach exhibits superior performance compared to state-of-the-art methods for pseudoknots prediction and secondary structure prediction including pseudoknots for long RNAs. The source code of DivideFold, along with all the datasets used in this study, is accessible at https://evryrna.ibisc.univ-evry.fr/evryrna/dividefold/home.

bioinformatics↗

bioRxiv2023

Comparison and benchmark of deep learning methods for non-coding RNA classification

The involvement of non-coding RNAs in biological processes and diseases has made the exploration of their functions crucial. Most non-coding RNAs have yet to be studied, creating the need for methods that can rapidly classify large sets of non-coding RNAs into functional groups, or classes. In recent years, the success of deep learning in various domains led to its application to non-coding RNA classification. Multiple novel architectures have been developed, but these advancements are not covered by current literature reviews. We present an exhaustive comparison of the different methods proposed in the state-of-the-art and describe their associated datasets. Moreover, the literature lacks objective benchmarks. We perform experiments to fairly evaluate the performance of various tools for non-coding RNA classification on popular datasets. The robustness of methods to non-functional sequences and sequence boundary noise is explored. We also measure computation time and CO2 emissions. With regard to these results, we assess the relevance of the different architectural choices and provide recommendations to consider in future methods. Datasets and reproducible codes are available at https://evryrna.ibisc.univ-evry.fr/evryrna/ncBench. Author summaryRNA can either encode proteins, which perform different functions in the genome, or be non-coding. Non-coding RNAs represent around 98% of the genome, and were long thought to be non-functional. It has now been proven that non-coding RNAs can have diverse biological functions and be involved in diseases. A large proportion of non-coding RNAs has not yet been studied. The function of specific non-coding RNAs can be studied experimentally, but experiments are costly and time-consuming. One possibility to massively characterize the function of non-coding RNAs is to use computational methods to classify them into functional groups, or classes. Recent computational methods for non-coding RNA classification are all based on deep learning, as it leads to faster runtime and improved performance. Our work presents and compares the different approaches adopted in the state-of-the-art, as well as the non-coding RNA datasets that are used. We also present a comprehensive benchmark, measuring classification performance in different conditions, computation time, and CO2 emissions. The descriptions and comparisons provided are meant to guide researchers in the field, whether wanting to use existing tools or to develop new ones.

bioinformatics↗

bioRxiv2022

Integrated molecular and pharmacological characterization of patient-derived xenografts from bladder and ureteral cancers identifies new potential therapies.

BackgroundMuscle-invasive bladder cancer (MIBC) and upper urinary tract urothelial carcinoma (UTUC) are molecularly heterogeneous. Despite chemotherapies, immunotherapies or anti-FGFR treatments, these tumors are still of poor outcome. Our objective was to develop a bank of patient-derived xenografts (PDXs) recapitulating molecular heterogeneity of MIBC and UTUC, to facilitate preclinical identification of therapies. MethodsFresh tumors were obtained from patients and subcutaneously engrafted into immune-compromised mice. Patient tumors and matched PDXs were compared regarding histopathology, transcriptomic (microarrays) and genomic profiles (targeted-NGS). Several PDXs were treated with chemotherapy (cisplatin/gemcitabine) or targeted therapies (FGFR and EGFR inhibitors). Results31 PDXs were established from 1 non-MIBC, 25 MIBC, 5 upper urinary tract tumors, including 28 urothelial (UCC) and 3 squamous-cell carcinomas (SCC). Integrated genomic and transcriptomic profiling identified PDXs of 3 different consensus molecular subtypes (Basal/Squamous, Luminal papillary and Luminal unstable), and included FGFR3-mutated PDXs. High histological and genomic concordance was found between matched patient tumor/PDX. Discordance in molecular subtypes, such as a basal/squamous patient tumor giving rise to a luminal papillary PDX, was observed (n=5) at molecular and histological levels. Ten models were treated with cisplatin-based chemotherapy and we did not observe association between subtypes and response. Of the 3 basal/squamous models treated with anti-EGFR therapy, two models were sensitive and one model, of sarcomatoid variant, was resistant. Treatment of 3 FGFR3-mutant PDXs with combined FGFR/EGFR inhibitors was more efficient than anti-FGFR3 treatment alone. ConclusionsWe developed preclinical PDX models that recapitulate the molecular heterogeneity of MIBCs and UTUC, including actionable mutations, which will represent an essential tool in therapy development. Pharmacological characterization of the PDXs suggested that upper urinary tract and MIBCs, UCC but also SCC, with similar molecular characteristics could benefit from the same treatments including anti-FGFR for FGFR3-mutated tumors and anti-EGFR for basal ones and showed a benefit for combined FGFR/EGFR inhibition in FGFR3-mutant PDXs, compared to FGFR inhibition alone.

cancer biology↗

bioRxiv2022

Epigenomic mapping identifies a super-enhancer repertoire that regulates cell identity in bladder cancers through distinct transcription factor networks

Muscle-invasive bladder cancer (BLCA) is an aggressive disease. Consensus BLCA transcriptomic subtypes have been proposed, with two major Luminal and Basal subgroups, presenting distinct molecular and clinical characteristics. However, how these distinct subtypes are regulated remains unclear. We hypothesized that epigenetic activation of distinct super-enhancers could drive the transcriptional programs of BLCA subtypes. Through integrated RNA-sequencing and epigenomic profiling of histone marks in primary tumours, cancer cell lines, and normal human urothelia, we established the first integrated epigenetic map of BLCA and demonstrated the link between subtype and epigenetic control. We identified the repertoire of activated super-enhancers and highlighted Basal, Luminal and Normal-associated SEs. We revealed the super-enhancer-regulated networks of candidate master transcription factors for Luminal and Basal subgroups including FOXA1 and ZBED2 respectively. FOXA1 CRISPR-Cas9 mutation triggered a shift from Luminal to Basal phenotype, confirming its role in Luminal identity regulation and induced ZBED2 overexpression. In parallel, we showed that both FOXA1 and ZBED2 play concordant roles in preventing inflammatory response in cancer cells through STAT2 inhibition. Our study furthers the understanding of epigenetic regulation of muscle-invasive BLCA and identifies a co-regulated network of super-enhancers and associated transcription factors providing potential targets for the treatment of this aggressive disease.

cancer biology↗

bioRxiv2021

FGFR3 activating mutations induce luminal-like papillary bladder tumor formation and favor a male gender bias.

BackgroundFGFR3 mutations are among the most frequent genetic alterations in bladder cancer and are enriched in the luminal papillary subtype of muscle-invasive tumors (MIBC) and luminal-like classes 1 and 3 of non-MIBC. To study their oncogenic properties in vivo, we developed here a genetically engineered mouse (GEM) model expressing the most frequent FGFR3 mutation, FGFR3-S249C, in urothelial cells. MethodsBladder tumorigenesis was monitored in FGFR3-S249C mice. FGFR3 expression was assessed by RT-qPCR in the transgenic mice urothelium and in various human epithelia. Transcriptomic data were obtained from mouse bladder tumors and crossspecies comparisons were performed. Sex bias in FGFR3-mutated tumors was evaluated in our GEM model and in the TCGA and UROMOL cohorts of patients including 408 MIBC and 419 NMIBC, respectively. The association of androgen receptor (AR) activity, based on the expression of its target genes, with FGFR3 mutations was examined in these two cohorts. Binding of AR to its response element and AR phosphorylation in FGFR3-dependent cell lines were evaluated. ResultsFGFR3-S249C expression in the urothelium of mice induced spontaneous low-grade papillary bladder tumors resembling the human counterpart at the histological and transcriptomic levels. Mutant-FGFR3 expression levels impacted tumor formation incidence in mice and mutant-FGFR3-driven human tumors were restricted to epithelia presenting high normal expression levels of FGFR3. The known bladder cancer male gender bias, also found in our model, was even higher in human FGFR3-mutated compared to wild-type tumors and associated with a higher AR regulon activity considering gender adjustment. AR phosphorylation and regulon activity were modulated by FGFR3 in FGFR3-dependent models. ConclusionsMutant-FGFR3 is an oncogene per se, inducing bladder tumorigenesis. Patients with early stage bladder lesions could thus potentially benefit from FGFR3 targeting. Our results also reinforce the interest in elucidating the role of AR in bladder carcinogenesis, specifically in FGFR3-mutated driven tumors. Finally, our results suggest FGFR3 expression level in epithelium as a determinant for the FGFR3-driven tumors tissue specificity.

cancer biology↗

bioRxiv2020

STAG2 loss-of-function affects short-range genomic contacts and modulates urothelial differentiation in bladder cancer cells

Cohesin exists in two variants, containing either STAG1 or STAG2. STAG2 is one of the most commonly mutated genes in human cancer, and a major bladder cancer tumor suppressor. Little is known about how its inactivation contributes to tumor development. Here, we analyze the genomic distribution of STAG1 and STAG2 and perform STAG2 loss-of-function experiments using RT112 bladder cancer cells; we then analyze the resulting genomic effects by integrating gene expression and chromatin interaction data. Cohesin-STAG2 is required for DNA contacts within topological domains, but not for compartment maintenance of domain boundary integrity. Cohesin-STAG2-mediated interactions are short-ranged and engage promoters and gene bodies with higher frequency than those mediated by cohesin-STAG1. STAG2 knockdown resulted in a modest but consistent down-regulation of the luminal urothelial differentiation signature, mirroring differences between STAG2-high and STAG2-low bladder tumors. Both lost and gained contacts were enriched among STAG1/STAG2 common sites as well as STAG2-enriched sites. Contacts lost upon depletion of STAG2 were significantly assortative, indicating their proximity at the 3D level, and were associated with changes in gene expression. Overall, our findings indicate that, in urothelial cells, STAG2 is required for the establishment and/or maintenance of DNA looping that, in turn, sustains the luminal differentiation program. This mechanism may contribute to the tumor suppressor function of STAG2 in bladder cancer.

cancer biology↗

bioRxiv2019

PPARG is a tumor suppressor in basal bladder tumors offering new potential therapeutic opportunities.

PPAR{gamma} activation is a critical event in luminal muscle-invasive bladder cancer (MIBC) tumorigenesis, favoring both tumor cell growth and microenvironment modulation toward tumor immune escape. Conversely, the down-regulation of PPAR{gamma} activity in basal MIBC suggests tumor suppressive effects in this subgroup. Here, we report genetic, epigenetic and functional evidence to support the tumor suppressor role for PPAR{gamma} in basal bladder tumors. We identified hemizygous deletions, DNA hyper-methylation and loss-of-function mutations of PPAR{gamma} in basal MIBC, associated with PPAR{gamma} under-expression and its decreased activity. Re-expression of PPAR{gamma} in basal tumor cells resulted in the activation of PPAR{gamma} -dependent transcription program that modulated fatty acid metabolism and cell differentiation and decreased cell growth, which could partly rely on EGFR down-regulation. Structure-function studies of two PPAR{gamma} mutant revealed a destabilization of a region important for coactivator recruitment and should help develop potent molecules to activate PPAR{gamma} as a therapeutic strategy for basal MIBC. The identification of this subtype-dependent dual role of PPAR{gamma} in MIBC strengthens the critical role of PPAR{gamma} in bladder tumorigenesis and reinforces the interest in stratified medicine based on tumor molecular subtyping. One sentence summaryGenetic, epigenetic and functional evidence of a tumor suppressor role for PPAR{gamma} in basal bladder tumors offer new therapeutic opportunities for this subgroup.

cancer biology↗

bioRxiv2019

Role of a kinesin motor in cancer cell mechanics

Molecular motors play important roles in force generation, migration and intracellular trafficking. Changes in specific motor activities are altered in numerous diseases. KIF20A, a motor protein of the kinesin-6 family, is overexpressed in bladder cancer, and KIF20A levels correlate negatively with the clinical outcome. We report here a new role for the KIF20A kinesin motor protein in intracellular mechanics. Using optical tweezers to probe intracellular mechanics and surface AFM to probe cortical mechanics, we first confirm that bladder cells soften with increasing cancer grade. We then show that inhibiting KIF20A makes the intracellular environment softer for both high and low grade bladder cancer cells. Upon inhibition of KIF20A cortical stiffness also decreases in lower grade cells, while it surprisingly increases in higher grade malignant cells. Changes in cortical stiffness correlate with the interaction of KIF20A with myosin IIA. Moreover, KIF20A negatively regulates bladder cancer cell motility irrespective of the underlying substrate stiffness. Our results reveal a central role for a microtubule motor in cell mechanics and migration in the context of bladder cancer.

biophysics↗

bioRxiv2018

The consensus molecular classification of muscle-invasive bladder cancer

Muscle-Invasive Bladder Cancer (MIBC) is a molecularly diverse disease with heterogeneous clinical outcomes. Several molecular classifications have been proposed, yielding diverse sets of subtypes, which hampers the clinical implications of such knowledge. Here, we report the results of a large international effort to reach a consensus on MIBC molecular subtypes. Using 1750 MIBC transcriptomes and a network-based analysis of six independent MIBC classification systems, we identified a consensus set of six molecular classes: Luminal Papillary (24%), Luminal Non-Specified (8%), Luminal Unstable (15%), Stroma-rich (15%), Basal/Squamous (35%), and Neuroendocrine-like (3%). These consensus classes differ regarding underlying oncogenic mechanisms, infiltration by immune and stromal cells, and histological and clinical characteristics. This consensus system offers a robust framework that will enable testing and validating predictive biomarkers in future clinical trials.

cancer biology↗

bioRxiv2018

Stabilized Independent Component Analysis outperforms other methods in finding reproducible signals in tumoral transcriptomes

MotivationMatrix factorization methods are widely exploited in order to reduce dimensionality of transcriptomic datasets to the action of few hidden factors (metagenes). Applying such methods to similar independent datasets should yield reproducible inter-series outputs, though it was never demonstrated yet.\n\nResultsWe systematically test state-of-art methods of matrix factorization on several transcriptomic datasets of the same cancer type. Inspired by concepts of evolutionary bioinformatics, we design a new framework based on Reciprocally Best Hit (RBH) graphs in order to benchmark the methods reproducibility. We show that a particular protocol of application of Independent Component Analysis (ICA), accompanied by a stabilisation procedure, leads to a significant increase in the inter-series output reproducibility. Moreover, we show that the signals detected through this method are systematically more interpretable than those of other state-of-art methods. We developed a user-friendly tool BIODICA for performing the Stabilized ICA-based RBH meta-analysis. We apply this methodology to the study of colorectal cancer (CRC) for which 14 independent publicly available transcriptomic datasets can be collected. The resulting RBH graph maps the landscape of interconnected factors that can be associated to biological processes or to technological artefacts. These factors can be used as clinical biomarkers or robust and tumor-type specific transcriptomic signatures of tumoral cells or tumoral microenvironment. Their intensities in different samples shed light on the mechanistic basis of CRC molecular subtyping.\n\nAvailabilityThe BIODICA tool is available from https://github.com/LabBandSB/BIODICA.\n\nContactlaura.cantini@curie.fr and andrei.zinovyev@curie.fr\n\nSupplementary informationSupplementary data are available at Bioinformatics online.

systems biology↗