Biology subjects

Capraz, T.

Publications and source records attributed to Capraz, T..

6 recordsLinked to original sources

bioRxiv2026

A human lung organoid co-culture model of early bone sarcoma metastasis reveals contact-dependent epithelial remodeling at the metastatic interface

Lung metastasis drives mortality across cancer types, yet how infiltrating tumor cells remodel the lung epithelium to establish metastatic niches remains poorly understood. Here we establish MESCUL (MEtastatic Sarcoma Co-CULture), a co-culture platform combining human lung organoids with patient-derived bone sarcoma cells to model early tumor-lung epithelial interactions in a physiologically relevant 3D system. MESCUL reveals that direct tumor-epithelial contact induces rapid, reproducible lung epithelial remodeling across Ewing sarcoma (ES) and osteosarcoma (OS) models and multiple organoid donor backgrounds, which is contact-dependent and not recapitulated by paracrine signaling. Single-cell RNA sequencing identifies LIMES (Lung Interface Metastasis Signature), a shared transcriptional program encompassing focal adhesion assembly, matrix metalloprotease (MMP) upregulation, and emergence of a damage-associated transitional cell state, in both ES and OS. Mechanistically, tumor-derived fibronectin (FN1) engages epithelial integrin receptors to activate focal adhesion kinase (FAK), driving amphiregulin (AREG) induction and MMP-mediated remodeling; FN1 alone phenocopies this response, and FAK inhibition attenuates it, nominating the FN1-integrin-FAK-AREG axis as a candidate therapeutic vulnerability. The LIMES program, identified through the MESCUL co-culture model, is spatially confined to the tumor-lung interface in patient metastases of both ES and OS, as demonstrated by spatial transcriptomics across nine patients. Massons trichrome staining of matched patient sections reveals pronounced collagen deposition in the peri-tumoral lung parenchyma, consistent with LIMES acting upstream of a wound-healing cascade that proceeds to structural fibrotic remodeling in patient tissue. Together, these findings establish the lung epithelium as an active participant in metastatic colonization, characterize a pharmacologically targetable, spatially restricted epithelial remodeling response at the bone sarcoma-lung interface across OS and ES, and introduce MESCUL as a tractable 3D platform for investigating lung metastasis, with possible implications for tumor types beyond bone sarcomas.

cancer biology↗

bioRxiv2024

Semi-supervised Omics Factor Analysis (SOFA) disentangles known sources of variation from latent factors in multi-omics data

A fundamental design pattern in biomolecular studies is to assay the same set of samples (organisms, tissue biopsies, or individual cells) by multiple different omics assays. Group Factor Analysis (GFA) and its adaptation to high-dimensional settings, Multi-Omics Factor Analysis (MOFA), are widely used as a first-line approach to analyse such data and are effective in detecting patterns of correlation, organize them into so-called latent factors, and identify common and assay-specific factors. However, in many applications a subset of the found factors just rediscovers already known covariates (e.g., disease subtypes, environmental covariates) while others may represent genuine novelty. Here, we present Semi-supervised Omics Factor Analysis (SOFA), a method that incorporates known covariates into the model upfront and focuses the factor discovery on novel sources of variation. We show SOFAs effectiveness for discovering novel patterns by applying it to cancer, brain development and heart failure multi-omic data sets.

bioinformatics↗

bioRxiv2023

Feature selection by replicate reproducibility and non-redundancy

MotivationA fundamental step in many analyses of high-dimensional data is dimension reduction. Two basic approaches are introduction of new, synthetic coordinates, and selection of extant features. Advantages of the latter include interpretability, simplicity, transferability and modularity. A common criterion for unsupervised feature selection is variance or dynamic range. However, in practice it can occur that high-variance features are noisy, that important features have low variance, or that variances are simply not comparable across features because they are measured in unrelated numeric scales or physical units. Moreover, users may want to include measures of signal-to-noise ratio and non-redundancy into feature selection. ResultsHere, we introduce the RNR algorithm, which selects features based on (i) the reproducibility of their signal across replicates and (ii) their non-redundancy, measured by linear independence. It takes as input a typically large set of features measured on a collection of objects with two or more replicates per object. It returns an ordered list of features i1, i2, ..., ik, where feature i1 is the one with the highest reproducibility across replicates, i2 that with the highest reproducibility across replicates after projecting out the dimension spanned by i1, and so on. Applications to microscopy based imaging of cells and proteomics experiments highlight benefits of the approach. AvailabilityThe RNR method is implemented in the R package FeatSeekR and is available via Bioconductor (Huber et al., 2015) under the GPL-3 open source license. Contacttuemay.capraz@embl.de

bioinformatics↗

bioRxiv2023

Proteomic profiling identifies Serpin B9 as mediator of resistance to CAR T-cell and bispecific antibody treatment in B-cell lymphoma

Although T-cell-engaging therapies are highly effective in patients with relapsed and/or refractory B-cell non-Hodgkin lymphoma (B-NHL), responses are often not durable. To identify tumor-intrinsic drivers of resistance, we quantified in-vitro response to CD19-directed chimeric antigen receptor T-cells (CD19-CAR) and bispecific antibodies (BsAb) across 46 B-NHL cell lines and measured their proteomic profiles at baseline. Among the proteins associated with poor in-vitro response was Serpin B9, an endogenous granzyme B inhibitor. Knock-out of SERPINB9 in cell lines with high intrinsic expression rendered them more susceptible to CD19-CAR and CD19-BsAb. Overexpression in cell lines with low intrinsic expression attenuated responses. Polatuzumab, vorinostat, lenalidomide, or checkpoint inhibitors improved response to CD19-CAR, although independently of Serpin B9 expression. Besides providing an important resource of therapy response and proteomic profiles, this study refines our understanding of resistance in T-cell engaging therapies, and suggests clinically relevant combination regimes.

cancer biology↗

bioRxiv2021

Structure-guided glyco-engineering of ACE2 for improved potency as soluble SARS-CoV-2 decoy receptor

Infection and viral entry of SARS-CoV-2 crucially depends on the binding of its Spike protein to angiotensin converting enzyme 2 (ACE2) presented on host cells. Glycosylation of both proteins is critical for this interaction. Recombinant soluble human ACE2 can neutralize SARS-CoV-2 and is currently undergoing clinical tests for the treatment of COVID-19. We used 3D structural models and molecular dynamics simulations to define the ACE2 N-glycans that critically influence Spike-ACE2 complex formation. Engineering of ACE2 N-glycosylation by site-directed mutagenesis or glycosidase treatment resulted in enhanced binding affinities and improved virus neutralization without notable deleterious effects on the structural stability and catalytic activity of the protein. Importantly, simultaneous removal of all accessible N-glycans from recombinant soluble human ACE2 yields a superior SARS-CoV-2 decoy receptor with promise as effective treatment for COVID-19 patients.

biochemistry↗

bioRxiv2021

ACE2 is the critical in vivo receptor for SARS-CoV-2 in a novel COVID-19 mouse model with TNF- and IFNγ-driven immunopathology

Despite tremendous progress in the understanding of COVID-19, mechanistic insight into immunological, disease-driving factors remains limited. We generated maVie16, a mouse-adapted SARS-CoV-2, by serial passaging of a human isolate. In silico modelling revealed how Spike mutations of maVie16 enhanced interaction with murine ACE2. MaVie16 induced profound pathology in BALB/c and C57BL/6 mice and the resulting mouse COVID-19 (mCOVID-19) replicated critical aspects of human disease, including early lymphopenia, pulmonary immune cell infiltration, pneumonia and specific adaptive immunity. Inhibition of the proinflammatory cytokines IFN{gamma} and TNF substantially reduced immunopathology. Importantly, genetic ACE2-deficiency completely prevented mCOVID-19 development. Finally, inhalation therapy with recombinant ACE2 fully protected mice from mCOVID-19, revealing a novel and efficient treatment. Thus, we here present maVie16 as a new tool to model COVID-19 for the discovery of new therapies and show that disease severity is determined by cytokine-driven immunopathology and critically dependent on ACE2 in vivo. Key pointsO_LIThe mouse-adapted SARS-CoV-2 strain maVie16 causes fatal disease in BALB/c mice and substantial inflammation, pneumonia and immunity in C57BL/6 mice C_LIO_LITNF/IFN{gamma} blockade ameliorates maVie16-induced immunopathology C_LIO_LIMaVie16 infection depends on ACE2 and soluble ACE2 inhalation can prevent disease C_LI

immunology↗