Léa Bellenger

Summary: One-third of skin melanomas arise from melanocytic nevi, benign skin lesions composed of clustered melanocytes. Benign nevi are associated with overactivation of the mitogen-activated protein kinase RAS/ERK pathway, resulting from driver mutations, most commonly in the BRAF or NRAS gene. However, this overactivation in melanocytes is insufficient to induce melanoma formation, as only a minority of benign nevi give rise to melanoma. Overactivation of the RAS/ERK pathway promotes genetic and epigenetic alterations by inducing aneuploidy, but the processes by which nevi evolve into melanoma via RAS/ERK pathway-dependent aneuploidy are only partially understood. Using single-nucleus RNA sequencing after overactivation of the RAS/ERK pathway in the chicken embryo, we discovered that RIPOR2 is a positive transcriptional target of this pathway, including in melanocyte precursors. Similar transcriptional control of RIPOR2 by RAS/ERK is conserved in human melanoma cells. RIPOR2 emerged as an attractive target because it encodes an atypical RHOA inhibitory protein involved in the development of physiologically multinucleated cell types. Multinucleation in cancer has been shown to promote aneuploidy, which correlates with tumor aggressiveness. We found that RIPOR2 is ectopically expressed in human nevi and skin melanomas and functionally promotes multinucleation in both an animal model and human tumor-derived cells, including melanoma cell lines. Our results suggest that RIPOR2 expression, downstream of RAS/ERK overactivation in skin melanocytes, promotes the emergence of multinucleated cells, a previously overlooked step in melanoma formation.

Tight developmental coordination of muscle attachments with skeletal muscle is fundamental for building functional limbs and allowing locomotion. Muscle attachments include connective tissue fibroblasts of tendon and muscle connective tissue. Although muscle attachments play critical roles in development, our understanding of connective tissue fibroblast developmental programs lags behind that of other components of the musculoskeletal system, mainly because fibroblasts are highly heterogeneous and poorly characterized. Combining single-cell RNA-sequencing-based strategies including trajectory inference and in situ hybridization analyses, we address the diversity of connective tissue fibroblasts and their developmental trajectories during chicken limb fetal development. We show that fibroblasts switch from a program providing positional information to a program of lineage diversification at the onset of fetal period. Muscle connective tissue and tendon contain several fibroblast populations that emerge asynchronously. Once the final muscle pattern is set, in silico fibroblast populations that are close in transcriptional identity are found in neighbouring locations in limbs, prefiguring the adult fibroblast layers associated to muscle. We propose that the limb connective tissue is organised in a continuum of promiscuous fibroblast identities, allowing for the robust and efficient connection of the muscular system to bone and skin. Highlights: Dissection of connective tissue heterogeneity using single-cell RNA-sequencing and in situ RNA staining In silico fibroblast populations map to distinct fibroblast layers in limbs Molecular signatures of fetal fibroblast populations at the origin of limb adult connective tissue Epimysium and endomysium fibroblasts are transcriptionally distinct at fetal stage Limb fibroblasts display promiscuous identities reflecting their interconnected 3D-network Fibroblasts switch from providing positional information to differentiation during development

T-cell acute lymphoblastic leukemia (T-ALL) is a hematopoietic malignancy characterized by an increased proliferation and incomplete maturation of T-cell progenitors. Despite therapeutic improvements, relapses are often of bad prognosis. Therapeutic vulnerabilities and chemoresistance mechanisms arising from cell plasticity induced by the bone marrow (BM) microenvironment remain an important field of investigation. Employing single cell RNA sequencing (scRNAseq) of human T-ALL cells recovered from adipocyte-rich and -poor BM, a distinct leukemic stem cell (LSC) population defined by quiescence and elevated CD44 level (Ki67 ^neg/low CD44 ^high ) expression is identified in both territories. In vivo chemotherapy demonstrated that the LSC population evades drug treatment. Patient sample analyses confirmed the presence of Ki67 ^neg/low CD44 ^high LSC both at diagnosis and relapse that displayed a specific transcriptomic signature. Interestingly, the intense expression of CD44 in T-ALL Ki67 ^neg/low LSC was associated with E-selectin binding. Importantly, when 39 human T-ALL samples were analyzed, the E-selectin binding ability was found significantly higher in Relapse/Refractory compared to drug-sensitive patients. These findings characterize a T-ALL LSC population with chemoresistant properties and shade light on new strategies for prognostic stratification while opening avenues for novel therapeutic options.

Positional information driving limb muscle patterning is contained in connective tissue fibroblasts but not in myogenic cells. Limb muscles originate from somites, while connective tissues originate from lateral plate mesoderm. With cell and genetic lineage tracing we challenge this model and identify an unexpected contribution of lateral plate-derived fibroblasts to the myogenic lineage, preferentially at the myotendinous junction. Analysis of single-cell RNA-sequencing data from whole limbs at successive developmental stages identifies a population displaying a dual muscle and connective tissue signature. BMP signalling is active in this dual population and at the tendon/muscle interface. In vivo and in vitro gain- and loss-of-function experiments show that BMP signalling regulates a fibroblast-to-myoblast conversion. These results suggest a scenario in which BMP signalling converts a subset of lateral plate mesoderm-derived cells to a myogenic fate in order to create a boundary of fibroblast-derived myonuclei at the myotendinous junction that controls limb muscle patterning.

Background: The vast ecosystem of single-cell RNA-sequencing tools has until recently been plagued by an excess of diverging analysis strategies, inconsistent file formats, and compatibility issues between different software suites. The uptake of 10x Genomics datasets has begun to calm this diversity, and the bioinformatics community leans once more towards the large computing requirements and the statistically driven methods needed to process and understand these ever-growing datasets.Results: Here we outline several Galaxy workflows and learning resources for single-cell RNA-sequencing, with the aim of providing a comprehensive analysis environment paired with a thorough user learning experience that bridges the knowledge gap between the computational methods and the underlying cell biology. The Galaxy reproducible bioinformatics framework provides tools, workflows, and trainings that not only enable users to perform 1-click 10x preprocessing but also empower them to demultiplex raw sequencing from custom tagged and full-length sequencing protocols. The downstream analysis supports a range of high-quality interoperable suites separated into common stages of analysis: inspection, filtering, normalization, confounder removal, and clustering. The teaching resources cover concepts from computer science to cell biology. Access to all resources is provided at the singlecell.usegalaxy.eu portal.Conclusions: The reproducible and training-oriented Galaxy framework provides a sustainable high-performance computing environment for users to run flexible analyses on both 10x and alternative platforms. The tutorials from the Galaxy Training Network along with the frequent training workshops hosted by the Galaxy community provide a means for users to learn, publish, and teach single-cell RNA-sequencing analysis.

Positional information driving limb muscle patterning is contained in lateral plate mesoderm-derived tissues, such as tendon or muscle connective tissue but not in myogenic cells themselves. The long-standing consensus is that myogenic cells originate from the somitic mesoderm, while connective tissue fibroblasts originate from the lateral plate mesoderm. We challenged this model using cell and genetic lineage tracing experiments in birds and mice, respectively, and identified a subpopulation of myogenic cells at the muscle tips close to tendons originating from the lateral plate mesoderm and derived from connective tissue gene lineages. Analysis of single-cell RNA-sequencing data obtained from limb cells at successive developmental stages revealed a subpopulation of cells displaying a dual muscle and connective tissue signature, in addition to independent muscle and connective tissue populations. Active BMP signalling was detected in this junctional cell sub-population and at the tendon/muscle interface in developing limbs. BMP gain- and loss-of-function experiments performed in vivo and in vitro showed that this signalling pathway regulated a fibroblast-to-myoblast conversion. We propose that localised BMP signalling converts a subset of lateral plate mesoderm-derived fibroblasts to a myogenic fate and establishes a boundary of fibroblast-derived myonuclei at the muscle/tendon interface to control the muscle pattern during limb development.

Background: The vast ecosystem of single-cell RNA-seq tools has until recently been plagued by an excess of diverging analysis strategies, inconsistent file formats, and compatibility issues between different software suites. The uptake of 10x Genomics datasets has begun to calm this diversity, and the bioinformatic community leans once more towards the large computing requirements and the statistically-driven methods needed to process and understand these ever-growing datasets. Results: Here we outline several Galaxy workflows and learning resources for scRNA-seq, with the aim of providing a comprehensive analysis environment paired with a thorough user learning experience that bridges the knowledge gap between the computational methods and the underlying cell biology. The Galaxy reproducible bioinformatic framework provides tools, workflows and trainings that not only enable users to perform one-click 10x preprocessing, but also empowers them to demultiplex raw sequencing data manually. The downstream analysis supports a wide range of high quality interoperable suites separated into common stages of analysis: inspection, filtering, normalization, confounder removal and clustering. The teaching resources cover an assortment of different concepts from computer science to cell biology. Access to all resources is provided at the singlecell.usegalaxy.eu portal. Conclusions: The reproducible and training-oriented Galaxy framework provides a sustainable HPC environment for users to run flexible analyses on both 10x and alternatively derived datasets. The tutorials from the Galaxy Training Network along with the frequent training workshops hosted by the Galaxy Community provide a means for users to learn, publish and teach scRNA-seq analysis.