Biology subjects

Leonetti, M.

Publications and source records attributed to Leonetti, M..

12 recordsLinked to original sources

bioRxiv2026

A high-resolution protein-RNA binding atlas for structure-aware interaction prediction and post-transcriptional regulatory modeling

RNA-binding proteins (RBPs) control post-transcriptional gene expression by regulating RNA processing, localization, stability, and translation. However, our understanding of RNA regulation remains limited by the resolution, scale, and uniformity of available RBP binding maps. Here we introduce eclair CLIP (ecCLIP), a high-throughput CLIP-seq platform that combines endogenous protein tagging for uniform affinity capture with ligation-free library preparation to generate high-resolution RNA-protein binding profiles. We applied ecCLIP to 67 endogenously tagged RBPs in HEK293T cells, generating a high-quality transcriptome-scale atlas that recovered known binding preferences and captured post-transcriptional regulons linked to RNA half-life. To test the utility of the atlas, we incorporated ecCLIP into Borzuki, a Borzoi-derived sequence-to-function model trained on complementary RNA degradation, expression, CLIP, and Perturb-seq data. Inclusion of ecCLIP data improved RNA half-life prediction and assessment of 3' UTR variant effects on RNA stability, demonstrating that our atlas helps capture RNA dynamics. To directly model the cis-regulatory grammar underlying these RBP-RNA interactions, we developed Pythia, a structure-aware neural network architecture that incorporates context-free grammar-inspired fixed convolutions to explicitly encode RNA base-pairing features. Pythia predicted RBP binding from sequence and structural context and decomposed predictions into sequence- and structure-dependent components. We tested this decomposition for five RBPs using massively parallel reporter CLIP (MPRC) binding assays in cells, using synthetic RNA libraries containing mutations designed to preserve or disrupt predicted RNA structures across nominated binding sites. We used the same MPRC strategy to confirm allele-specific binding effects for variants prioritized by Pythia. Together, ecCLIP and Pythia provide a resource for structure-aware decoding of RNA-protein interaction grammar and interpretation of post-transcriptional regulatory variation.

genomics↗

bioRxiv2025

A low-dose immunotherapy targeting Fc-gamma Receptors and Heparan Sulfate Proteoglycan to impact myeloid cells and control tumor growth in cancers with varying immunosuppressive profiles

BackgroundMyeloid cells play a central role in cancer-associated immunosuppression. Indeed, their accumulation and functional reprogramming attenuate effective immune responses and may facilitate tumor progression. To modulate the activity of the myeloid subsets, we chose to target Fc{gamma} receptors (Fc{gamma}Rs), since all myeloid cell populations variably express Fc{gamma}Rs. As Fc{gamma}RIIb provides an inhibitory signaling that might alter efficacy, we developed an immunotherapy, active at a low dose to limit binding via this low-affinity receptor while still interacting with higher-affinity Fc{gamma}Rs. We engineered a Fc-based fusion protein whose activity is potentiated by its ability to engage both Fc{gamma}Rs and a coreceptor, Heparan Sulfate Proteoglycan (HSPGs). We designed and fused a HSPG-ligand, named T54, to a human IgG1-Fc molecule to produce the Fc-T54 fusion protein. MethodsFc-T54 was produced using recombinant technologies. Binding characteristics were assessed using ELISA and flow-cytometric assays. Immune activity was investigated using cell-culture assays. Ability to affect tumor growth was investigated using four syngeneic tumor models with deserted to inflamed characteristics that differ in their sensitivity to immune checkpoint inhibitors (ICIs). Tumor microenvironment (TME) was analysed by flow cytometry. ResultsCompared to Fc, Fc-T54 demonstrates superior binding to low-affinity Fc{gamma}Rs and interacts more effectively with human leukocytes, including neutrophils, and B lymphocytes, as well as with monocytes and dendritic cells (DCs) within peripheral blood mononuclear cells. In activation assays Fc-T54 increases the number of monocytes/macrophages and B-lymphocytes, decreases neutrophil abundance, and enhances DC activation. Fc-T54 also demonstrates enhanced interaction with murine DCs accompanied by increased activation. Subcutaneous administration of low dose Fc-T54 - or its murine surrogate - significantly inhibits tumor growth in immune-deserted and immune-excluded mouse models, and synergizes with anti-PD-1 therapy in the immune-inflamed model. TME analysis in the MB49 bladder cancer model reveals that the immunotherapy decreases the proportion of granulocytic myeloid-derived suppressor cells while increasing CD8+ T-cells and natural killer cells, promoting a microenvironment more prone to tumor control. ConclusionsThis new Fc{gamma}R/HSPG-engaging immunotherapy, administered via subcutaneous route, offers a novel approach to modulate the myeloid compartment and expand therapeutic options for ICI-resistant, deserted/excluded tumors, and for inflamed tumors when used in combination regimens.

cancer biology↗

bioRxiv2024

Bone pain in Fibrous dysplasia does not rely on aberrant sensory nerve sprouting or neuroma formation

Bone pain is a major symptom of many skeletal disorders. Fibrous dysplasia (FD) is a genetic disease with mono or polyostotic skeletal phenotype due to the post-zygotic occurrence of the causative Gs mutation. Bone pain in FD often associates with skeletal deformities and fractures or nerve impingement by the pathological tissue. However, even in the absence of complications, FD patients often complain of a chronic pain that does not correlate with their disease burden. Multiple hypotheses have been made to explain this pain. However, its pathogenetic mechanisms remain, as yet, largely unexplored. In this study, we first demonstrate that the FD mouse model EF1-GsR201C develops a painful-like behavior and an altered response to nociceptive stimuli that, as in FD patients, do not correlate with the severity of their phenotype, thus providing a reliable model to study bone pain in FD. Then, we show that in EF1-GsR201C mice, the overall pattern of skeletal innervation is preserved and that within FD lesions, sensory fibers are variably and focally distributed, mainly at perivascular sites. Finally, we provide the first analysis of a series of human FD bone biopsies showing that sensory nerve fibers are rarely detected within the pathological tissue. These data confirm that bone pain is an intrinsic and reproducible feature of FD. They also show that, albeit sensory nerve fibers are found within FD lesions and may contribute to the unpleasant sensation that accompanies the disease, pathological sensory nerve sprouting or formation of neuromas are not detected in the Gs-mutated skeleton.

pathology↗

bioRxiv2023

PROTOSPACEJAM: AN OPEN-SOURCE, CUSTOMIZABLE AND WEB-ACCESSIBLE DESIGN PLATFORM FOR CRISPR/CAS9 INSERTIONAL KNOCK-IN

CRISPR/Cas9-mediated knock-in of DNA sequences enables precise genome engineering for research and therapeutic applications. However, designing effective guide RNAs (gRNAs) and homology-directed repair (HDR) donors remains a bottleneck. Here, we present protoSpaceJAM, an open-source algorithm to automate and optimize gRNA and HDR donor design for CRISPR/Cas9 insertional knock-in experiments. protoSpaceJAM utilizes biological rules to rank gRNAs based on specificity, distance to insertion site, and position relative to regulatory regions. protoSpaceJAM can introduce recoding mutations (silent mutations and mutations in non-coding sequences) in HDR donors to prevent re-cutting and increase knock-in efficiency. Users can customize parameters and design double-stranded or single-stranded donors. We validated protoSpaceJAMs design rules by demonstrating increased knock-in efficiency with recoding mutations and optimal strand selection for single-stranded donors. An additional module enables the design of genotyping primers for next-generation sequencing of edited alleles. Overall, protoSpaceJAM streamlines and optimizes CRISPR knock-in experimental design in a flexible and modular manner to benefit diverse research and therapeutic applications. protoSpaceJAM is available open-source as an interactive web tool at protospacejam.czbiohub.org or as a standalone Python package at github.com/czbiohub-sf/protoSpaceJAM.

bioinformatics↗

bioRxiv2022

Correlative imaging of spatio-angular dynamics of molecular assemblies and cells with multimodal instant polarization microscope

Biological function depends on the spatio-angular architecture of macromolecules - for example, functions of lipid membrane and cytoskeletal polymers arise from both the spatial and the angular organization of the constituent molecules. Correlative imaging of cellular and molecular architecture is valuable across cell biology and pathology. However, current live imaging methods primarily focus on spatial component of the architecture. Imaging the dynamic angular architecture of cells and organelles requires fast polarization-, depth-, and wavelength-diverse measurement of intrinsic optical properties and fluorophore concentration, but remains challenging with current designs. We report a multimodal instant polarization microscope (miPolScope) that combines a broadband polarization-resolved detector, automation, and reconstruction algorithms to enable label-free imaging of phase, retardance, and orientation, multiplexed with fluorescence imaging of concentration, anisotropy, and orientation of molecules at diffraction-limited resolution and high speed. miPolScope enabled multimodal imaging of myofibril architecture and contractile activity of beating cardiomyocytes, cell and organelle architecture of live HEK293T and U2OS cells, and density and anisotropy of white and grey matter of mouse brain tissue across the visible spectrum. We anticipate these developments in joint quantitative imaging of density and anisotropy to enable new studies in tissue pathology, mechanobiology, and imaging-based screens.

bioengineering↗

bioRxiv2021

mTOR controls neurotoxic lysosome exocytosis in inflammatory reactive astrocytes

Astrocytes respond and contribute to neuroinflammation by adopting inflammatory reactive states. Although recent efforts have characterized the gene expression signatures associated with these reactive states, the cell biology underlying inflammatory reactive astrocyte phenotypes remains under-explored. Here, we used CRISPR-based screening in human iPSC-derived astrocytes to identify mTOR activation a driver of cytokine-induced endolysosomal system remodeling, manifesting as alkalinization of endolysosomal compartments, decreased autophagic flux, and increased exocytosis of certain endolysosomal cargos. Through endolysosomal proteomics, we identified and focused on one such cargo - IL-32, a disease-associated pro-inflammatory cytokine not present in rodents, whose secretion mechanism is not well understood. We found that IL-32 was partially secreted in extracellular vesicles likely to be exosomes. Furthermore, we found that IL-32 was involved in the polarization of inflammatory reactive astrocyte states, was upregulated in astrocytes in multiple sclerosis lesions, and preferentially co-localized with astrocytes in hypoxic-ischemic encephalopathy. We believe that our results advance our understanding of cell biological pathways underlying inflammatory reactive astrocyte phenotypes and identify potential therapeutic targets.

cell biology↗

bioRxiv2021

A technique for resolution assessment in blind-SIM experiments

Super resolution techniques are an excellent alternative to wide field microscopy, providing high resolution also in (typically fragile) biological sample. Among the various super resolution techniques, Structured Illumination Microscopy (SIM) improve resolution by employing multiple illumination patterns to be deconvolved with a dedicated software. In the case of blind SIM techniques, unknown patterns, such as speckles, are used, thus providing super resolved images, nearly unaffected by aberrations with a simplified experimental setup. Scattering Assisted Imaging, a special blind SIM technique, exploits an illumination PSF (speckle grains size), smaller than the collection PSF (defined by the collection objectives), to surpass the typical SIM resolution enhancement. However, if SAI is used, it is very difficult to extract the resolution enhancement from a priori considerations. In this paper we propose a protocol and experimental setup for the resolution measurement, demonstrating the resolution enhancement for different collection PSF values.

biophysics↗

bioRxiv2021

Self-Supervised Deep-Learning Encodes High-Resolution Features of Protein Subcellular Localization

Elucidating the diversity and complexity of protein localization is essential to fully understand cellular architecture. Here, we present cytoself, a deep-learning approach for fully self-supervised protein localization profiling and clustering. cytoself leverages a self-supervised training scheme that does not require pre-existing knowledge, categories, or annotations. Training cytoself on images of 1,311 endogenously labeled proteins from the OpenCell database reveals a highly resolved protein localization atlas that recapitulates major scales of cellular organization, from coarse classes such as nuclear, cytoplasmic and vesicular, to the subtle localization signatures of individual protein complexes. We quantitatively validate cytoselfs ability to cluster proteins into organelles and protein complex clusters using a clustering score, and show that cytoself attains higher scores than previous unsupervised or self-supervised approaches. Finally, to better understand the inner workings of our model, we dissect the emergent features from which our clustering is derived, interpret these features in the context of the fluorescence images, and analyze the performance contributions of the different components of our approach.

cell biology↗

bioRxiv2020

Getting stuck in a rut as an emergent feature of a dynamic decision-making system

Human sensorimotor decision-making has a tendency to get stuck in a rut, being biased towards selecting a previously implemented action structure ( hysteresis). Existing explanations cannot provide a principled account of when hysteresis will occur. We propose that hysteresis is an emergent property of a dynamical system learning from the consequences of its actions. To examine this, 152 participants moved a cursor to a target on a tablet device whilst avoiding an obstacle. Hysteresis was observed when the obstacle moved sequentially across the screen between trials, but not with random obstacle placement. Two further experiments (n = 20) showed an attenuation when time and resource constraints were eased. We created a simple computational model capturing dynamic probabilistic estimate updating that showed the same patterns of results. This provides the first computational demonstration of how sensorimotor decision-making can get stuck in a rut through the dynamic updating of its probability estimates. Significance StatementHumans show a bias to select the organisational structure of a recently carried out action, even when an alternative option is available with lower costs. This hysteresis is said to be more efficient than creating a new plan and it has been interpreted as a design feature within decision-making systems. We suggest such teleological arguments are redundant, with hysteresis being a naturally emergent property of a dynamic control system that evolved to operate effectively in an uncertain and partially observable world. Empirical experimentation and simulations from a first principle computational model of decision-making were consistent with our hypothesis. The identification of such a mechanism can inform robotics research, suggesting how robotic agents can show human-like flexibility in complex dynamic environments.

neuroscience↗

bioRxiv2020

Mapping of the nucleolar proteome reveals spatiotemporal organization related to intrinsic protein disorder

The nucleolus is essential for ribosome biogenesis and is involved in many other cellular functions. We performed a systematic spatiotemporal dissection of the human nucleolar proteome using confocal microscopy. In total, 1,318 nucleolar proteins were identified; 287 were localized to fibrillar components, and 157 were enriched along the nucleoplasmic border, indicating a potential fourth nucleolar subcompartment (nucleoli rim). We found 65 nucleolar proteins (36 uncharacterized) to relocate to the chromosomal periphery during mitosis. Interestingly, we observed temporal partitioning into two recruitment phenotypes: early (prometaphase) and late (after metaphase), suggesting phase-specific functions. We further show that expression of MKI67 is critical for this temporal partitioning. We provide the first proteome-wide analysis of intrinsic protein disorder for the human nucleolus and show that nucleolar proteins in general, and mitotic chromosome proteins in particular, have significantly higher intrinsic disorder level compared to cytosolic proteins. In summary, this study provides a comprehensive and essential resource of spatiotemporal expression data for the nucleolar proteome as part of the Human Protein Atlas.

cell biology↗

bioRxiv2019

Conserved function of ether lipids and sphingolipids in the early secretory pathway

Sphingolipids have been shown to play important roles in physiology and cell biology, but a systematic examination of their functions is lacking. We performed a genome-wide CRISPRi screen in sphingolipid-depleted cells and identified hypersensitive mutants in genes of membrane trafficking and lipid biosynthesis, including ether lipid synthesis. Systematic lipidomic analysis showed a coordinate regulation of ether lipids with sphingolipids, where depletion of one of these lipid types resulted in increases in the other, suggesting an adaptation and functional compensation. Biophysical experiments on model membranes show common properties of these structurally diverse lipids that also share a known function as GPI anchors in different kingdoms of life. Molecular dynamics simulations show a selective enrichment of ether phosphatidylcholine around p24 proteins, which are receptors for the export of GPI-anchored proteins and have been shown to bind a specific sphingomyelin species. Our results support a model of convergent evolution of proteins and lipids, based on their physico-chemical properties, to regulate GPI-anchored protein transport and maintain homeostasis in the early secretory pathway.

biochemistry↗

bioRxiv2019

Revealing architectural order with polarized light imaging and deep neural networks

Quantitative imaging of biological architecture with fluorescent labels is not as scalable as genomic or proteomic measurements. Here, we combine quantitative label-free imaging and deep neural networks for scalable analysis of complex structures. We reconstruct quantitative three-dimensional density, anisotropy, and orientation in live cells and tissue slices from polarization- and depth-resolved images. We report a computationally efficient variant of U-Net architecture that predicts a 3D fluorescent structure from its morphology and physical properties. We evaluate the performance of our models by predicting F-actin and nuclei in mouse kidney tissue. Further, we report label-free imaging of axon tracts and predict level of myelination in human brain tissue sections. We demonstrate the model's ability to rescue inconsistent labeling. We anticipate that the proposed approach will enable quantitative analysis of architectural order across scales of organelles to tissues.

biophysics↗