Biology subjects

Rompais, M.

Publications and source records attributed to Rompais, M..

3 recordsLinked to original sources

bioRxiv2026

Dynamic 2'-O-Me of rRNA regulates ribosome translation in Arabidopsis nucleolin1 mutants and during seed-to-seedling transition

Across all kingdoms of life, 2'-O-ribose methylation (2'-O-Me) is the most abundant modification of ribosomal RNAs (rRNAs), which are central to ribosome assembly and function. The 2'-O-Me of specific rRNA sites is catalysed by C/D-box small nucleolar ribonucleoprotein (snoRNP) complexes. However, how and when the landscape of plant rRNA 2'-O-Me modifications is remodelled remains largely unexplored. To study the regulation of rRNA 2'-O-Me and its impact on ribosome translation, we characterised Arabidopsis NUCLEOLIN 1 (NUC1)-depleted plants, which exhibit reduced 2'-O-Me at dozens specific rRNA sites, and investigated the impact of 2'-O-Me in ribosome translation during plant development. We demonstrated that 2'-O-Me is dynamically regulated during seed-to-seedling transition. Our analysis identified differentially methylated rRNA sites, which are specifically located in structurally critical regions of the ribosome. Notably, this hypomethylation is uncoupled from the global expression, assembly, or subcellular localisation of C/D-box snoRNPs. We further observed that polysomes with low 2'-O-Me levels not only accumulate but also exhibit differential mRNA loading. These polysomes characterised by reduced 2'-O-Me at specific rRNA sites, influence multiples aspects of translation, including ribosomes stalling at the initiation, elongation and termination steps. Together, these findings advance our understanding of the dynamic regulation of rRNA 2'-O-methylation and its functional role in ribosome-mediated translation in plants.

plant biology↗

bioRxiv2025

Therapeutic poxviruses induce the secretion of immunostimulating and anti-tumoral extracellular vesicles

Poxvirus-based vectors provide a versatile cancer immunotherapy platform, enabling the expression of immunostimulatory molecules and cancer-specific antigens. While infections with pathogenic viruses are well known to modulate extracellular vesicle (EV) biogenesis and function, the extent to which therapeutic poxviral vectors influence EV secretion by immune cells and thereby affect therapeutic efficacy remains underexplored. In this study, we showed that poxviruses, including the clinically relevant Modified Vaccinia Ankara (MVA), stimulate the secretion of small EVs (sEVs) containing viral proteins and immune-related signatures from peripheral blood mononuclear cells (PBMCs). Using an engineered MVA vector, we demonstrated the transfer of virus-encoded therapeutic payloads to sEVs, including the model ovalbumin (OVA)-derived peptide SIINFEKL presented by the class I major histocompatibility complex (MHC I) and the immune activators interleukin-12 (IL-12) and CD40 ligand (CD40L). Depending on the isolation method, these sEVs stimulated SIINFEKL-specific CD8 T cells with varying efficiencies in vitro. Remarkably, intravenous injection of these sEVs into E.G7-OVA lymphoma-bearing mice reduced tumor growth to an extent comparable to the virus itself. Taken together, our findings indicate that EVs released from immune cells infected with engineered therapeutic poxviruses exert potent antitumor activity. These vesicles represent actionable mediators whose secretion and functionalization can be harnessed to improve viral vector-based immunotherapies, as well as being considered as therapeutic vectors in their own. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=95 SRC="FIGDIR/small/677320v1_ufig1.gif" ALT="Figure 1"> View larger version (20K): org.highwire.dtl.DTLVardef@1eae18borg.highwire.dtl.DTLVardef@17d6fb5org.highwire.dtl.DTLVardef@311577org.highwire.dtl.DTLVardef@7839c7_HPS_FORMAT_FIGEXP M_FIG C_FIG

cancer biology↗

bioRxiv2025

DeepLC introduces transfer learning for accurate LC retention time prediction and adaptation to substantially different modifications and setups

While LC retention time prediction of peptides and their modifications has proven useful, widespread adoption and optimal performance are hindered by variations in experimental parameters. These variations can render retention time prediction models inaccurate and dramatically reduce the value of predictions for identification, validation, and DIA spectral library generation. To date, mitigation of these issues has been attempted through calibration or by training bespoke models for specific experimental setups, with only partial success. We here demonstrate that transfer learning can successfully overcome these limitations by leveraging pre-trained model parameters. Remarkably, this approach can even fit highly performant models to substantially different peptide modifications and LC conditions than those on which the model was originally trained. This impressive adaptability of transfer learning makes it a highly robust solution for accurate peptide retention time prediction across a very wide variety of imaginable proteomics workflows.

bioinformatics↗