Biology subjects

Koszul, R.

Publications and source records attributed to Koszul, R..

At least 19 recordsLinked to original sources

bioRxiv2026

Genome-wide chromatin reset at the onset of leg regeneration in pancrustaceans

Regeneration requires injured tissues to selectively reactivate certain developmental programs while triggering regeneration-specific mechanisms and responding to the immediate consequences of tissue damage. How chromatin states are reorganized during this transition, and whether the underlying sequence of regulatory events is conserved across regenerating animals, remain poorly understood. We address these questions in the two-spotted cricket Gryllus bimaculatus, in which limb regeneration progressively declines during nymphal development and is absent in adults, and extend our analysis to the crustacean Parhyale hawaiensis. We first establish a chromosome-scale genomic and epigenomic framework for the cricket and then follow chromatin and transcriptional changes after limb amputation. Amputation triggers a striking genome-wide shutdown of enhancer-associated chromatin, marked by loss of accessible regulatory regions and depletion of H3K27ac and H3K4me1. This is followed, in regeneration-competent tissue, by selective reopening of a restricted population of regulatory elements and strong induction of the pioneer transcription factor Zelda (GbZelda). The GbZelda locus is unusually marked by both H3K27me3 and H3K9me3 in intact leg tissue; after injury, H3K9me3 is reduced broadly, whereas H3K27me3 is selectively relieved in regenerating nymphs. Finally, analysis of Parhyale reveals the same sequence of chromatin events during appendage regeneration: an initial enhancer shutdown followed by selective regulatory reopening concomitant with Zelda upregulation. Thus, rather than restoring the pre-injury regulatory landscape wholesale, appendage regeneration in these two pancrustacean models proceeds through a conserved sequence of genome-wide chromatin resetting, selective regulatory reopening and activation of a pioneer transcription factor.

developmental biology↗

bioRxiv2026

Chromatin is dispensable for bacterial life

Inside cells, DNA is intimately associated with proteins, forming chromatin. The protein constituents of chromatin vary across the tree of life: histones are the principal building blocks of chromatin in eukaryotes and many archaea, whereas bacteria typically encode a collection of nucleoid-associated proteins (NAPs) that wrap, bend, bridge or coat the DNA. Although chromatin proteins appear to be a universal feature of cellular life, DNA-templated processes such as transcription, replication, and DNA repair can take place in vitro in the absence of chromatin, raising the possibility that cellular systems might exist -- or could be built -- that lack chromatin proteins. To explore this possibility, the molecular consequences and potential systemic adjustments required for life without chromatin, we serially deleted the nine most abundant NAPs from E. coli (hupA , hupB , ihfA , ihfB , hns , stpA , fis , dps , lrp), resulting in a strain ({Delta}NAP9) that lacks its native chromatin. Using an array of different techniques, we document change -- and sometimes surprising lack thereof -- in compaction, composition and 3D architecture of the nucleoid, supercoiling, prophage activity, growth, viability, and genetic make-up of {Delta}NAP9. Most notably, we find that {Delta}NAP9 exhibits global dysregulation of gene expression, marked by a striking homogenization of transcriptional output across the genome that is reminiscent of the effects of histone depletion in eukaryotic cells. Our results reinforce the notion that chromatin plays a key role in compartmentalizing the use of genomic information, enabling both the localized suppression of selfish elements and dynamic reprogramming of genome activity in response to environmental change. At the same time, the successful construction of {Delta}NAP9 demonstrates that bacterial cells can carry out basic cellular functions in the absence of co-evolved chromatin proteins, highlighting the potential for radical (re-)engineering of prokaryotic chromatin and systems of gene expression.

microbiology↗

bioRxiv2025

Bacteriophage PAK_P3 genome structuration and dynamics during infection of Pseudomonas aeruginosa reveal specific interactions patterns

Bacteriophages, or phages, are highly abundant and diverse genomic entities that play an important role in microbial ecology, evolution, and horizontal gene transfer. While the dynamic changes in genome organization in cellular organisms have been well described, the 3D folding of phage genomes during the infection of their host is extremely limited. Understanding how phage genomes fold, invade and rearrange the genome of their host to functionally organize the optimal expression of their genes remains unknown. Here, we explore the spatial dynamics of the virulent double-stranded DNA phage PAK_P3 during infection of its host Pseudomonas aeruginosa and reveal how its genome rapidly decondenses to adopt a specific 3D organization that reflects its transcriptional program. Concomitantly, the host genome decondenses as gene expression wanes and transcription induced domains vanish. We also uncover specific and discrete bridging of the host and the phage genomes, showing how the phage genome exploits the spatial genome architecture of its host to succeed in its infection cycle. Our data highlights an unprecedented level of genome folding and gene expression regulation during a viral infection.

microbiology↗

bioRxiv2025

Multi-modal data integration for machine learning applications

The integration of multi-modal genomic data, encompassing sequences, annotations, and coverage tracks, remains a major bottleneck in bioinformatics, both for exploratory data analysis and machine learning applications. Current approaches rely on several specialized tools for different data modalities, leading to inefficient workflows and computational overhead. Here, we present momics, a unified framework to consolidate multi-omics data in a single repository and interrogate it with a high-performance query engine. Compared to existing tools, momics ingests genomic sequences, feature annotations, and unlimited coverage tracks into TileDB-backed repositories, and provides a scalable query engine for concurrent multi-modal queries across millions of genomic loci. Our benchmarks demonstrate up to 20-fold better data compression and up to 100-fold speed improvements over standard tools like pyBigWig, with a sublinear time complexity ideal for large-scale queries. Momics provides a python library optimized for exploratory data analysis and machine learning workflows, natively supporting current state-of-the-art bioinformatic ecosystems and cloud storage systems. We demonstrate momics utility through two real-world applications: (1) multi-modal data integration of hundreds of ChIP-seq datasets together with genomic sequence, and (2) multi-modal deep learning for chromatin accessibility prediction. By eliminating the need for multiple data parsing tools and providing a unified interface for all genomic data types, momics represents a paradigm shift in how large-scale multi-omics data can be managed and analyzed. Key pointsO_LImomics is a unified framework to consolidate sequences, annotations, and coverage tracks into a single queryable repository, addressing the critical bottleneck in genomic data analysis where researchers must juggle multiple specialized tools for different data modalities. C_LIO_LIWe show that momics can achieve up to 20-fold better data compression and 100-fold speed improvements over standard tools, with sublinear time complexity when querying millions of genomic positions simultaneously. C_LIO_LIWe use momics to formally demonstrate that multi-modal deep learning models can outperform single-modality approaches in predicting chromatin accessibility, achieving correlation of 0.84 when training with a combination of genomic sequence and MNase data. C_LIO_LIOur results establish a new paradigm for reproducible multi-omics modeling, where entire multi-omics analysis workflows from data storage to machine learning model training can be replicated. C_LI

bioinformatics↗

bioRxiv2025

Dynamics and interplay of gene expression and chromosome organization across a predatory lifecycle

The obligate predatory bacterium Bdellovibrio bacteriovorus alternates between a motile attack phase and a growth phase inside another bacterium, during which it undergoes multiple rounds of DNA replication followed by non-binary division and progeny release. How its chromosome is dynamically reorganized across these contrasting states remains unclear. Here, we used synchronized predatory infections combined with time-resolved RNA-seq, Hi-C, ChIP-seq, and quantitative microscopy to map transcriptional activity and chromosome architecture across seven key stages of the predatory lifecycle. We uncover a dramatic shift in nucleoid organization - from an ultra-compact, transcriptionally restricted state in attack-phase cells to a progressively decompacted, ori-centered conformation during intracellular replication. Notably, distinct transcriptional and chromosome folding states not only mark the canonical attack and growth phases, but also critical transitions such as prey invasion, replication onset, and predator division. Temporal gene expression profiles reflect the coordinated regulation of core cellular functions across the predatory cycle. Furthermore, spatial localization of RNA polymerase, ribosomes, DNA-binding proteins, and freely diffusing proteins highlight the tight, cell cycle-dependent coupling between nucleoid compaction, chromosome accessibility, and transcriptional activity. Altogether, our findings reveal the intricate dynamics and coordination of genome folding and gene expression in a complex bacterial cell cycle driven by prey-predator interactions.

microbiology↗

bioRxiv2025

Hicberg: Reconstruction of contact signals from repeated elements

In the course of their evolution, genomes can acquire various repeated elements, such as transposons, ribosomal DNA, duplicated genes or tandem repeats. These types of sequences cannot be processed directly by current high-throughput sequencing pipelines, as they generate short reads that cannot be unambiguously localized on reference genomes. We propose an algorithm called Hicberg that uses statistical inference with the computation of probability distributions to precisely reassign the positions of reads from repeated sequences in different paired omics data, such as Hi-C data. We show that Hicberg can generate new insights into the impact of repeated elements on the spatial organisation of genomes. Significance StatementThe genomes of microorganisms can contain various types of repeated sequences: duplicated genes, low-complexity sequences and transposons. The question of their potential impact on the spatial organization of genomes is now wide open. We propose Hicberg, an algorithm capable of reconstructing contact signals from repeated elements. It computes statistical trends on the unambiguous part of the genome and then, by statistical inference, reassigns the position of multi-mapping reads. The complete chromosome contact maps thus reveal new observations on the impact of repeated elements on chromosome architecture. In particular, they suggest the involvement of certain retrotransposons in the positioning of cohesins, the molecular motors behind chromosome loops. Classification: Biophysics and Computational Biology section

genomics↗

bioRxiv2025

Chromosome-level genome assembly and annotation of the chaetognath Flaccisagitta enflata

Chaetognaths, a phylum of enigmatic marine predators, present a significant challenge to phylogenetic reconstruction due to their uncertain evolutionary placement. While transcriptome analyses have suggested affinities with the Gnathifera clade, genomic data for this group remain scarce, hindering a comprehensive understanding of their evolution. Here, we present the first chromosome-level genome assembly of Flaccisagitta enflata, a species within the Aphragmophora order. The genome assembly includes 9 chromosome candidates with a total size of 794 Mb and a BUSCO score of 91.3% against the Metazoa lineage. This high-quality genome assembly provides a crucial resource for comparative genomic analyses within Chaetognatha and the broader Gnathifera clade, and it will facilitate investigations into chaetognath evolution and their phylogenetic relationships, addressing long-standing questions regarding their placement within the animal kingdom.

genomics↗

bioRxiv2025

Replication coordination marks the domestication of large extrachromosomal replicons in bacteria

Bacterial genomes often include extrachromosomal replicons (ERs), ranging from small plasmids to nearly chromosome-sized elements, that foster genome plasticity and adaptation. Despite their prevalence, the mechanisms underlying ER domestication and their long-term adaptation within bacterial hosts remain largely unexplored. By analyzing over 40,000 complete bacterial genomes, we identified two main ER categories: small ERs with diverse GC content and large ERs ([≥]10% the size of the main chromosome) that closely match the GC content of the chromosome. Across multiple phyla, marker frequency analyses showed that large ERs maintain a 1:1 copy number with the chromosome. Another key finding of this study is that large ERs terminate replication in synchrony with the chromosome. Hi-C contact maps revealed consistent ori-ori interactions between chromosomes and ERs. In large ERs, inter-replichore and ter-ter interactions, along with the recruitment of key chromosomal segregation motifs, suggest the co-option of chromosome-associated replication and segregation machineries. Together, our findings indicate that as ERs become larger, they become increasingly reliant on chromosome-driven processes for stable inheritance, potentially explaining why they do not exceed the size of the chromosome.

microbiology↗

bioRxiv2025

The CST complex mediates a post-resection non-homologous end-joining repair pathway and promotes local deletions

Repair of a DNA double-strand break (DSB) by non-homologous end-joining (NHEJ) generally leaves an intact or minimally modified DNA sequence. Resection initiation exposes single-stranded DNA and directs repair towards homology-dependent pathways and away from NHEJ. Therefore, NHEJ is not thought to be an available repair pathway once the DSB is resected. Here, we report that the Cdc13/Stn1/Ten1 (CST) complex, well characterized for its telomere-associated functions, acts after resection initiation to mediate a backup NHEJ repair. We found a CST-specific mutation signature after DSB repair, characterized by deletions of 5-85 bp, mostly dependent on NHEJ. In contrast, NHEJ-mediated small deletions of 1-4 bp and insertions are not affected in CST mutants. The interaction between CST and Pol-primase is critical for these intermediate size deletions, suggesting a role for fill-in synthesis. Consistently, in stn1{Delta} and Pol-primase mutant deficient for interaction with CST, resection is increased, leading to larger deletions of several kilobases mediated by microhomologies. Collectively, these results depict a more complex picture of repair pathway choice where CST allows a post-resection NHEJ repair, promoting local deletions but guarding against much larger and potentially more deleterious deletions and rearrangements.

molecular biology↗

bioRxiv2024

HU promotes higher-order chromosome organisation and influences DNA replication rates in Streptococcus pneumoniae

Nucleoid-associated proteins (NAPs) are crucial for maintaining chromosomal compaction and architecture and are actively involved in DNA replication, recombination, repair, and gene regulation. In the opportunistic pathogen Streptococcus pneumoniae, HU is the only identified NAP, and its role in chromosome conformation and other essential processes has not yet been investigated. Here, we use a multi-scale approach to explore the role of HU in chromosome conformation and segregation dynamics. By combining superresolution microscopy and whole-genome binding analysis, we describe the nucleoid as a dynamic structure where HU binds transiently across the entire nucleoid, with a preference for the origin of replication over the terminus. Reducing cellular HU levels impacts nucleoid maintenance and disrupts robust nucleoid scaling with cell size. This effect is similar to the distortion caused by fluoroquinolone-antibiotics, supporting earlier observations that HU is essential for maintaining DNA supercoiling. Furthermore, in cells lacking HU, the replication machinery is misplaced, and cells are unable to initiate and proceed with on-going replication. Chromosome conformation capture (Hi-C) experiments revealed that HU is required to maintain cohesion between the two chromosomal arms, in a similar way to the structural maintenance of the chromosome complex SMC. Together, we show that by promoting long-range chromosome interactions and supporting the architecture of the domain encompassing the origin, HU is fundamental for chromosome integrity and the intimately related processes of chromosome replication and segregation. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=101 SRC="FIGDIR/small/615122v1_ufig1.gif" ALT="Figure 1"> View larger version (22K): org.highwire.dtl.DTLVardef@10cfa9eorg.highwire.dtl.DTLVardef@119ae12org.highwire.dtl.DTLVardef@f35cd4org.highwire.dtl.DTLVardef@1537e7b_HPS_FORMAT_FIGEXP M_FIG C_FIG

microbiology↗

bioRxiv2024

Micro-C reveals MORC/ApiAP2-mediated links between distant, functionally related genes in the human malaria parasite

Genome organization plays a significant role in silencing heterochromatinized genes in the most virulent human malaria parasite, Plasmodium falciparum. However, it remains unclear how heterochromatinized genes spatially cluster or if active genes are also organized in a specific manner. We used Micro-C to achieve a near-nucleosome resolution DNA-DNA contact map, which revealed new inter- and intrachromosomal heterochromatic and euchromatic structures in the blood stage parasite. We observed subtelomeric fold structures that facilitate interactions amongst heterochromatinized genes involved in antigenic variation. In addition, we identified long-range intra- and interchromosomal interactions amongst active, stage-specific genes. Both structures are mediated by AP2-P, an ApiAP2 transcription factor, and a putative MORC chromatin remodeler, and functional specificity is achieved via combinatorial binding with other sequence-specific DNA-binding factors. This study provides unprecedented insight into the organizational machinery used by this medically important eukaryotic parasite to spatially coordinate genes underlying antigenic variation and to co-activate stage-specific genes.

molecular biology↗

bioRxiv2024

Identification of a new cell cycle variant during multiciliated cell differentiation

2A complex and conserved regulatory network drives the cell cycle. Individual components of this network are sometimes used in differentiated cells, i.e. to control organelle destruction in mammalian lens cells or light response in land plants. Some differentiated cells co-opt cell-cycle regulators more largely, to increase their ploidy using a cell cycle variant named endoreplication. Using single-cell RNA-seq profiling and functional assays in differentiating multiciliated cells, we identified a novel type of cell cycle variant that supports cytoplasmic organelle, rather than nuclear content amplification. This variant operates in post-mitotic, centriole-amplifying differentiating multiciliated cells and is characterized by (i) a circular trajectory of the transcriptome, (ii) sequential expression of more than 70% of the genes involved in S, G2 and M-like progression along this trajectory, and (iii) successive waves of cyclins. This cell cycle variant is tailored by the expression of the non-canonical cyclins O and A1 - which replace the transcriptionally silent cyclins E2 and A2 - and by the silencing of the APC/C inhibitor Emi1, two switches also detected in male meiosis, another variant of the canonical cell cycle where centriole and DNA replications are uncoupled. Re-expressing Cyclin E2, cyclin A2 or Emi1 is sufficient to induce partial replication and mitosis, suggesting that change in the regulation of expression of a few cell cycle key players drives a qualitative and quantitative tuning of Cdk activity, allowing the diversion of the cell cycle in the multiciliation variant. We also propose that this new cell cycle variant relies on the existence of a cytoplasmic - or centriolar - Cdk threshold, lower than the S-phase threshold, which affects only the cytoplasmic reorganization. One-Sentence SummaryMCC progenitors undergo a final, tailored iteration of the cell cycle during differentiation, to drive centriole amplification without DNA replication or mitosis.

cell biology↗

bioRxiv2024

Cyclin O controls entry into the cell-cycle variant required for multiciliated cell differentiation

Multiciliated cells (MCC) ensure proper fluid circulation in various organs in metazoans. Their differentiation is marked by the massive ampliication of cilia-nucleating centrioles and is known to be controlled by various cell cycle components. Tn a companion study, we show that the differentiation of MCC is driven by a genuine cell-cycle variant characterized by sequential and wave-like expression of canonical and non-canonical cyclins such as Cyclin O (CCNO). Patients with CCNO mutations exhibit a subtype of Primary Ciliary Dyskinesia (PCD) designated as Reduced Generation of Multiple Motile Cilia (RGMC), yet the role of CCNO during MCC differentiation remains unclear. Here, using mice and human cellular models, single cell transcriptomics and functional studies, we show that Cena is activated during a strategic temporal window at the crossroads between the onset of MCC differentiation, the entry into the MCC cell cycle variant, and the activation of the centriole biogenesis program. We ind that the absence of Cena leads to a block of MCC progenitor differentiation at the G1/S-like transition, just before the beginning of centriole formation. This leads to a complete lack of centrioles and cilia in mouse brain and human airway MCC. Altogether, our study identifies CCNO as a core regulator of entry into the MCC cell cycle variant and shows that the coupling of centriole biogenesis to an S-like phase, maintained in MCC, is dependent on CCNO. One sentence summaryCyclin O is necessary for multiciliated cells to enter their differentiation cell cycle variant and allows the massive amplification of centrioles, which serve as basal bodies for cilia nucleation.

cell biology↗

bioRxiv2024

MatP delays the segregation of chromosome termini independently of cell division and tetramer formation in Vibrio cholerae

Vibrio cholerae harbours a primary chromosome derived from the monochromosomal ancestor of the Vibrionales (ChrI) and a secondary chromosome derived from a megaplasmid (ChrII). The coordinated segregation of the replication terminus of both chromosomes (TerI and TerII) determines when and where cell division occurs. ChrI encodes a homolog of Escherichia coli MatP, a protein that binds to a DNA motif (matS) that is overrepresented in replication termini. Here, we use a combination of deep sequencing and fluorescence microscopy techniques to show that V. cholerae MatP structures TerI and TerII into macrodomains, targets them to mid-cell during replication, and delays their segregation, thus supporting that ChrII behaves as a bona fide chromosome. We further show that the extent of the segregation delay mediated by MatP depends on the number and local density of matS sites, and is independent of its assembly into tetramers and any interaction with the divisome, in contrast to what has been previously observed in E. coli.

microbiology↗

bioRxiv2024

Epigenomics coverage data extraction and aggregation in R with tidyCoverage

SummaryThe tidyCoverage R package provides a framework for intuitive investigation of collections of genomic tracks over genomic features, relying on the principle of tidy data manipulation. It defines two data structures, CoverageExperiment and AggregatedCoverage classes, directly extending the SummarizedExperiment fundamental class, and introduces a principled approach to exploring genome-wide data. This infrastructure facilitates the extraction and manipulation of genomic coverage track data across individual or multiple sets of thousands of genomic loci. This allows the end user to rapidly visualize track coverage at individual genomic loci or aggregated coverage profiles over sets of genomic loci. tidyCoverage seamlessly combines with the existing Bioconductor ecosystem to accelerate the integration of genome-wide track data in epigenomic analysis workflows. tidyCoverage emerges as a valuable tool, contributing to the advancement of epigenomics research by promoting consistency, reproducibility, and accessibility in data analysis. Availability and implementationtidyCoverage is an R package freely available from Bioconductor [≥] 3.19 (https://www.bioconductor.org/packages/tidyCoverage) for R [≥] 4.4. The software is distributed under the MIT License and is accompanied by example files and data. Contactjacques.serizay@pasteur.fr Supplementary informationAdditional documentation is available from https://js2264.github.io/tidyCoverage/ and https://js2264.github.io/tidyCoverage/articles/tidyCoverage.html.

bioinformatics↗

bioRxiv2024

Viral proteins activate PARIS-mediated tRNA degradation and viral tRNAs rescue infection

Viruses compete with each other for limited cellular resources, and some viruses deliver defense mechanisms that protect the host from competing genetic parasites. PARIS is a defense system, often encoded in viral genomes, that is composed of a 53 kDa ABC ATPase (AriA) and a 35 kDa TOPRIM nuclease (AriB). Here we show that AriA and AriB assemble into a 425 kDa supramolecular immune complex. We use cryo-EM to determine the structure of this complex which explains how six molecules of AriA assemble into a propeller-shaped scaffold that coordinates three subunits of AriB. ATP-dependent detection of foreign proteins triggers the release of AriB, which assembles into a homodimeric nuclease that blocks infection by cleaving the host tRNALys. Phage T5 subverts PARIS immunity through expression of a tRNALys variant that prevents PARIS-mediated cleavage, and thereby restores viral infection. Collectively, these data explain how AriA functions as an ATP-dependent sensor that detects viral proteins and activates the AriB toxin. PARIS is one of an emerging set of immune systems that form macromolecular complexes for the recognition of foreign proteins, rather than foreign nucleic acids.

biochemistry↗

bioRxiv2023

Transcription promotes discrete long-range chromatin loops besides organizing cohesin-mediated DNA folding

The multi-layered arrangement of eukaryotic genomes and chromosome spatial organization dynamics are of functional importance for gene expression, DNA replication and segregation. SMC complexes are essential instruments of chromosome folding by carrying out long range intra-chromatid DNA looping. Cohesin, in addition to tether sister chromatids, also ensures dynamic regulation of gene expression in mammals by promoting interaction between distal regulatory elements and promoters whereas transcription affects genome folding in numerous organisms and in multiple ways. Here, we comprehensively dissect the relative contributions of transcription and cohesin complexes, as well as their interplay, on the yeast S. cerevisiae genome organization through DNA borders and loops. Transcription activation specifically induces appearance of DNA borders and loops, independently of SMC complexes, while also directly interfering in addition with cohesin-mediated loop expansion.

genomics↗

bioRxiv2023

Chromosome-level genome assembly of the European Green woodpecker Picus viridis

The European Green Woodpecker, Picus viridis, is a widely distributed species found in the Western Palearctic region. Here we assembled a highly contiguous genome assembly for this species using a combination of short and long reads sequencing and scaffolded with chromatin conformation capture (Hi-C). The final genome assembly was 1.28 Gb and features a scaffold N50 of 37Mb and a scaffold L50 of 39.165 Mb. The assembly incorporates 89.4% of the genes identified in birds in OrthoDB. Gene and repetitive content annotation on the assembly detected 15,805 genes and a [~]30.1% occurrence of repetitive elements, respectively. Analysis of synteny demonstrates the fragmented nature of the Picus viridis genome when compared to the chicken (Gallus gallus). The assembly and annotations produced in this study will certainly help for further research into the genomics of P. viridis and the comparative evolution of woodpeckers.

genomics↗