Florian Roisné-Hamelin

Telomere biology is often framed as a uniquely eukaryotic problem, yet the challenges of maintaining DNA ends are shared by all organisms harboring linear DNA, including viral entities. Telomeres enable complete replication of chromosome ends while shielding the termini from DNA repair factors and innate immune surveillance systems that would otherwise recognize them as damaged or foreign DNA. A particular threat is posed by Mre11‐Rad50, a nuclease complex conserved from bacteria to humans that specializes in the detection and processing of DNA ends. The solutions to these DNA end‐related challenges that have emerged across the tree of life converge on a small number of recurring strategies. Here, we review how diverse biological systems, from bacteriophages to mammals, have solved the DNA end‐related challenges, with a particular focus on recent advances in our understanding of end protection.

Telomere capping largely depends on telomere length. Abnormally short telomeres are prone to activate DNA damage checkpoint, undergo unscheduled chromosomal fusions through nonhomologous end-joining (NHEJ) and be resected. All these processes are mediated by the Mre11–Rad50–Xrs2^NBS1 (MRX^MRN) complex. The response to telomere length is thought to correlate with the number of DNA-bound telomeric proteins but the mechanisms translating this number into functional protection remain unclear. Here, combining genetic and reconstitution biochemistry approaches, we explain why NHEJ inhibition and DNA resection suppression are proportional to the length of Rap1-bound arrays adjacent to DNA ends in Saccharomyces cerevisiae. Our findings reveal that MRX has a capacity to distinguish short from long protein blocks at DNA breaks, which translates into the ability of Rap1 protein–DNA arrays to suppress MRX on the basis of their length. Unlike the previously described direct inhibition of MRX^MRN by Rif2 and TRF2 iDDR, the process described here does not rely on direct protein–protein interactions. Instead, Rap1 inhibits MRX by steric hindrance, determined both by the length of DNA it shields and by the coverage density. These results explain how cells detect short telomeres, a crucial process for telomere length homeostasis and telomere-driven senescence.

Le complexe Mre11-Rad50-Xrs2/NBS1 (MRX/N) joue un rôle fondamental dans le maintien de la stabilité du génome et dans la réparation des cassures double-brin (CDB). Il est essentiel au NHEJ (Non-Homologous End-Joining), et critique pour l’initiation de la résection 5’, première étape de la recombinaison homologue. MRX est aussi impliqué dans la maintenance des extrémités des chromosomes. MRX favorise l’action de la télomérase préférentiellement aux télomères les plus courts via l’activation régulée de la kinase Tel1/ATM. L’action de MRX aux télomères doit être finement contrôlée pour éviter une fusion ou une dégradation des télomères. Ce contrôle est assuré par un petit nombre de facteurs spécifiquement enrichis aux extrémités des chromosomes. Chez la levure Saccharomyces cerevisiae, une protéine, Rap1, lie directement les répétitions télomériques et recrute un éventail de protéines assurant les différentes fonctions des télomères. Parmi ces protéines, Rif2 est impliquée dans l’inhibition de plusieurs fonctions qui dépendent de MRX (inhibition du NHEJ, de la résection 5’ et de l’élongation des télomères par la télomérase). Nous avons établi qu’un petit domaine N-terminal de Rif2, connu pour bloquer l’élongation des télomères, était aussi capable d’inhiber le NHEJ, la résection et le chargement du complexe MRX sur l’ADN. Nous avons identifié une interaction entre les têtes ATPase de Rad50 et ce motif N-terminal de Rif2. De plus, nous avons montré que ce motif de Rif2 agit à courte distance en cis. Par des approches de prédictions structurales couplées à un crible génétique, nous avons identifié des résidus de Rad50 essentiels pour l’interaction avec Rif2 et sa capacité d’inhibition du complexe. La position de ces résidus sur une surface conservée de Rad50 nous a permis de proposer un modèle pour l’inhibition du complexe MRX aux télomères par Rif2, où Rif2 bloque la conformation de MRX permettant l’association stable du complexe sur l’ADN. En parallèle, nous avons étudié le rôle de Rap1 dans la protection des télomères en l’absence de ses partenaires impliqués dans le blocage du NHEJ. Nous avons établi que la force de cette inhibition du NHEJ par Rap1 était étroitement liée à la densité et au nombre de sites fixés par Rap1, et de la proximité de Rap1 de l’extrémité d’ADN.

DNA loop-extruding SMC complexes play vital roles in genome maintenance and DNA immunity. However, how these ring-shaped DNA motors navigate large DNA-bound obstacles has remained unclear. Here, we demonstrate that a bacterial SMC Wadjet complex can efficiently bypass obstacles larger than the SMC coiled coil lumen when they are tethered to the extruded DNA by a single-stranded DNA or RNA linker. This bypass is mediated by the selective entrapment of the linker within the SMC hinge channel, which functions as an obligate gate—permitting passage of the linker while retaining double-stranded DNA stably inside the SMC ring and keeping associated obstacles outside. We further show that eukaryotic SMC hinges similarly accommodate ssDNA, and that the hinge is dispensable for loop extrusion by Wadjet, altogether suggesting that obstacle bypass represents the conserved, long-sought function of the SMC hinge toroid. By integrating hinge bypass with loop extrusion, we provide a mechanistic framework for how DNA-entrapping SMC complexes can generate chromosomal loops densely decorated with obstacles.

Safeguarding chromosome ends against fusions via nonhomologous end joining (NHEJ) is essential for genome integrity. Paradoxically, the conserved NHEJ core factor Ku binds telomere ends. How it is prevented from promoting NHEJ remains unclear, as does the mechanism that allows Ku to coexist with telomere-protective DNA binding proteins, Rap1 in Saccharomyces cerevisiae. Here, we find that Rap1 directly inhibits Ku’s NHEJ function at telomeres. A single Rap1 molecule near a double-stand break suppresses NHEJ without displacing Ku in cells. Furthermore, Rap1 and Ku form a complex on short DNA duplexes in vitro. Cryo-EM shows Rap1 blocks Ku’s inward translocation on DNA – an essential step for NHEJ at DSBs. Nanopore sequencing of telomere fusions confirms this mechanism protects native telomere ends. These findings uncover a telomere protection mechanism where Rap1 restricts Ku’s inward translocation. This switches Ku from a repair-promoting to a protective role preventing NHEJ at telomeres.

DNA loop-extruding SMC complexes play vital roles in genome maintenance and DNA immunity. However, the ability of these ring-shaped DNA motors to navigate large DNA-bound obstacles, such as DNA/RNA polymerases and ribosomes in prokaryotes, has remained unclear. Here, we demonstrate that a bacterial SMC Wadjet complex can efficiently bypass obstacles larger than the SMC coiled coil lumen if they are linked to the extruded DNA by single-stranded DNA (ssDNA). This process is mediated by the selective entrapment of the ssDNA linker within the SMC hinge channel, which acts as an obligate gate—permitting ssDNA passage while excluding double-stranded DNA (dsDNA). By threading the linker through the hinge gate as dsDNA is translocated through the SMC motor, the obstacle can follow a trajectory distinct from its associated dsDNA, remaining outside the SMC ring. This mechanism enables continuous dsDNA entrapment while successfully bypassing obstacles. We show that hinges of diverse SMC complexes including eukaryotic condensin and Smc5/6 also selectively accommodate ssDNA, suggesting that hinge bypass is a conserved feature across SMC proteins. We demonstrate that while the hinge domain is essential for obstacle bypass, it is dispensable for loop extrusion, showing that these functions are independent, provided by separate SMC modules. These findings uncover the long-sought function of the SMC hinge toroid and provide the first mechanistic explanation for how DNA-entrapping SMC complexes can extrude loops on chromosomal DNA densely populated with obstacles. Integrating hinge bypass into the DNA loop extrusion model, we propose that the hinge may also facilitate the passage of RNA and peptide, enabling bypass of transcription-associated ribonucleoprotein complexes and other SMC complexes.

Summary: Structural Maintenance of Chromosomes proteins are essential DNA-folding motors that facilitate critical cellular functions, including chromosome segregation and DNA repair. Wadjet systems are prokaryotic SMC complexes specialized in cellular immunity against plasmids. Type I Wadjet systems restrict plasmids via a DNA extrusion-cleavage reaction. Two other Wadjet types (II and III) have also been identified, however, their molecular characteristics are unclear. Here, we reconstituted a representative type II Wadjet system from Neobacillus vireti . We show that this system shares substrate selection and cleavage properties with type I but exhibits distinctive structural features, including a long elbow-distal coiled coil, a channel-less hinge, and a tandem KITE subunit. These features help identify the common and distinguishing architectural elements in the family of Wadjet systems and raise intriguing questions about the evolution of prokaryotic SMC complexes.

Multisubunit protein complexes are central to many cellular processes, and studying their activities and structures in vitro requires reconstitution via recombinant expression and purification. Obtaining targets at sufficient purity and scale typically involves screening several protein variants and expression hosts. Existing cloning strategies enable co-expression but are often time-consuming, labor-intensive, and host-specific, or involve error-prone steps. We present a novel vector set and assembly strategy to overcome these limitations, enabling expression construct generation for multisubunit complexes in a single step. This modular system can be extended to additional hosts or include new tags. We demonstrate its utility by constructing expression vectors for structural maintenance of chromosomes complexes in various hosts, streamlining workflows, and improving productivity.

Multi-subunit protein complexes are at the heart of many cellular processes, and studying their biochemical activities and structures in vitro requires their reconstitution by recombinant expression and purification. Obtaining targets at sufficient purity and scale typically requires the screening of several protein variants and expression hosts. Existing cloning strategies allow to produce constructs for co-expression of proteins, but are often time-consuming, labour-intensive, host-specific, or involving error-prone assembly steps. Here we present a unique set of vectors together with a novel assembly strategy designed to overcome these limitations. It allows for the generation of expression constructs for multi-subunit protein complexes for various hosts in a single cloning step. Its modular nature allows the system to be easily extended to target additional expression hosts or to include new tags or regulatory sequences. As a proof of principle, we present the parallel construction of expression vectors for several Structural Maintenance of Chromosomes (SMC) complexes, allowing us to devise strategies for the recombinant production of these targets in bacteria, insect cells, and human cells, respectively. This work will help laboratories working on protein complexes to streamline their workflow, increase their productivity and improve the quality of the purified material.

Summary: DNA loop-extruding SMC complexes play crucial roles in chromosome folding and DNA immunity. Prokaryotic SMC Wadjet (JET) complexes limit the spread of circular plasmids through DNA cleavage; yet the mechanisms for target recognition are unresolved. We show that artificial DNA circularization renders linear DNA susceptible to JET cleavage. Unlike free DNA, JET cleaves immobilized plasmid DNA at a specific site, the plasmid-anchoring point, showing that the anchor hinders DNA extrusion but not DNA cleavage implying that residual unextruded DNA is cleaved. Structures of plasmid-bound JetABC reveal two presumably stalled SMC motor units that are drastically rearranged from the resting state, together entrapping a U-shaped DNA segment, which is further converted to kinked V-shaped cleavage substrate by JetD nuclease binding. Our findings uncover mechanical bending of residual unextruded DNA as principle for non-self DNA recognition and molecular signature for plasmid cleavage. We elucidate key elements of SMC loop extrusion including motor directionality and the structure of a DNA-holding state.