CENPS

CENP-S (MHF1) is a histone-fold protein that forms an obligate heterodimer with CENP-X (MHF2) and operates at the intersection of kinetochore assembly and DNA damage repair (PMID:20347429, PMID:19620631, PMID:24522885). At centromeres, CENP-S/X assemble de novo during S and G2 phase, increasing several-fold in abundance, and join CENP-T to form a centromere-bound complex positioned near histone H3 rather than CENP-A; their kinetochore localization depends on CENP-T and CENP-K (PMID:19620631, PMID:24522885). This CENP-S/X subcomplex is required for stable outer kinetochore assembly, with its loss reducing outer plate size, increasing intrakinetochore distance, and producing abnormal mitosis (PMID:19620631). In fission yeast the orthologous Mhf1-Mhf2 promote spindle assembly checkpoint signaling and proper kinetochore-microtubule attachments, and support kinetochore localization of CCAN components and Aurora B (PMID:36537249). Outside centromeres the heterodimer is excluded from the large soluble nucleoplasmic CCAN complex, indicating its centromeric assembly is spatially restricted (PMID:29509805). In its second role, CENP-S/X binds double-stranded DNA through a positively charged surface of the (MHF1-MHF2)4 octamer and enhances the branch migration activity of FANCM, stabilizing FANCM and promoting the Fanconi anemia pathway including FANCD2 monoubiquitination and chromatin loading of the FA core complex (PMID:20347429, PMID:23886707). Live-cell microirradiation places CENP-S/X recruitment to double-strand breaks immediately after ATM and RNF8-RNF168 activity, with removal coinciding with RPA loading and RAD51 assembly, positioning the complex in early nucleosome remodeling that precedes resection (PMID:40450933). Genetic dissection in fission yeast shows the DNA repair and centromere functions are separable, the former dependent on the FANCM ortholog Fml1 and the latter independent of it (PMID:24026537).

1. 2009 High

   It was unknown which factors stabilize the outer kinetochore; this established CENP-S/X as an essential CCAN subcomplex for stable outer plate assembly.

   Evidence Gene knockout in chicken DT40 cells and siRNA depletion in HeLa cells with EM, live imaging, and CCAN localization epistasis

   PMID:19620631

   Open questions at the time

   • Did not resolve the molecular structure of the CENP-S/X-containing centromeric complex
   • Mechanism linking outer plate size to chromosome segregation fidelity not defined
2. 2010 High

   The molecular partner and biochemical activity of CENP-S beyond the kinetochore were unknown; this identified MHF1(CENP-S)-MHF2 as a DNA-binding heterodimer that stimulates FANCM branch migration and supports the Fanconi anemia pathway.

   Evidence Co-IP, biochemical reconstitution, DNA binding and branch migration assays, siRNA knockdown with FANCD2 foci and chromosome aberration readouts

   PMID:20347429

   Open questions at the time

   • Structural basis of DNA binding not yet defined
   • Did not address how the same protein partitions between centromere and repair roles
3. 2013 Medium

   Whether the kinetochore and repair functions are mechanistically coupled was unclear; ortholog genetics separated a Fml1(FANCM)-dependent repair role from a Fml1-independent centromere/segregation role.

   Evidence Fission yeast genetics, epistasis, fluorescence microscopy, and co-IP

   PMID:24026537

   Open questions at the time

   • Separability shown genetically in yeast, not biochemically demonstrated in human cells
   • Whether the same molecular pool serves both functions unresolved
4. 2013 Medium

   The structural mechanism of chromatin anchoring was unknown; SAXS and crystallography revealed an (MHF1-MHF2)4 octamer with a positively charged patch that mediates dsDNA binding.

   Evidence SAXS combined with crystallography and biochemical DNA binding assays

   PMID:23886707

   Open questions at the time

   • Limited functional mutagenesis to validate the DNA-binding patch in cells
   • Octamer relevance to centromeric assembly versus repair not distinguished
5. 2014 Medium

   The timing, stoichiometry, and partners of centromeric assembly were undefined; biophysics in live cells showed S/G2-phase de novo assembly, that CENP-S/X exist principally as a complex, and that CENP-T is a strong binding partner near H3.

   Evidence FCCS, FRET, FRAP, conditional labeling, and fluorescent two-hybrid in live human cells

   PMID:24522885

   Open questions at the time

   • Differential FRAP exchange of CENP-X versus CENP-S not mechanistically explained
   • Relationship of the distinct DNA-damage-site binding kinetics to the repair function not directly linked
6. 2018 Medium

   Whether CENP-S/X belongs to the soluble nucleoplasmic CCAN was unknown; FCCS showed they are excluded from the large soluble CENP-C/H/I/K/M/T/W/N/L complex outside centromeres.

   Evidence FCCS in living human cells with dissociation constant measurement

   PMID:29509805

   Open questions at the time

   • Did not define what restricts CENP-S/X assembly to centromeres
   • Functional consequence of exclusion not tested
7. 2023 Medium

   The role of CENP-S/X in checkpoint signaling was unclear; yeast deletion showed Mhf1-Mhf2 promote the spindle assembly checkpoint and govern kinetochore localization of CCAN and Aurora B.

   Evidence Live-cell microscopy, deletion mutant genetics, immunofluorescence, and CCAN localization epistasis in fission yeast

   PMID:36537249

   Open questions at the time

   • SAC role demonstrated in yeast, not confirmed in human cells
   • Mechanism by which it controls Aurora B localization unresolved
8. 2025 Medium

   The temporal placement of CENP-S/X within the DSB response was unknown; microirradiation kinetics positioned their recruitment immediately after ATM and RNF8-RNF168 activity and their removal at RPA/RAD51 loading.

   Evidence Live-cell laser microirradiation with cell-cycle enrichment and kinetics calibrated to published DDR timelines in HeLa cells

   PMID:40450933

   Open questions at the time

   • Molecular function at the early DSB step (e.g. nucleosome remodeling versus pathway choice) not biochemically defined
   • Direct interactors at the break site not identified

• How a single CENP-S/X heterodimer is partitioned between stable centromeric assembly and transient DSB recruitment, and what governs this switch in human cells, remains unresolved.
• No mechanism defining the molecular switch between centromere and repair pools
• Human-cell evidence for the SAC role absent
• Direct structural model of CENP-S/X within the kinetochore not established