Centromere

A centromere is a specialized region of a chromosome that provides the foundation for assembly of the kinetochore, the protein structure that connects chromosomes to spindle microtubules during cell division. Centromeres thereby enable genetic material to be distributed to daughter cells. They are distinct from both telomeres, which protect chromosome ends, and centrosomes, which organize microtubules in many animal cells. Although commonly depicted as a chromosome’s narrow primary constriction, centromeric activity can instead be distributed along much of a chromosome’s length. (elifesciences.org)

Centromeres combine specialized DNA, chromatin, and associated proteins, but their identity is not determined by a universal DNA sequence. In many eukaryotes, nucleosomes containing the histone variant CENP-A provide an inherited epigenetic foundation for centromere maintenance and kinetochore assembly; some lineages use alternative molecular arrangements. Their essential role in chromosome inheritance, coupled with the rapid evolution of their underlying sequences, makes centromeres important subjects in molecular biology, genetics, and evolutionary research. This article examines their chromosome-level organization, molecular composition, inheritance, segregation functions, evolutionary dynamics, experimental investigation, and consequences of dysfunction. (nature.com)

Chromosomal location and cytological identification

A centromere is a functional domain of a chromosome that specifies where the kinetochore assembles. The distinction is between a chromosomal substrate—specialized chromatin—and the protein apparatus built upon it: spindle microtubules bind to the kinetochore, not directly to centromeric DNA. In chromosomes with a single localized centromere, the kinetochore-forming domain lies within a broader cytologically recognizable region. The surrounding pericentromere comprises neighboring chromatin, commonly including heterochromatin and regions involved in sister-chromatid cohesion; its boundaries need not coincide precisely with those of the visible constriction. Neither structure should be confused with telomeres, which protect the ends of linear chromosomes, or with the centrosome, a cellular microtubule-organizing structure rather than a chromosomal domain. (pmc.ncbi.nlm.nih.gov)

In condensed monocentric chromosomes, particularly those examined at metaphase of mitosis, the centromeric region usually appears as the primary constriction, a narrowed zone distinguishing the chromosome’s two arms. By cytogenetic convention, the shorter arm is designated p and the longer arm q. These labels describe positions relative to the centromere, not separate DNA molecules. A replicated metaphase chromosome contains two sister chromatids, each bearing its own centromeric domain and kinetochore; their close association in this region contributes to the familiar constricted appearance. Despite its name, a centromere need not occupy the chromosome’s midpoint and may lie near an end. Moreover, the light-microscopic outline of the primary constriction does not resolve the smaller molecular domain upon which the kinetochore assembles. (genome.gov)

Centromere position provides a morphological classification of monocentric chromosomes. Metacentric chromosomes have approximately equal arms; submetacentric chromosomes have an off-center centromere and appreciably unequal arms; acrocentric chromosomes have a centromere near one end, leaving a very short p arm; and telocentric chromosomes have an effectively terminal centromere and only one readily distinguishable arm. Albert Levan, Karl Fredga, and Avery A. Sandberg’s 1964 nomenclature study formalized descriptions of centromeric position using chromosome-arm measurements. These are morphological categories, not classifications of centromeric DNA sequence or protein composition. Human autosomes 13, 14, 15, 21, and 22 provide familiar acrocentric examples: their short arms remain substantial genomic regions despite appearing small in condensed chromosome preparations. (onlinelibrary.wiley.com)

Identification by cytogenetics combines chromosome shape with staining landmarks. A primary constriction supplies positional evidence, whereas chromosome banding supplies a reproducible pattern by which the chromosome and its surrounding segments can be recognized. G-, Q-, and R-banding distinguish landmarks along chromosome arms; C-banding preferentially highlights constitutive heterochromatin, often concentrated around centromeres. A dark C-band therefore identifies a chromatin compartment associated with many centromeric regions, rather than directly demonstrating a functional kinetochore. Banding and constriction morphology provide complementary anatomical evidence, but neither alone establishes the molecular boundaries or activity of the kinetochore-forming domain. (pubmed.ncbi.nlm.nih.gov)

Localization of centromere-associated proteins adds evidence about functional organization. In human chromosome preparations, immunofluorescence detection of CENP-A or CENP-C can mark active centromeric sites, including sites whose interpretation from morphology or DNA staining is uncertain. Marker choice matters: in a 1989 study, William C. Earnshaw, H. Ratrie III, and G. Stetten examined a stable chromosome carrying two centromeric regions and found CENP-B at both the active and inactive sites, whereas CENP-C was undetectable at the inactive site. A later experimental protocol combined CENP-C immunofluorescence with chromosome-specific DNA hybridization to identify active centromeres in fixed cells. Such observations distinguish the presence of centromere-associated DNA or proteins from assembly of an active centromeric apparatus. (pubmed.ncbi.nlm.nih.gov)

Point, regional, and holocentric organization

Point centromeres are specified by short, recognizable DNA sequences. In the budding yeast Saccharomyces cerevisiae, each of the 16 chromosomes contains an approximately 125-base-pair centromere comprising three centromere DNA elements: CDEI, CDEII, and CDEIII. CDEI is an 8-base-pair recognition site for Cbf1; CDEII is a 78–86-base-pair, strongly adenine- and thymine-rich segment; and CDEIII is an approximately 25–26-base-pair conserved element bound by the four-subunit CBF3 complex. Sequence recognition provides an entry point for kinetochore assembly: the Cep3 subunit of CBF3 contacts the essential CCG motif in CDEIII, and mutations in this motif can abolish centromere activity. CDEII contributes to the chromatin environment containing Cse4, the yeast centromeric histone H3 variant. Thus, “point” denotes a compact, sequence-defined domain, not an attachment site lacking chromatin or associated proteins. (pmc.ncbi.nlm.nih.gov)

Regional centromeres extend over much larger domains but lack a universally conserved DNA sequence that specifies their identity. In the fission yeast Schizosaccharomyces pombe, the three centromeric regions encompass approximately 35–110 kilobases, including flanking repetitive DNA. Their central domains carry Cnp1, the fission-yeast counterpart of CENP-A/CenH3, whereas outer repeats generally form heterochromatin. The central core and the complete repeat-containing region are therefore different units of measurement: the 2002 genome study described a 4.1-kilobase central core within the approximately 35-kilobase centromere 1. Experiments published in 2019 showed that central-domain DNA from S. octosporus and S. cryophilus could support functional centromeres in S. pombe despite lacking sequence homology with its centromeres. Mammalian centromeres likewise occupy extended chromatin domains, commonly within large satellite arrays; mapping of human centromeres in 2022 distinguished kinetochore-associated CENP-A domains from the larger surrounding repeat arrays. These observations separate regional organization from any requirement for one conserved nucleotide motif. (nature.com)

In holocentric chromosomes, centromeric activity is distributed along much of the chromosome rather than confined to one localized region. In the nematode Caenorhabditis elegans, genome-wide mapping by Reto Gassmann and colleagues in 2012 detected low-density CENP-A incorporation in domains cumulatively covering about half the embryonic genome. Distributed activity therefore does not imply that every base pair is centromeric or that CENP-A forms an uninterrupted coating. Holocentricity also differs from dicentric organization: a dicentric chromosome has two discrete centromeric sites, whereas a holocentric chromosome has a longitudinally distributed system. Butterflies and moths, and many hemipteran insects, provide additional well-established examples of holocentric organization. (nature.com)

Holocentric plants demonstrate that distributed activity can have different underlying DNA architectures. In Rhynchospora beaksedges, studied chromosomes contain hundreds of short, approximately 20-kilobase arrays of the Tyba satellite repeat, interspersed along their length and associated with CENH3. These separate units collectively constitute a holocentromere rather than one continuous satellite block. Woodrushes of the genus Luzula provide further variation: research published in 2024 found that L. sylvatica CENH3 chromatin was predominantly associated with the Lusy1 and Lusy2 satellite repeats, but also occupied satellite-free, gene-poor regions. Consequently, neither repeat enrichment nor repeat independence defines all holocentric chromosomes, even among plants. (nature.com)

The molecular machinery underlying these organizational classes is not universal. C. elegans and the holocentric plants described above retain CENP-A/CenH3, but Ines Drinnenberg, Dakota deYoung, Steven Henikoff, and Harmit Malik reported a different pattern in 2014. Their comparative genomic and transcriptomic analysis inferred at least four independent losses of CenH3 in insect lineages associated with independent transitions to holocentricity. The sampled holocentric insects retained numerous kinetochore components, particularly proteins involved in spindle attachment, despite lacking identifiable CenH3. These findings establish that chromosome-wide centromeric organization can operate without this histone variant; they do not establish one shared replacement mechanism for all such insects or show that CenH3 loss alone causes holocentricity. (elifesciences.org)

Centromeric DNA and repetitive-sequence architecture

Human centromeric DNA is dominated by alpha-satellite DNA, a family of tandem repeats whose basic units, or monomers, are approximately 171 base pairs long. Monomers differing in sequence are arranged in an ordered series to form a higher-order repeat (HOR); repeated copies of that composite unit then build arrays extending over megabases. Chromosome-associated HORs differ in their monomer sequences, number, and order, so alpha satellite is not an identical sequence shared by every centromere. For example, the DXZ1 array of the human X chromosome has a canonical HOR comprising 12 monomers, approximately 2 kilobases in total. Adjacent monomers within a HOR can be substantially divergent, while successive copies of the complete HOR are much more similar. This nested organization distinguishes an alpha-satellite array from a simple repetition of one invariant 171-base-pair sequence. (pmc.ncbi.nlm.nih.gov)

Some alpha-satellite monomers contain a 17-base-pair CENP-B box, recognized directly by the sequence-specific DNA-binding protein CENP-B. This provides a molecular connection between repeat sequence and centromere-associated proteins: experiments reported by Daniele Fachinetti and colleagues in 2015 showed that DNA-bound CENP-B stabilizes centromeric CENP-C and contributes to chromosome-segregation fidelity. Nevertheless, CENP-B binding is not an obligatory specification mechanism. The human Y centromere lacks CENP-B boxes, and functional human neocentromeres can form on DNA lacking alpha satellite and bound CENP-B. Conversely, CENP-B can remain associated with an inactive alpha-satellite centromere. Its binding therefore neither universally defines nor, by itself, establishes centromeric activity. (pmc.ncbi.nlm.nih.gov)

An entire satellite array must also be distinguished from the active centromeric subdomain within it. In their 2022 analysis of the T2T-CHM13 human genome, Nicolas Altemose and colleagues mapped the centromeric histone variant CENP-A to one active alpha-satellite HOR array per chromosome. Windows of strong CENP-A enrichment spanned 190–570 kilobases, occupying only 7–24% of their respective active arrays. These windows identify protein-enriched chromatin domains, not the full extent of centromeric repetitive DNA. Strong enrichment occurred near and within locally hypomethylated regions termed centromere dip regions, but extended beyond them. The reported dimensions apply to the CHM13 sample and the mapping criteria used: the authors explicitly did not exclude weaker CENP-A occupancy outside those windows or variation among cells and cell types. They are therefore not a universal size range for human centromeres. (pmc.ncbi.nlm.nih.gov)

Other organisms use different repeat architectures. Rice centromeres contain CentO satellite monomers, commonly 155 base pairs long, interspersed with centromere-associated retrotransposons. Sequencing of the rice chromosome 8 centromeric region in 2004 identified 68.5 kilobases of CentO repeats in three clusters within a 1.97-megabase interval, together with more than 220 transposable-element-related sequences. Satellite clusters were interrupted by these elements and differed in orientation. In maize, the approximately 156-base-pair CentC satellite and CRM retrotransposons are major centromeric components; CENH3-associated DNA includes both satellite and retrotransposon-derived sequences. Thus, tandem satellite abundance is only one aspect of centromeric sequence composition. (pubmed.ncbi.nlm.nih.gov)

Centromeric activity can also occur on nonrepetitive DNA. The 2009 horse genome study by C. M. Wade and colleagues identified a functional centromere on chromosome 11 without centromeric satellite DNA. Subsequent mapping exploited this satellite-free sequence to locate its CENP-A-associated domain. Together with inactive human alpha-satellite loci that retain CENP-B binding, such cases demonstrate why repeat annotation alone cannot identify a functional centromere: sequence composition must be distinguished from evidence of centromeric chromatin and kinetochore assembly. (pmc.ncbi.nlm.nih.gov)

Centromeric chromatin and kinetochore assembly

In humans and many other eukaryotes, centromeric chromatin contains CENP-A, a centromere-associated variant of histone H3 also termed CenH3. CENP-A replaces H3 in a subset of centromeric nucleosomes, rather than throughout the entire domain; canonical H3-containing nucleosomes remain interspersed with CENP-A-containing particles. Beth Sullivan and Gary Karpen documented this mixed organization in human and Drosophila centromeric chromatin in 2004. In 2011, Hiroaki Tachiwana and colleagues determined a crystal structure of a reconstituted human CENP-A nucleosome containing two copies each of CENP-A, H4, H2A, and H2B. Its DNA forms a left-handed superhelix around the histone octamer, but the DNA entrance and exit regions are more flexible than in conventional H3 nucleosomes. These findings establish CENP-A as a specialized chromatin component, not a separate spindle-binding structure. (pmc.ncbi.nlm.nih.gov)

Two inner-kinetochore proteins, CENP-C and CENP-N, recognize distinctive surfaces of CENP-A nucleosomes. Hidenori Kato and colleagues showed in 2013 that CENP-C contacts the hydrophobic C-terminal tail of CENP-A while docking onto the acidic patch formed by histones H2A and H2B. CENP-N uses a different interface: its N-terminal region recognizes the exposed L1 loop of CENP-A and also contacts adjacent nucleosomal DNA. A 2018 cryo-electron microscopy study resolved a human CENP-N–CENP-A nucleosome complex at 3.9 Å, supporting this combination of histone-specific recognition and DNA-mediated stabilization. CENP-N’s C-terminal region associates with CENP-L, connecting nucleosome recognition to the wider inner kinetochore. These interactions distinguish recognition of a chromatin particle from recognition of a particular DNA sequence. (pubmed.ncbi.nlm.nih.gov)

The human constitutive centromere-associated network (CCAN) comprises 16 proteins organized into interacting modules, including CENP-C, CENP-LN, CENP-HIKM, CENP-TWSX, and CENP-OPQUR. This assembly provides the foundation of the inner kinetochore. CENP-C acts as a scaffold whose separate binding regions engage centromeric nucleosomes, other CCAN components, and outer-kinetochore proteins. CCAN attachment is not mediated exclusively through histone contacts: structural studies in 2022 demonstrated extensive interactions with DNA, including a channel formed by CENP-LN and neighboring subunits that accommodates extranucleosomal DNA. Consequently, the isolated CENP-N–nucleosome interface does not by itself describe every arrangement of CENP-N within assembled CCAN; protein–protein contacts and DNA geometry also determine inner-kinetochore organization. (pmc.ncbi.nlm.nih.gov)

Connections to the outer kinetochore principally involve CENP-C and CENP-T. CENP-C directly binds the MIS12 complex, which links it to the KMN network—KNL1, the MIS12 complex, and the NDC80 complex. CENP-T supplies a parallel, phosphorylation-regulated recruitment pathway. In biochemical reconstitution experiments reported in 2016, phosphorylation by CDK1–cyclin B enabled one CENP-T molecule to recruit two NDC80 complexes directly and another through MIS12; these stoichiometries describe the reconstructed assembly, not an invariant count for an entire kinetochore. NDC80 supplies a major microtubule-binding interface: the human four-subunit complex contains NDC80/HEC1, NUF2, SPC24, and SPC25, with its microtubule-binding and kinetochore-anchoring regions positioned at opposite ends of an elongated structure. (sciencedirect.com)

Centromeric DNA therefore serves as the substrate on which nucleosomes and DNA-binding kinetochore components assemble, whereas CENP-A provides a recognizable histone surface, CCAN proteins organize and anchor the inner kinetochore, and outer-kinetochore complexes establish contact with the mitotic spindle. The mechanical connection runs through these protein assemblies rather than through direct attachment of spindle microtubules to centromeric DNA. DNA-binding activity, nucleosome recognition, scaffolding, and microtubule binding are distinct molecular contributions, even when an individual protein participates in more than one of them. (pmc.ncbi.nlm.nih.gov)

Establishment and epigenetic inheritance

In human cells and many other eukaryotes with regional centromeres, centromeric identity is propagated through chromatin rather than specified anew by a unique DNA sequence at every division. Pre-existing CENP-A nucleosomes provide a positional memory that directs the replenishment of centromeric chromatin. In 2013, Daniele Fachinetti and colleagues used gene targeting and histone-replacement experiments in human cells and fission yeast to distinguish this memory from kinetochore assembly. Histone H3 carrying the centromere-targeting domain of CENP-A could maintain centromeric positioning and support continued chromatin propagation; additional CENP-A terminal regions were needed to recruit proteins supporting kinetochore function. These experiments identified separable molecular requirements for inheriting a centromeric site and making that site competent for chromosome segregation. This is epigenetic inheritance, but it does not imply that local DNA sequence and chromatin context have no influence. (nature.com)

During DNA replication, parental CENP-A is retained and redistributed between the two sister centromeres rather than immediately supplemented with an equivalent amount of newly synthesized CENP-A. Lars Jansen and colleagues demonstrated this in 2007 using SNAP-tag pulse–chase labeling, which distinguished previously incorporated CENP-A from newly synthesized molecules. The parental pool was approximately equally partitioned during S phase, leaving each replicated centromere with about half its preceding CENP-A complement. New CENP-A incorporation followed passage through mitosis, principally during early G1 in the human cells examined. Replication therefore temporarily dilutes the inherited mark without erasing its location. A 2018 study further showed that HJURP participates in retaining parental CENP-A during S phase and interacts with the replication-associated MCM2 protein, revealing a recycling function distinct from its role in depositing new CENP-A. (pubmed.ncbi.nlm.nih.gov)

Replenishment requires the CENP-A-specific histone chaperone HJURP, which binds soluble CENP-A–H4 and facilitates its assembly into nucleosomes. Human centromeres recruit this chaperone through the Mis18 machinery, comprising Mis18α, Mis18β and M18BP1, also called KNL2. Depletion experiments reported in 2007 showed that loss of any of these three components abolished efficient recruitment of newly synthesized CENP-A. Biochemical reconstitution subsequently established an assembly containing four Mis18α, two Mis18β and two M18BP1 molecules. HJURP binds directly to this machinery through its centromere-targeting repeats, connecting a licensed chromosomal site to the incoming CENP-A–H4 cargo. Thus, HJURP supplies histone specificity and assembly activity, whereas Mis18 and M18BP1 help determine where that activity occurs. (pubmed.ncbi.nlm.nih.gov)

A feedback pathway links existing centromeric chromatin to its replenishment: CENP-A chromatin recruits CENP-C, and CENP-C facilitates M18BP1 recruitment, connecting the inherited mark to machinery that deposits more CENP-A nearby. Temporal regulation prevents this positive feedback from operating continuously. CDK-dependent phosphorylation regulates Mis18-complex assembly and HJURP interactions; declining CDK activity at mitotic exit permits productive recruitment and loading. The early-G1 schedule is a human-cell example, not a universal property of centromeres: fission yeast replenishes its CENP-A, Cnp1, during G2, with recruitment mechanisms adapted to its nuclear organization. Spatial feedback and cell-cycle control therefore act together to reproduce centromeric chromatin at an established site. (pmc.ncbi.nlm.nih.gov)

Maintenance of an existing domain is better characterized than the spontaneous initiation of a new one. In 2011, experiments tethering HJURP to a noncentromeric LacO DNA array induced stable CENP-A incorporation, recruitment of kinetochore proteins and microtubule attachments, demonstrating that targeted delivery of an assembly factor can initiate centromeric activity without conventional centromeric DNA. Such engineered recruitment bypasses the unresolved step of selecting a site naturally. Likewise, neocentromeres demonstrate that alternative chromosomal locations can sustain centromere function, but do not by themselves explain how the first self-maintaining CENP-A domain arises. Questions remain about the initial recruitment signal, the minimum effective chromatin seed, and the local conditions that allow a nascent domain to persist rather than disappear. (pubmed.ncbi.nlm.nih.gov)

Pericentromeric organization and chromosome segregation

Pericentromeric chromatin is enriched in cohesin, a protein complex that maintains sister-chromatid association after replication. This association supplies resistance to the opposing forces exerted through sister kinetochores by spindle microtubules: before anaphase, pulling can stretch the intervening chromatin without separating the sisters completely. Pericentromeric architecture also organizes this mechanical response. In Saccharomyces cerevisiae, a 2020 study showed that centromeres promote cohesin loading, while convergently transcribed genes trap cohesin at pericentromere boundaries. Reorienting these boundary genes enlarged the pericentromeric domain and impaired chromosome biorientation, providing experimental evidence that chromatin organization—not simply the presence of a spindle attachment site—contributes to segregation. (pmc.ncbi.nlm.nih.gov)

In fission yeast, Schizosaccharomyces pombe, pericentromeric heterochromatin connects histone modification to cohesion. The methyltransferase Clr4 methylates lysine 9 of histone H3, producing H3K9 methylation recognized by the chromodomain of Swi6, an HP1-family protein. Andrew Bannister and colleagues demonstrated this methylation-dependent recognition in 2001. Nobuhiro Nonaka and colleagues subsequently reported in 2002 that Swi6 is required for preferential cohesin accumulation at centromeric repeats and interacts with the cohesin subunit Psc3. These experiments established a molecular link between heterochromatin and cohesin enrichment rather than attributing cohesion directly to repetitive DNA. (nature.com)

Heterochromatin also contributes to recruiting cohesion-protection machinery. In 2008, Yuya Yamagishi and colleagues showed that Swi6 associates directly with the meiotic shugoshin protein Sgo1. The Sgo1 substitution V242E disrupted this interaction and impaired Sgo1 localization and function. Artificially targeting Sgo1 to centromeres restored proper meiotic segregation in cells lacking Swi6; artificially enriching cohesin largely bypassed the mitotic segregation requirement for heterochromatin but did not by itself restore meiotic segregation. Thus, in these experiments, cohesin accumulation and protection of that cohesin were separable requirements. Sgo1 recruitment also depends on the kinetochore-associated kinase Bub1, as shown by Tomoya Kitajima, Shigehiro Kawashima and Yoshinori Watanabe in 2004. (nature.com)

The attachment geometry differs between divisions. In mitosis, sister kinetochores normally biorient toward opposite spindle poles. In meiosis I, sister kinetochores co-orient toward the same pole, while homologous chromosomes orient toward opposite poles. Cleavage of arm cohesin resolves the chiasmata maintained after crossing over, allowing homologues to separate; centromeric cohesion persists, keeping sisters together. Meiosis II then requires sister-kinetochore biorientation and release of the remaining cohesion. Shugoshin recruits protein phosphatase 2A (PP2A) to protect centromeric cohesin during meiosis I. Two studies published in 2006 showed that this localized phosphatase counteracts phosphorylation associated with cohesin removal; in yeast, preventing Rec8 phosphorylation reduces its susceptibility to separase-mediated cleavage. (nature.com)

Protection must subsequently be disabled, but disappearance of shugoshin from the chromosome is not the only mechanism. In mouse oocytes, Sgo2 recruits PP2A and protects Rec8 during meiosis I; during metaphase II, Sgo2 redistributes toward kinetochores, away from the protected cohesin domain. Mammalian somatic-cell experiments likewise demonstrated tension-associated separation of shugoshin and cohesin. Mitotic protection principally opposes the early, nonproteolytic cohesin-release pathway, whereas separase cleaves the remaining cohesin at anaphase. These findings distinguish regulated spatial displacement of protection from destruction of cohesin itself. (nature.com)

Attachment correction and checkpoint signaling operate through centromere-associated proteins. Aurora B phosphorylates kinetochore substrates to destabilize inappropriate attachments. In 2009, Dan Liu and colleagues used phosphorylation biosensors to show that increased separation of kinetochore substrates from inner-centromeric Aurora B reduced phosphorylation and stabilized attachments, supporting a spatial mechanism linking tension to attachment stability. Unattached kinetochores separately catalyse assembly of the mitotic checkpoint complex, containing MAD2, BUBR1, BUB3 and CDC20, which inhibits the anaphase-promoting complex/cyclosome and delays anaphase. The pericentromere therefore supplies a mechanical and organizational setting for correction and surveillance; neither kinase-mediated correction nor checkpoint inhibition is an enzymatic function of centromeric DNA. (pubmed.ncbi.nlm.nih.gov)

Centromeric transcription, replication, and DNA repair

Centromeric repeats undergo transcription rather than constituting uniformly silent DNA. In human cells, RNA polymerase II produces noncoding RNAs from alpha-satellite arrays, including arrays that do not assemble an active kinetochore; transcription alone therefore does not identify a functional centromere. Experiments have identified associations between these RNAs and centromeric proteins. In 2014, Delphine Quénet and Yamini Dalal reported a centromeric RNA associated with CENP-A and its deposition chaperone HJURP; inhibition of transcription impaired their recruitment to human centromeres. In 2017, Shannon McNulty and colleagues demonstrated chromosome- and array-specific alpha-satellite transcripts associated with chromatin-bound CENP-A and CENP-C. Depleting particular transcripts impaired new CENP-A loading at the corresponding arrays, supporting a local contribution to centromeric chromatin organization rather than an interchangeable pool of satellite RNA. (elifesciences.org)

DNA replication must duplicate extensive repeat arrays while preserving their chromatin organization. Human alpha-satellite monomers can support replication initiation: a 2014 study found origins within endogenous arrays and replication during mid-to-late S phase, with CENP-A-associated chromatin replicating earlier than surrounding heterochromatic arrays. An engineered human artificial chromosome showed a different distribution, with initiation preferentially occurring in transcriptionally active regions. Replication timing and origin use consequently depend on chromosomal context, not simply on the presence of satellite repeats. Repetitive sequences can also adopt noncanonical DNA structures, while transcription creates potential encounters between RNA polymerases and replication forks. A R-loop consists of an RNA–DNA hybrid and a displaced DNA strand; persistent hybrids can impede fork progression through transcribed repeats. (academic.oup.com)

A 2021 human-cell study directly connected CENP-A chromatin to suppression of these replication obstacles. Rapid CENP-A removal during S phase increased centromeric transcripts and R-loops, slowed replication-fork progression, and produced unfinished replication, elevated recombination at alpha-satellite repeats, and subsequent centromeric chromosome breakage. Removing CENP-A at other cell-cycle stages did not reproduce the same replication-dependent effects. Expression of RNase H1, which degrades RNA within RNA–DNA hybrids, reduced centromeric damage and abnormal recombination; targeting this enzyme to alpha-satellite DNA also suppressed these defects. These experiments establish a protective function for CENP-A during replication, although the proposed mechanism whereby its chromatin might sequester transcripts and prevent their reannealing remains a model rather than a demonstrated molecular sequence. (pmc.ncbi.nlm.nih.gov)

Centromeric DNA repair can nevertheless use RNA–DNA hybrids beneficially in a different context. In 2021, Duygu Yilmaz and colleagues demonstrated recruitment of homologous recombination machinery to experimentally induced centromeric DNA breaks in human G1 cells, despite the absence of a sister chromatid. Dimethylation of histone H3 lysine 4, H3K4me2, supported damage-induced transcription and increased hybrid formation, promoting DNA-end resection. CENP-A and HJURP interacted with the deubiquitinase USP11, enabling formation of a RAD51–BRCA1–BRCA2 complex and RAD51 recruitment. Inhibiting homologous recombination in G1 increased centromeric instability and chromosome translocations. Thus, hybrids that obstruct undisturbed replication can also participate in the processing of a DNA break; their effects depend on their location, persistence, and cell-cycle context. (nature.com)

These findings derive principally from cultured human cells and do not establish a universal RNA requirement or repair-pathway hierarchy across organisms. Transcript depletion tests the contribution of particular RNAs, whereas polymerase inhibition also alters the act of transcription and its effects on nucleosomes; the two interventions are not mechanistically equivalent. Likewise, activation of recombination at induced G1 breaks does not resolve which homologous template is used at every endogenous lesion or how repair avoids inappropriate exchanges among similar repeats. The balance between RNA-supported protein assembly, replication-associated hybrid removal, and damage-induced hybrid formation therefore remains an experimentally defined, context-dependent problem in maintaining repeat-array integrity. (pmc.ncbi.nlm.nih.gov)

Neocentromeres, inactivation, and evolutionary change

Neocentromeres are functional centromeres established at chromosomal locations not previously serving that role, often without conventional centromeric satellite DNA. Their identity depends on assembling centromeric chromatin and a functional kinetochore, rather than merely acquiring a cytological constriction. In 1997, Theresa Depinet and colleagues characterized eight human accessory marker chromosomes lacking detectable alpha-satellite DNA: they appeared mitotically stable and recruited CENP-C and CENP-E, although CENP-B was absent. Such observations provide evidence for epigenetic specification: DNA normally occupying a noncentromeric position can support centromere activity without acquiring the repeat architecture characteristic of ordinary human centromeres. They do not imply that every genomic location is equally permissive for establishment. (scholars.duke.edu)

Centromere inactivation is the loss of activity at an existing site; it need not involve deletion of its satellite array. A 1995 study of a human dicentric X;15 translocation detected alpha-satellite DNA at both centromeric regions, but CENP-C and a primary constriction consistently identified only the X-derived site as active. Thus, retention of centromeric DNA does not establish continued kinetochore function. Centromere repositioning, in its strict evolutionary sense, instead denotes a change in the location of centromeric activity without alteration of the surrounding marker order. This differs from an inversion or translocation that changes a centromere’s apparent position by rearranging the chromosome. Giuseppe Montefalcone and colleagues documented repositioning in comparative primate mapping in 1999; subsequent work in Equus demonstrated analogous changes with conserved marker order. (pubmed.ncbi.nlm.nih.gov)

Satellite arrays also evolve through expansion, contraction, and sequence homogenization. Unequal exchange between misaligned repeat copies can generate reciprocal gains and losses, while repeated exchange can amplify particular sequence variants; gene conversion can transfer sequence information without reciprocal crossing-over. Their relative contributions remain incompletely resolved. A 1998 analysis of a 17-kilobase alpha-satellite fragment from human chromosome 21 identified complementary deletions consistent with reciprocal products of unequal recombination. Sequence comparisons in other arrays support local homogenization, but do not uniquely identify the responsible mechanism. Nicolas Altemose and colleagues’ 2022 analysis of T2T-CHM13 provided genome-wide evidence for “layered” evolution: comparatively young, homogeneous repeat blocks lie within older, more divergent satellite layers, whose accumulated mutations and transposable-element insertions help establish relative ages. The layered-expansion model proposes successive emergence and amplification of new variants, displacing older repeats toward array margins; the observed spatial and sequence patterns support this model without directly demonstrating every historical expansion event or its cause. (sciencedirect.com)

Centromere drive is biased transmission of competing centromeres during asymmetric female meiosis, in which one meiotic product contributes to the egg while others become polar bodies. Mouse-oocyte experiments by Takashi Akera and colleagues in 2017 showed that proximity to the cell cortex produces spindle asymmetry, with more tyrosinated microtubules on the cortical side. Centromeres with greater microtubule-destabilizing activity can exploit this asymmetry by preferentially detaching from cortical-side attachments and reorienting toward the egg. Their 2019 experiments in CF-1 × CHPO hybrid oocytes linked this behavior to centromeres carrying approximately six- to tenfold more minor-satellite repeats. Of the observed reorientation events, 81% directed the larger centromere toward the egg side; this measures flipping bias, not an 81% transmission rate. Increased BUB1-dependent recruitment of shugoshin-2 and destabilizing factors supplied a molecular mechanism connecting centromeric differences to biased orientation. (pubmed.ncbi.nlm.nih.gov)

The drive–suppression hypothesis proposes that transmission-enhancing centromeric variants can favor compensatory evolution of centromere-associated proteins. In 2021, Tomohiro Kumon and colleagues tested parallel kinetochore and heterochromatin pathways for recruiting microtubule-destabilizing effectors. Expression of divergent CENP-C alleles, including rat CENP-C in mouse oocytes, weakened kinetochore-pathway recruitment and reduced functional differences between competing centromeres. These experiments support a possible suppression mechanism. They do not reconstruct the historical selective pressures on particular protein substitutions: sequence signatures of adaptive evolution and cross-species replacement effects alone cannot establish that centromere drive caused every evolutionary change in CENP-A, CENP-C, or their interacting proteins. (pmc.ncbi.nlm.nih.gov)

Centromere dysfunction and genome instability

An acentric chromosome fragment lacks a functional centromere and therefore cannot assemble the normal kinetochore attachment to the spindle. Such fragments, produced by chromosome breakage, may segregate unequally, be lost from daughter nuclei, or become enclosed in micronuclei. Loss removes the fragment’s genetic material from the affected lineage; retention in a micronucleus can expose it to defective replication and further rearrangement. These outcomes are not inevitable: experiments in Drosophila neuroblasts have demonstrated transport and nuclear incorporation of acentric fragments through DNA tethers, microtubule bundles, and nuclear-envelope channels. Thus, lacking a centromere compromises reliable inheritance without necessarily causing immediate elimination. (pmc.ncbi.nlm.nih.gov)

A dicentric chromosome contains two centromeric regions, commonly following chromosome fusion or rearrangement. If both assemble active kinetochores and attach to opposite spindle poles, the intervening chromatin forms an anaphase bridge. Bridge breakage and subsequent fusion of unprotected chromosome ends can generate repeated breakage–fusion–bridge cycles, with deletions, inverted duplications, and amplification of chromosome segments. However, two centromeric DNA regions do not necessarily imply two competing spindle attachments. Inactivation of one centromere can leave a structurally dicentric chromosome functionally monocentric and stably inherited. Constitutional human Robertsonian translocations illustrate this distinction: a 1998 investigation of 15 translocation-containing cell lines detected CENP-C at only one centromere in a fraction of cells in every line. Closely spaced centromeres can also permit stability when both remain active. (pmc.ncbi.nlm.nih.gov)

Potential defects must be distinguished at three levels: centromeric DNA structure, the chromatin state established on that DNA, and the abundance or localization of centromere-associated proteins. These levels interact but are not interchangeable. A 2016 study of human chromosome 17 associated variation within the D17Z1 alpha-satellite array with alternative centromere placement at D17Z1-B; some variant D17Z1 arrays supporting centromere activity had reduced centromere-protein recruitment and impaired chromosome stability. This finding does not make satellite variation intrinsically pathological, because alternative active arrays can support stable inheritance. Conversely, a chromatin or protein-localization defect need not require a change in the underlying DNA sequence. Experimental CENP-A overexpression, for example, deposits this centromeric histone variant outside its normal domain. (pubmed.ncbi.nlm.nih.gov)

Impaired centromeric cohesion can separate sister chromatids before anaphase, undermining coordinated segregation and producing aneuploidy. Ho-Soo Lee and colleagues demonstrated a specific chromatin-mediated mechanism in 2018: loss of the remodeler RSF1 disrupted centromeric recruitment of HDAC1, allowing acetylation of histone H2A at lysine 118 to interfere with Bub1-dependent phosphorylation at threonine 120. The resulting failure to concentrate shugoshin 1 at centromeres weakened cohesion protection and caused premature sister-chromatid separation; experimentally tethering HDAC1 to centromeres suppressed this phenotype. Constitutional disease provides a separate example. Cells from individuals with Roberts syndrome, caused by biallelic pathogenic variants in ESCO2, characteristically show premature centromere separation and heterochromatin repulsion. A 2014 investigation additionally detected abnormal separation of sister pericentromeric regions during interphase. These observations concern cohesion establishment rather than absence of centromeric DNA. (nature.com)

In cancer research, elevated CENP-A expression and ectopic localization have been reported in tumor tissues and cultured cancer cells, but association alone does not establish that either initiates malignancy. Nicolas Lacoste and colleagues showed in 2014 that ectopic CENP-A deposition in human cells depends on the histone chaperone DAXX. A 2017 experimental study linked induced CENP-A overexpression and mislocalization to segregation defects, reduced centromeric CENP-T and kinetochore Nuf2, and increased micronucleus formation; DAXX depletion reduced mislocalization and rescued segregation defects in that system. These perturbation experiments support a causal mechanism for chromosome instability in the tested cells, not a general demonstration of cancer causation in patients. (pubmed.ncbi.nlm.nih.gov)

Discovery and experimental investigation

Early cytogenetic identification relied on the primary constriction visible in condensed chromosomes; molecular experiments subsequently made centromeric activity testable independently of chromosome shape. In 1980, Louise Clarke and John Carbon isolated a 1.6-kilobase DNA segment containing CEN3 from chromosome III of Saccharomyces cerevisiae. Adding this segment to a plasmid carrying a yeast replication origin produced minichromosomes that were stable during mitosis and segregated through both meiotic divisions. This established a functional assay for centromeric DNA rather than merely a positional description. In 1985, William C. Earnshaw and Naomi F. Rothfield used human anticentromere autoantibodies, immunoblotting, and immunofluorescence to identify CENP-A, CENP-B, and CENP-C, with apparent molecular masses of 17, 80, and 140 kilodaltons. Biochemical work published in 1987 showed that CENP-A copurified with core histones and nucleosome particles, linking the cytological landmark to a specialized chromatin component. (pubmed.ncbi.nlm.nih.gov)

Chromosome spreads, fluorescence in situ hybridization (FISH), and immunofluorescence provide complementary spatial information. Spreads expose condensed chromosomes for examining constrictions and rearrangements; fluorescent DNA probes identify particular chromosomes or satellite arrays, whereas antibodies locate centromere-associated proteins. These measurements need not coincide: an alpha-satellite FISH signal identifies DNA, not necessarily an active centromere. Combined immunofluorescence–FISH can therefore distinguish a repeat-bearing site from one retaining CENP-C and other components associated with functional centromeres. A 2016 study demonstrated this combination on methanol–acetic-acid-fixed human cell lines and also documented a methodological limitation: fixation can abolish recognition by some antibodies, including a widely used CENP-A antibody. Absence of staining consequently requires interpretation in light of sample preparation and antibody performance. (link.springer.com)

Live-cell imaging adds a temporal dimension that fixed preparations cannot provide. Fluorescently tagged centromere proteins can be followed during the cell cycle, while experimental protein depletion or gene disruption tests whether an observed component is required for assembly or segregation. Inducible degradation systems permit rapid removal, helping separate immediate consequences from defects accumulated over several divisions. In a 2019 study, researchers replaced endogenous CENP-A with variants carrying substitutions at serine 7 and followed chromosome alignment, lagging chromosomes, and micronucleus formation by live imaging. The substitutions did not reproduce the proposed loss of centromere function under the tested conditions, illustrating how depletion-and-rescue experiments can test a specific biochemical hypothesis rather than infer necessity from localization alone. (pmc.ncbi.nlm.nih.gov)

Chromatin immunoprecipitation (ChIP) maps protein-associated DNA by fragmenting chromatin, recovering antibody-bound complexes, and identifying their DNA, commonly by sequencing. Native ChIP avoids chemical crosslinking and is particularly useful for examining nucleosomal proteins. CUT&RUN, introduced by Peter J. Skene and Steven Henikoff in 2017, instead uses antibody-targeted micrococcal nuclease to cleave near the selected protein in situ; released fragments are collected for sequencing. Both approaches provide genomic occupancy maps, although near-identical repeats complicate assignment of individual reads. Structural methods address a different scale: cryo-electron microscopy and X-ray crystallography reveal molecular interfaces in purified complexes. For example, structural and mutational studies reported in 2018 showed that CENP-N recognizes the L1 loop of CENP-A while also contacting nucleosomal DNA, connecting a resolved interface to kinetochore assembly. (elifesciences.org)

Centromeric arrays resisted conventional genome assembly because short reads could not reliably distinguish or order long stretches of nearly identical repeats. The 2022 T2T-CHM13 project combined highly accurate PacBio HiFi reads with Oxford Nanopore ultralong reads, exploiting subtle repeat differences and longer-range sequence continuity. Its reference assembly resolved all 22 autosomes and the X chromosome; it did not include a Y chromosome. Nicolas Altemose and colleagues then characterized 189.9 megabases of centromeric and pericentromeric repeats—6.2% of that assembly—and integrated sequence information with CENP-A chromatin profiling to locate active domains. These maps describe a particular, largely homozygous cell-line genome, not population-wide centromere architecture: comparisons of X centromeres among individuals revealed substantial sequence, structural, and epigenetic variation. (eichlerlab.gs.washington.edu)

Artificial centromeres and unresolved research questions

In April 1997, John J. Harrington, Gil Van Bokkelen, Robert W. Mays, Karen Gustashaw, and Huntington F. Willard reported first-generation human artificial chromosomes (HACs) constructed by introducing long synthetic alpha-satellite arrays, telomeric DNA, and genomic DNA into human HT1080 cells. The resulting linear microchromosomes were estimated to contain 6–10 megabases of DNA, recruited proteins characteristic of active centromeres, and remained mitotically and cytogenetically stable without selection for up to six months in culture. These experiments demonstrated de novo centromere construction: introduced material acquired a functional centromere rather than merely retaining one from an existing chromosome. Their inheritance provided a functional test beyond detecting centromeric proteins on introduced DNA. (pubmed.ncbi.nlm.nih.gov)

Artificial-chromosome assays distinguish sequence features that promote centromere establishment from mechanisms that subsequently maintain it. In 2002, Jun-ichirou Ohzeki and colleagues compared synthetic alphoid arrays with intact or mutated CENP-B-binding sites, alongside nonalphoid repeats carrying or lacking those sites. In their HT1080-cell assay, both alphoid sequence context and functional CENP-B boxes were required for artificial-chromosome formation; adding CENP-B boxes to unrelated repeats was insufficient. This was an experimentally defined requirement, not a universal rule for centromeres. In 2019, Glennis A. Logsdon and colleagues generated HACs without alpha-satellite DNA by transiently targeting the CENP-A assembly factor HJURP to engineered DNA. Such results separate a sequence’s capacity to initiate centromeric chromatin from the capacity of an established domain to undergo epigenetic propagation. (pmc.ncbi.nlm.nih.gov)

Targeted recruitment also tests the sufficiency of particular assembly steps. Meghan C. Barnhart and colleagues showed in 2011 that tethering HJURP through a LacI–LacO system to a noncentromeric chromosomal array deposited CENP-A, recruited inner-centromere proteins, and assembled a functional kinetochore. Artificial targeting bypassed the normal requirement for Mis18-mediated HJURP recruitment, identifying local delivery of the chaperone as an experimentally sufficient initiating step. Conversely, chromatin manipulation can disrupt an established artificial centromere. In a 2011 study, tethering the demethylase LSD1 to a synthetic HAC depleted histone H3 lysine-4 dimethylation, reduced underlying satellite transcription and HJURP recruitment, impaired new CENP-A incorporation, and eventually inactivated the kinetochore. These experiments test causal contributions of assembly factors and histone modifications rather than relying solely on their colocalization. (pmc.ncbi.nlm.nih.gov)

How centromeric domains acquire their size and boundaries remains incompletely explained. In engineered chicken DT40 cells, a 2015 study found that tethered full-length HJURP produced a CENP-A domain of approximately 120 kilobases, whereas its N-terminal 1–254-amino-acid fragment produced a domain of approximately 40 kilobases. These construct-specific measurements implicate HJURP regions in domain expansion but do not establish a universal minimum or optimal centromere size. Outstanding questions include how local chromatin restricts expansion, how much CENP-A is needed for durable inheritance, and which conditions permit ectopic chromatin to mature into a complete kinetochore. Human-cell experiments showing that dispersed CENP-A accumulation after overexpression need not produce functional kinetochores indicate that ectopic deposition alone is insufficient. (pmc.ncbi.nlm.nih.gov)

The relationship between repeat-array variation, epigenetic plasticity, and segregation performance likewise lacks a general predictive model. A 2024 comparison of two complete human centromere sets found up to threefold differences in centromere size and at least a 4.1-fold increase in single-nucleotide variation relative to unique flanking sequences. These observations document variation but do not establish that a larger or more divergent array segregates less accurately. Engineered chromosomes offer controlled tests, although their own architecture can complicate interpretation: earlier HAC constructs frequently multimerized and rearranged during establishment. In 2024, Craig W. Gambogi and colleagues reported efficient single-copy HAC formation using larger input constructs, enabling experiments in which chromosome architecture is more directly specified. Their cultured-cell inheritance results establish an experimental chromosome-engineering platform, not an established clinical application; proposed therapeutic uses remain distinct from demonstrated delivery, safety, and efficacy in patients. (pmc.ncbi.nlm.nih.gov)

References

1. The Centromere: Chromatin Foundation for the Kinetochore Machinerypmc.ncbi.nlm.nih.gov
2. Chromosomes Fact Sheetgenome.gov
3. Centrosomegenome.gov
4. Chromosome Map — Genes and Diseasencbi.nlm.nih.gov
5. Revisiting the question: When is a centromere not a kinetochore?pmc.ncbi.nlm.nih.gov
6. Genetics, Chromosomesncbi.nlm.nih.gov
7. Recombination between heterologous human acrocentric chromosomespmc.ncbi.nlm.nih.gov
8. Chromosome banding techniquespubmed.ncbi.nlm.nih.gov
9. Visualization of centromere proteins CENP-B and CENP-C on a stable dicentric chromosome in cytological spreadspubmed.ncbi.nlm.nih.gov
10. Active centromere and chromosome identification in fixed cell linespmc.ncbi.nlm.nih.gov
11. Determination of the binding constants of the centromere protein Cbf1 to all 16 centromere DNAs of Saccharomyces cerevisiaepmc.ncbi.nlm.nih.gov
12. Architecture of the CBF3-Centromere Complex of the Budding Yeast Kinetochorepmc.ncbi.nlm.nih.gov
13. Probing the Saccharomyces cerevisiae centromeric DNA (CEN DNA)–binding factor 3 (CBF3) kinetochore complex by using atomic force microscopypmc.ncbi.nlm.nih.gov
14. The genome sequence of Schizosaccharomyces pombenature.com
15. Interspecies conservation of organisation and function between nonhomologous regional centromerespmc.ncbi.nlm.nih.gov
16. Complete genomic and epigenetic maps of human centromerespubmed.ncbi.nlm.nih.gov
17. Meiotic recombination dynamics in plants with repeat-based holocentromeres shed light on the primary drivers of crossover patterningnature.com
18. Repeat-based holocentromeres of the woodrush Luzula sylvatica reveal insights into the evolutionary transition to holocentricitypmc.ncbi.nlm.nih.gov
19. Human centromeres produce chromosome-specific and array-specific alpha satellite transcripts that are complexed with CENP-A and CENP-Cpmc.ncbi.nlm.nih.gov
20. Telomere-to-telomere assembly of a complete human X chromosomepmc.ncbi.nlm.nih.gov
21. DNA sequence-specific binding of CENP-B enhances the fidelity of human centromere functionpmc.ncbi.nlm.nih.gov
22. Centromere Protein B Null Mice are Mitotically and Meiotically Normal but Have Lower Body and Testis Weightspmc.ncbi.nlm.nih.gov
23. Complete genomic and epigenetic maps of human centromerespmc.ncbi.nlm.nih.gov
24. Composition and structure of the centromeric region of rice chromosome 8pubmed.ncbi.nlm.nih.gov
25. The CentO satellite confers translational and rotational phasing on cenH3 nucleosomes in rice centromerespmc.ncbi.nlm.nih.gov
26. Genome sequence, comparative analysis and population genetics of the domestic horse (Equus caballus)pmc.ncbi.nlm.nih.gov
27. Centromere sliding on a mammalian chromosomepmc.ncbi.nlm.nih.gov
28. Centromeric chromatin exhibits a histone modification pattern that is distinct from both euchromatin and heterochromatinpmc.ncbi.nlm.nih.gov
29. Crystal structure of the human centromeric nucleosome containing CENP-Anature.com
30. A conserved mechanism for centromeric nucleosome recognition by centromere protein CENP-Cpubmed.ncbi.nlm.nih.gov
31. Structural mechanisms of centromeric nucleosome recognition by the kinetochore protein CENP-Npmc.ncbi.nlm.nih.gov
32. Structure of the human inner kinetochore CCAN complex and its significance for human centromere organizationpmc.ncbi.nlm.nih.gov
33. Structural insights into human CCAN complex assembled onto DNApmc.ncbi.nlm.nih.gov
34. Molecular basis of outer kinetochore assembly on CENP-Tpubmed.ncbi.nlm.nih.gov
35. Implications for kinetochore-microtubule attachment from the structure of an engineered Ndc80 complexpubmed.ncbi.nlm.nih.gov
36. A two-step mechanism for epigenetic specification of centromere identity and functionnature.com
37. Propagation of centromeric chromatin requires exit from mitosispubmed.ncbi.nlm.nih.gov
38. Inheritance of CENP-A Nucleosomes during DNA Replication Requires HJURPpubmed.ncbi.nlm.nih.gov
39. Centromere-specific assembly of CENP-a nucleosomes is mediated by HJURPpubmed.ncbi.nlm.nih.gov
40. Priming of centromere for CENP-A recruitment by human hMis18alpha, hMis18beta, and M18BP1pubmed.ncbi.nlm.nih.gov
41. Molecular basis for Cdk1-regulated timing of Mis18 complex assembly and CENP-A depositionpubmed.ncbi.nlm.nih.gov
42. Mechanism of centromere recruitment of the CENP-A chaperone HJURP and its implications for centromere licensingpubmed.ncbi.nlm.nih.gov
43. CENP-C facilitates the recruitment of M18BP1 to centromeric chromatinpmc.ncbi.nlm.nih.gov
44. Mitotic regulator Mis18β interacts with and specifies the centromeric assembly of molecular chaperone holliday junction recognition protein (HJURP)pubmed.ncbi.nlm.nih.gov
45. Direct recruitment of Mis18 to interphase spindle pole bodies promotes CENP-A chromatin assemblypubmed.ncbi.nlm.nih.gov
46. HJURP is a CENP-A chromatin assembly factor sufficient to form a functional de novo kinetochorepubmed.ncbi.nlm.nih.gov
47. Individual pericentromeres display coordinated motion and stretching in the yeast spindlepmc.ncbi.nlm.nih.gov
48. Convergent genes shape budding yeast pericentromeresnature.com
49. Selective recognition of methylated lysine 9 on histone H3 by the HP1 chromo domainnature.com
50. Recruitment of cohesin to heterochromatic regions by Swi6/HP1 in fission yeastnature.com
51. Heterochromatin links to centromeric protection by recruiting shugoshinnature.com
52. The conserved kinetochore protein shugoshin protects centromeric cohesion during meiosisnature.com
53. Protein phosphatase 2A protects centromeric sister chromatid cohesion during meiosis Inature.com
54. Shugoshin collaborates with protein phosphatase 2A to protect cohesinnature.com
55. Unified mode of centromeric protection by shugoshin in mammalian oocytes and somatic cellsnature.com
56. Structure of cohesin subcomplex pinpoints direct shugoshin-Wapl antagonism in centromeric cohesionnature.com
57. Sensing chromosome bi-orientation by spatial separation of aurora B kinase from kinetochore substratespubmed.ncbi.nlm.nih.gov
58. Juxtaposition of Bub1 and Cdc20 on phosphorylated Mad1 during catalytic mitotic checkpoint complex assemblynature.com
59. CENP-A chromatin prevents replication stress at centromeres to avoid structural aneuploidypmc.ncbi.nlm.nih.gov
60. Activation of homologous recombination in G1 preserves centromeric integritynature.com
61. Characterization of neo-centromeres in marker chromosomes lacking detectable alpha-satellite DNA.scholars.duke.edu
62. Further evidence that CENP-C is a necessary component of active centromeres: studies of a dic(X; 15) with simultaneous immunofluorescence and FISHpubmed.ncbi.nlm.nih.gov
63. Centromere repositioning.pubmed.ncbi.nlm.nih.gov
64. Phylogeny of horse chromosome 5q in the genus Equus and centromere repositioning.pubmed.ncbi.nlm.nih.gov
65. Differential Rates of Local and Global Homogenization in Centromere Satellites From Arabidopsis Relativespmc.ncbi.nlm.nih.gov
66. Complete genomic and epigenetic maps of human centromereseichlerlab.gs.washington.edu
67. Spindle asymmetry drives non-Mendelian chromosome segregationpubmed.ncbi.nlm.nih.gov
68. Molecular strategies of meiotic cheating by selfish centromerespmc.ncbi.nlm.nih.gov
69. Parallel pathways for recruiting effector proteins determine centromere drive and suppressionpmc.ncbi.nlm.nih.gov
70. Chromothripsis as an on-target consequence of CRISPR-Cas9 genome editingpmc.ncbi.nlm.nih.gov
71. ESCRT-III–mediated membrane fusion drives chromosome fragments through nuclear envelope channelspmc.ncbi.nlm.nih.gov
72. MECHANISMS GENERATING CANCER GENOME COMPLEXITY FROM A SINGLE CELL DIVISION ERRORpmc.ncbi.nlm.nih.gov
73. Chromosome stability is maintained by short intercentromeric distance in functionally dicentric human Robertsonian translocationspubmed.ncbi.nlm.nih.gov
74. Genomic variation within alpha satellite DNA influences centromere location on human chromosomes with metastable epiallelespubmed.ncbi.nlm.nih.gov
75. The chromatin remodeler RSF1 controls centromeric histone modifications to coordinate chromosome segregationnature.com
76. 3D-FISH analysis reveals chromatid cohesion defect during interphase in Roberts syndromepmc.ncbi.nlm.nih.gov
77. Mislocalization of the centromeric histone variant CenH3/CENP-A in human cells depends on the chaperone DAXXpubmed.ncbi.nlm.nih.gov
78. Mislocalization of centromeric histone H3 variant CENP-A contributes to chromosomal instability (CIN) in human cellspmc.ncbi.nlm.nih.gov
79. Isolation of a yeast centromere and construction of functional small circular chromosomespubmed.ncbi.nlm.nih.gov
80. Identification of a family of human centromere proteins using autoimmune sera from patients with sclerodermahero.epa.gov
81. A 17-kD centromere protein (CENP-A) copurifies with nucleosome core particles and with histonespmc.ncbi.nlm.nih.gov
82. Active centromere and chromosome identification in fixed cell lineslink.springer.com
83. Phosphorylation of CENP-A on serine 7 does not control centromere functionpmc.ncbi.nlm.nih.gov
84. Structural mechanisms of centromeric nucleosome recognition by the kinetochore protein CENP-Npubmed.ncbi.nlm.nih.gov
85. The complete sequence of a human genomeeichlerlab.gs.washington.edu
86. Complete genomic and epigenetic maps of human centromeresescholarship.org
87. Formation of de novo centromeres and construction of first-generation human artificial microchromosomespubmed.ncbi.nlm.nih.gov
88. CENP-B box is required for de novo centromere chromatin assembly on human alphoid DNApmc.ncbi.nlm.nih.gov
89. Human Artificial Chromosomes that Bypass Centromeric DNApmc.ncbi.nlm.nih.gov
90. HJURP is a CENP-A chromatin assembly factor sufficient to form a functional de novo kinetochorepmc.ncbi.nlm.nih.gov
91. Epigenetic engineering shows H3K4me2 is required for HJURP targeting and CENP-A assembly on a synthetic human kinetochorepmc.ncbi.nlm.nih.gov
92. HJURP is involved in the expansion of centromeric chromatinpmc.ncbi.nlm.nih.gov
93. The variation and evolution of complete human centromerespmc.ncbi.nlm.nih.gov
94. Efficient Formation of Single-copy Human Artificial Chromosomespmc.ncbi.nlm.nih.gov