Centromeric chromatin: what makes it unique?

https://doi.org/10.1016/j.gde.2005.01.004Get rights and content

Centromeres represent the final frontier of eukaryotic genomes. Although they are defining features of chromosomes — the points at which spindle microtubules attach — the fundamental features that distinguish them from other parts of the chromosome remain mysterious. The function of centromeres is conserved throughout eukaryotic biology, but their DNA sequences are not. Rather, accumulating evidence favors chromatin-based centromeric identification. To understand how centromeric identity is maintained, researchers have studied DNA–protein interactions at native centromeres and ectopic ‘neocentromeres’. Other studies have taken a comparative approach focusing on centromere-specific proteins, of which mammalian CENP-A and CENP-C are the prototypes. Elucidating the assembly and structure of chromatin at centromeres remain key challenges.

Introduction

Centromeres were described even before the rediscovery of Mendel's laws of inheritance. By the 1880s, cytologists realized that chromosomes are characterized by a constriction that corresponds to the site of spindle attachment during mitosis [1]. As the only part of the chromosome that is directly involved in the process of mitosis, centromeres are fundamental to eukaryotic biology. Centromeric DNA attaches to spindle microtubules through a proteinaceous structure referred to as the kinetochore [2]. The outer kinetochore assembles when chromosomes condense and microtubules attach to it for mitosis, and it disassembles after the chromosomes have segregated. The inner kinetochore remains with the DNA throughout the cell cycle and consists of centromeric chromatin and associated ‘foundation’ proteins [3].

What has made centromeres so mysterious is the lack of any conserved sequence across species [4] despite the fact that all centromeres have an identical function — to organize the kinetochore at mitosis. This is not to say that centromeric DNA lacks any distinguishing features, because most centromeres comprise long stretches of short tandem repetitive ‘satellite’ DNA sequences that are found only there and in surrounding pericentric heterochromatin. The highly repetitive nature of centromeric satellites has made centromeres nearly intractable to sequence analysis except in special cases.

A key insight into the basis for centromere identity came from the discovery that a mammalian centromere-specific protein, centromere protein A (CENP-A), is homologous to histone H3 and is packaged into chromatin [5]. H3 is one of the four histones that form an octamer that packages the rest of the genome and that is assembled into nucleosomes during replication. Finding that H3 is replaced in centromeric chromatin by an H3-like variant suggests that centromeric chromatin is unique because of its histone complement and not because of its DNA sequence. Support for this hypothesis comes from the fact that CENP-A and its centromeric H3 (CenH3) counterparts in other organisms are found at all centromeres and are essential for centromere formation, whereas this is not the case for centromeric repeats [3].

There have been numerous reviews in recent years discussing diverse aspects of centromere structure, function and evolution [2, 4, 6, 7, 8, 9, 10, 11]. In this review, we emphasize the progress that has been made in the understanding of centromeric chromatin since the subject was last reviewed in this series [9].

Section snippets

Centromere sequence organization

Centromeres in budding yeast are short and simple, and consist of common sequence elements that span just 125 bp [12]. One of these elements, CDEIII, is the binding site for the CDF3 multiprotein complex [13]. This complex is responsible for targeting the apparently single Cse4p-containing nucleosome — Cse4p is the yeast CenH3 [14]. Budding yeast centromeres, therefore, are well defined by DNA sequence alone.

Fission yeast centromeres are defined in a very different way than those of budding

DNA–protein interactions

Although dozens of proteins localize to the kinetochore [18, 19], the large majority of these are seen only at mitosis, and only two of these, CenH3 and centromere protein C (CENP-C), are known to bind DNA and to be widely distributed in evolution [8]. In every case that has been examined, CENP-C localization depends on the presence of CenH3 but not vice versa [27, 28, 29, 30, 31]. CenH3-containing nucleosomes appear, therefore, to provide the chromatin framework for centromeres, and much

Centromere evolution

Despite evidence for specificity between centromere-binding proteins and centromeric DNA, no sequence determinants have been identified for any complex centromere. This conclusion comes primarily from the existence of ∼70 different human neocentromeres that entirely lack α-satellite DNA, which is usually discovered from karyotype analysis that follows a diagnosis of possible aneuploidy [7]. Rarely, neocentromeres are discovered in otherwise karyotypically normal individuals, and three such

Assembly of centromeric chromatin and kinetochore function

There seems to be little doubt that centromere identity and the presence of CenH3-containing nucleosomes are inseparable. It is still an open question, however, whether or not there are undiscovered centromere determinants, because overproduction of CENP-A in human cells recruited CENP-C but did not lead to ectopic centromere formation [30]. Furthermore, human artificial centromeres have been produced using amplified α-satellite repeats found at native centromeres but not from α-satellite

Conclusions

Despite considerable recent progress in defining centromeric components, we still lack a clear understanding of how centromeres are distinguished from ordinary regions of the genome that are not centromeres. Maintenance of centromeric chromatin, which forms the foundation for spindle attachment, is extraordinary insofar as centromeres remain in the same cytological position over tens of millions of years. Yet the occasional appearance of neocentromeres in numerous regions of ordinary sequence

References and recommended reading

Papers of particular interest, published within the annual period of review, have been highlighted as:

  • • of special interest

  • •• of outstanding interest

Acknowledgements

We thank Takè Furuyama and Paul Talbert for comments on the manuscript.

References (52)

  • M.D. Blower et al.

    Conserved organization of centromeric chromatin in flies and humans

    Dev Cell

    (2002)
  • K.F. Sullivan

    A solid foundation: functional specialization of centromeric chromatin

    Curr Opin Genet Dev

    (2001)
  • A. Alonso et al.

    Genomic microarray analysis reveals distinct locations for the CENP-A binding domains in three human chromosome 13q32 neocentromeres

    Hum Mol Genet

    (2003)
  • K. Ahmad et al.

    Histone H3 variants specify modes of chromatin assembly

    Proc Natl Acad Sci USA

    (2002)
  • W. Flemming

    Zellsubstanz, Kern und Zelltheilung

    (1882)
  • D.K. Palmer et al.

    Purification of the centromere-specific protein CENP-A and demonstration that it is a distinctive histone

    Proc Natl Acad Sci USA

    (1991)
  • P.E. Warburton

    Chromosomal dynamics of human neocentromere formation

    Chromosome Res

    (2004)
  • A.A. Hyman et al.

    Microtubule-motor activity of a yeast centromere-binding protein complex

    Nature

    (1992)
  • P. Bernard et al.

    Requirement of heterochromatin for cohesion at centromeres

    Science

    (2001)
  • S. Henikoff et al.

    The centromere paradox: stable inheritance with rapidly evolving DNA

    Science

    (2001)
  • V.S. Nekrasov et al.

    Interactions between centromere complexes in Saccharomyces cerevisiae

    Mol Biol Cell

    (2003)
  • P. De Wulf et al.

    Hierarchical assembly of the budding yeast kinetochore from multiple subcomplexes

    Genes Dev

    (2003)
  • S.L. Baldauf et al.

    A kingdom-level phylogeny of eukaryotes based on combined protein data

    Science

    (2000)
  • T. Hosouchi et al.

    Physical map-based sizes of the centromeric regions of Arabidopsis thaliana chromosomes 1, 2, and 3

    DNA Res

    (2002)
  • X. Sun et al.

    Sequence analysis of a functional Drosophila centromere

    Genome Res

    (2003)
  • M.K. Rudd et al.

    Sequence organization and functional annotation of human centromeres

    Cold Spring Harb Symp Quant Biol

    (2003)
  • View full text