The centromeres of eukaryotic chromosomes represent one of the most daunting challenges in genomic sequencing. Obtaining the sequence of entire chromosomes or portions of chromosomes requires the cloning and sequencing of manageable fragments of chromosomal DNA, typically cloned into bacterial (BAC) or P1 plasmid-derived (PAC) artificial chromosomes, and then rearranging the resulting sequences into the proper order. Typically, a minimal tiling path (a subset of clones with minimally overlapping sequences) encompassing the chromosome, or region of interest along a chromosome, is constructed from a genomic library by creating a physical map of the region and aligning selected clones with the use of known markers. Sequences obtained from this subset of clones then are arranged into unambiguous order based on their overlapping regions. Centromeric regions of most multicellular eukaryotic chromosomes are resistant to this process because they consist of long stretches of highly repetitive satellite DNA sequences and transposable elements. Even the complete genomic sequences of organisms such as Arabidopsis and humans contain major gaps in the centromeric regions.
Centromere complexity and the associated recalcitrance to sequencing have impeded efforts to understand the regulation of centromere function. Centromeres present one of the great conundrums of eukaryotic biology: their functions in promoting sister chromatid cohesion, kinetochore assembly, and spindle fiber attachment to ensure the faithful segregation of chromosomes during cell division are the same in all multicellular eukaryotes, yet there is a striking lack of conservation of centromere DNA sequences, even between closely related species. Centromeres of most eukaryotes exhibit the hallmarks of an epigenetic locus whose function is not defined by the primary DNA sequence but instead is determined by a self-replicating protein complex that interacts with the locus. Thus, centromeres might be defined as regions of chromatin that contain a distinct type of nucleosome in which histone H3 is replaced by a divergent centromere-specific histone, called CENH3 in plants and CENP-A in humans (Cleveland et al., 2003; Jiang et al., 2003).
The nondependence of centromeres on primary DNA sequence is illustrated by the occasional occurrence of neocentromeres, which are ectopic centromeres that originate from noncentromeric regions of chromosomes. At least 49 cases of human neocentromere have been described, all of which exhibit a complete lack of normal centromeric-type sequence (such as satellite repeats) and yet bind CENP-A, assemble a functional kinetochore, and exhibit mitotic stability (Choo, 2001). However, some dependence on primary DNA sequence or organization of the centromeric region generally is considered important because centromeres normally are maintained at the same loci from generation to generation.
Centromere sequencing efforts in several organisms, such as human, Drosophila, and Arabidopsis, have suggested that satellite repeats and transposable element DNA play a fundamental role in defining centromeres (reviewed in Lamb and Birchler, 2003). An important part of this role might be to provide a distinctive structure and spacing to stretches of DNA that encourages the binding of CENH3 rather than histone H3 and the consequent formation of centromere-specific nucleosomes that have specific functions in enabling kinetochore assembly. A full understanding of primary DNA sequence versus epigenetic mechanisms for centromere formation has yet to be achieved.
In a breakthrough for centromere biology, two groups now present the complete high-quality sequence of the centromere of rice chromosome 8 (Cen8). In this issue of The Plant Cell, Wu et al. (pages 967–976) present an analysis of 1.97 Mb of contiguous DNA sequence assembled from 18 BAC or PAC clones encompassing the centromeric region of Cen8. In the February issue of Nature Genetics, Nagaki et al. (2004) reported the sequencing and analysis of a slightly narrower region of Cen8, encompassing 1.65 Mb of contiguous DNA obtained from 12 BAC clones. These two papers reveal several important findings related to rice Cen8 structure and function and represent a major achievement that should pave the way for rapid advances in our understanding of centromere function and regulation.
Previous cytological work (Dong et al., 1998; Cheng et al., 2002) found that rice centromeres contain repeats of a 155-bp satellite DNA sequence called CentO. These reports also showed that several rice chromosomes contain a limited amount of satellite DNA relative to other species and that Cen8 contains the least amount of satellite DNA sequence (∼64 kb) of the 12 rice chromosomes. Therefore, this centromere was a natural first choice for attempts to obtain complete high-quality DNA sequence.
Wu et al. defined the position of rice Cen8 within a 2-Mb region from a previous study (Wu et al., 2002) of chromosomal mapping using rice ESTs and centromere-specific satellite sequences (Figure 1) and assembled a tiling path of this region from 18 BAC or PAC clones. The complete sequence was resolved by repeated subcloning and transposon-based sequencing of the highly repetitive regions and was confirmed by comparing sizes of fragments generated by restriction enzyme digests of each PAC/BAC clone with the virtual fragments generated from the assembled sequence. Wu et al. estimated the overall accuracy of the resulting 1.97-Mb contiguous sequence at >99.99%. Nagaki et al. (2004) also mapped the Cen8 region using CentO and centromere-specific retrotransposon sequence probes and constructed a minimal tiling path of 12 BAC clones encompassing this region. These authors used similar methods to those of Wu et al. (e.g., transposon-based sequencing for the assembly of highly repetitive sequence) and constructed a virtual contig of ∼1.65 Mb spanning Cen8. Overall, the results of both papers were very similar, as expected.
Figure 1.
Fluorescent in Situ Hybridization Analysis of rice (Oryza sativa japonica cv Nipponbare) Mitotic Chromosomes Using CentO Satellite Repeats as Probes. (Figure courtesy of H. Mizuno.)
Both groups found that known repetitive elements, mainly retrotransposons, comprise 58 to 59% of the Cen8 sequence. Wu et al. described the majority of these as gypsy-type retrotransposons. Nagaki et al. (2004) found an additional 14% of the virtual contig to be derived from miniature inverted repeats, unknown repeats, and truncated transposons and thus estimated that 72% of the total Cen8 sequence is repetitive. CentO satellite repeats were found mainly in three clusters in a 40- to 80-kb segment located near the center of the centromere. Nagaki et al. (2004) used antibody to CENH3 in chromatin immunoprecipitation assays to define the region of the centromere that underlies the kinetochore. They found that this region encompasses ∼750 kb and includes the major CentO repeat clusters.
One of the most interesting findings to emerge from these reports is the discovery that the centromere harbors actively transcribed genes. Gene prediction analysis by Wu et al. revealed the presence of 201 predicted open reading frames in the 1.97-Mb sequence spanning the centromere, 48 of which were highly similar to sequences of known rice proteins or cDNA clones (Figure 2). By contrast, Nagaki et al. (2004) identified 47 putative genes, 14 of which lie within the defined (CENH3 binding) kinetochore region. Nagaki et al. (2004) further analyzed expression of 39 intron-containing open reading frame sequences using RT-PCR. They found that 12 of the genes are expressed in rice leaf and root tissues, including four located in the CENH3 binding kinetochore region. These results contrast with reports that centromeres do not contain actively transcribed genes in humans (Schueler et al., 2001), Drosophila (Sun et al., 1997), and Arabidopsis (Hosouchi et al., 2002). Nagaki et al. (2004) propose that the presence of actively transcribed genes in rice Cen8, together with the relatively small amount of satellite DNA, reflects an early stage of centromere evolution. The centromere sequences and analyses presented in these reports thus represent important contributions to our knowledge of centromere structure and function, epigenetic mechanisms of inheritance, and genome evolution.
Figure 2.
Composition of the Predicted Genes in the 1.97-Mb Centromeric Region of Rice Chromosome 8.
The predicted exons (yellow or brown), models from gene prediction programs (green), and regions with homology to known proteins (blue), rice ESTs (orange), or rice full-length cDNA sequences (red) are shown. Detailed information can be accessed through the Rice Annotation Database at http://golgi.gs.dna.affrc.go.jp/SY-1102/rad/index.html. (Figure courtesy of T. Sasaki.)
References
- Cheng, Z., Dong, F., Langdon, T., Ouyang, S., Buell, C.R., Gu, M., Blattner, F.R., and Jiang, J. (2002). Functional rice centromeres are marked by a satellite repeat and a centromere-specific retrotransposon. Plant Cell 14, 1691–1704. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Choo, K.H.A. (2001). Domain organization at the centromere and neocentromere. Dev. Cell 1, 165–177. [DOI] [PubMed] [Google Scholar]
- Cleveland, D.W., Mao, Y., and Sullivan, K.F. (2003). Centromeres and kinetochores: From epigenetics to mitotic checkpoint signaling. Cell 112, 407–421. [DOI] [PubMed] [Google Scholar]
- Dong, F., Miller, J.T., Jackson, S.A., Wang, G.-L., Ronald, P.C., and Jiang, J. (1998). Rice (Oryza sativa) centromeric regions consist of highly complex DNA. Proc. Natl. Acad. Sci. USA 95, 8135–8140. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hosouchi, T., Kumekawa, N., Tsuruoka, H., and Kotani, H. (2002). Physical map-based sizes of the centromeric regions of Arabidopsis thaliana chromosomes 1, 2, and 3. DNA Res. 9, 117–121. [DOI] [PubMed] [Google Scholar]
- Jiang, J., Birchler, J.A., Parrott, W.A., and Dawe, R.K. (2003). A molecular view of plant centromeres. Trends Plant Sci. 8, 570–575. [DOI] [PubMed] [Google Scholar]
- Lamb, J., and Birchler, J. (2003). The role of DNA sequence in centromere formation. Genome Biol. 4, 214. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Nagaki, K., Cheng, Z., Ouyang, S., Talbert, P., Kim, M., Jones, K.M., Henikoff, S., Buell, C.R., and Jiang, J. (2004). Sequencing of a rice centromere uncovers active genes. Nat. Genet. 36, 138–145. [DOI] [PubMed] [Google Scholar]
- Schueler, M.G., Higgins, A.W., Rudd, M.K., Gustashaw, K., and Willard, H.F. (2001). Genomic and genetic definition of a functional human centromere. Science 294, 109–115. [DOI] [PubMed] [Google Scholar]
- Sun, X., Walstrom, J.M., and Karpen, G.H. (1997). Molecular structure of a functional Drosophila centromere. Cell 91, 1007–1019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wu, J., et al. (2002). A comprehensive rice transcript map containing 6591 expressed sequence tag sites. Plant Cell 14, 525–535. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wu, J., et al. (2004). Composition and structure of the centromeric region of rice chromosome 8. Plant Cell 16, 967–976. [DOI] [PMC free article] [PubMed] [Google Scholar]


