Kinetochore specification

Read Also

Specification of the site of kinetochore formation on a chromosome is crucial for correct chromosome segregation. It ensures chromosome-microtubule attachment and places the kinetochore in a unique location on a chromosome (Cheeseman and Desai, 2008). Preventing formation of multiple kinetochores sites minimizes the chance that a single chromatid attaches to both spindle poles, which may lead to chromatid fragmentation. The discrete chromatin region that directs the formation of the kinetochore is the centromere. The centromere was originally identified as the primary constriction on a condensed chromosome, based on light microscopy observations. A single centromere region (monocentric chromosomes) is found in most eukaryotes. In C.elegans, the centromere extends along the entire sister chromatid (holocentric chromosomes; Maddox et al, 2004).

Although the function of the centromere is conserved, its sequence varies between species (Cleveland et al., 2003). In the budding yeasts the centromere is defined by a sequence of only 125 base pairs. The sequence is critical to the centromere function in budding yeast and kinetochore assembly depends on sequence-specific DNA binding proteins. In other eukaryotes the centromeres are more complex and can extend even up to 10 Mb. In organisms other than budding yeast, no precise DNA sequence defining the kinetochore site has been found (Sullivan et al., 2001). Instead, kinetochore assembly is thought to be epigenetically controlled (Karpen and Allshire, 1997; Henikoff and Dalal, 2004; Allshire and Karpen, 2008). It is the specialized centromere chromatin, rather than primary DNA sequence, that constitutes an epigenetic mark determining the kinetochore position.

The features of centromeric chromatin are common. The core part of the centromere consists of nucleosomes distinct from the rest of a genome, as they contain a unique and conserved variant of histone H3, Cenp-A (Centromere Associated Protein-A; Cse4 in S. cerevisiae, Cnp1 in S. pombe, CID in Drosophila), interspersed between the canonical histone H3 (Allshire and Karpen, 2008). Cenp-A containing region is flanked by heterochromatin (Gascoigne and Cheeseman, 2011). Both domains are important for the centromere function, but it is the Cenp-A-containing domain, that is the foundation of the kinetochore (Allshire and Karpen, 2008). Depletion of Cenp-A, leads to mislocalization of most of kinetochore components in all eukaryotes studied (Howman et al., 2000; Blower and Karpen, 2001; Liu et al., 2006). Contrarily, localization of Cenp-A is independent of most kinetochore proteins. In Drosophila, Cenp-A is both necessary and sufficient to mediate kinetochore assembly; its overexpression or artificial targeting to non-centromeric regions leads to formation of ectopic kinetochores (Heun et al., 2006; Mendiburo et al., 2011). Hence, Cenp-A is considered the main determinant of kinetochore assembly.

Over generations, the centromere is propagated at the same site. As Cenp-A level at the centromere is diluted during replication; its amount has to be restored by a newly synthesized protein (Allshire and Karpen, 2008). However, unlike canonical histones, tethering Cenp-A to the centromere is independent of replication and occurs in different cell cycle stages for different organisms. While in S. pombe Cenp-A assembles in S or G2 phase (Takayama et al., 2008), in human cells it is deposited during telophase/G1 phase (Jansen et al., 2007). In Drosophila embryos Cenp-A homologue, CID, becomes incorporated during anaphase, while in cultured cells it is was shown to be recruited in metaphase (Schuh et al., 2007; Mellone et al., 2010).

The mechanism of Cenp-A deposition into the centromeric chromatin is poorly understood and likely requires a multistep regulation (Allshire and Karpen, 2008). It is thought to use pre-existing Cenp-A to deposit new Cenp-A (Phansalkar et al., 2012). HJURP/Scm3 has been recognized to deliver newly synthesised Cenp-A into Cenp-A centromeric domain in humans and yeast (Dunleavy et al., 2009; Pidoux et al., 2009). Cal1 has been suggested a functional equivalent of HJURP/Scm3 in Drosophila (Mellone et al., 2010; Phansalkar et al., 2012l; Chen et al., 2014). Cal1 is required for localization of CenpA/CID and low amount of Cal1 at centromeres also limits the amount of incorporated CenpA/CID (Erhardt et al., 2008; Schittenhelm et al., 2010). Cal1 has also been shown to be sufficient to assemble Cenp-A/CID nucleosomes, which directs kinetochore formation (Chen et al., 2014). In Drosphila culture cells, Cal1 precedes Cenp-A/CID deposition on centromere and interacts directly with pre-nucleosomal CenpA/CID (Mellone et al., 2010; Schittenhelm et al., 2010). Cal1 lozalization to centromeres also depends on Cenp-A/CID and a proteins associated with Cenp-A, Cenp-C (Erhardt et al., 2008; Schittenhelm et al., 2010). Thus a possibility that Cal1 acts to receive new Cenp-A/CID on chromatin dependently on pre-existing CenpA/CID (Mellone et al., 2010).