Enhancing togetherness: kinetochores and cohesion.

All organisms must correctly copy and equally distribute their genetic information each and every time a new cell is created. In eukaryotic cells, this action involves many different coordinated processes, including DNA replication, kinetochore and spindle formation, and chromosome cohesion. If errors occur in chromosome segregation, a chromosome may be lost or gained, resulting in an aneuploid cell. Aneuploidy is associated with cancer in humans, as well as a variety of developmental disorders including Downs syndrome. Because all organisms require the faithful transmission of their genetic information, many of these processes and proteins are evolutionarily conserved. Thus, gaining a basic understanding of the basic mechanisms that contribute to the accurate segregation of chromosomes is essential for understanding the underlying cause of some human diseases. The kinetochore is a complex multiprotein structure that forms at centromeres (for recent reviews, see McAinsh et al. 2003; Meraldi et al. 2006). The basic function of the kinetochore is to attach chromosomes to microtubules. Once attached, microtubules can then pull chromosomes to opposite spindle poles, thus enabling the physical separation of chromosomes into two cells. Kinetochores can consist of 70 proteins or more, which are often divided up into three layers: inner, central, and outer. The inner proteins are associated with the DNA and are linked to the outer layer by the central layer. The outer kinetochore proteins attach to microtubules and associate with several checkpoint proteins that monitor various aspects of kinetochore and spindle formation. Accurate attachment of kinetochores to microtubules is referred to as amphitelic attachment and refers specifically to the case in which one sister kinetochore is attached to microtubules emanating from one spindle pole and the second sister kinetochore is attached to microtubules from the other spindle pole (Fig. 1A). This situation is often referred to as biorientation. In some cases, the attachment between kinetochores and microtubules can be defective. The attachment can be monooriented, meaning that microtubules from only one of the two poles are involved in the attachment. There are two different cases of this: syntelic and monotelic. In syntelic attachment, both sister kinetochores are attached to microtubules from the same pole (Fig. 1B). In monotelic attachment, one of the kinetochores may be attached to microtubules from one pole and the other kinetochore is unattached (Fig. 1C). In a merotelic attachment, one kinetochore may be attached to microtubules from both poles. This occurs in organisms that have multiple microtubules attached to a single kinetochore. Defects in kinetochore attachment may lead to the activation of the spindle checkpoint. All eukaryotic cells studied to date have a mitotic spindle checkpoint. This refers to the fact that defects in the mitotic spindle trigger a cascade of events that lead to cell cycle delay or arrest. Defects that activate the checkpoint likely include both (1) a lack of kinetochore– microtubule attachment and (2) a lack of tension between sister kinetochores (Fig. 1D; for a recent review, see Pinsky and Biggins 2005). Despite ongoing debate, there is a good body of evidence accumulating that lack of tension can trigger the spindle checkpoint in budding yeast. The protein with the best evidence of being involved in tension sensing is Aurora kinase B, or Ipl1 in budding yeast (Biggins and Murray 2001; Stern and Murray 2001). In the presence of functional Ipl1, defective kinetochore–microtubule attachments are detached, presumably to allow for an increased chance at forming correct attachments (Tanaka et al. 2002; Pinsky et al. 2006). An Ipl1 mutant (ipl1-321) displays 85% chromosome missegregation following a G1 arrest (Biggins et al. 1999). These missegregation events in many cases appear to be due to syntelic attachments since sister chromatids are observed segregating to the same spindle pole (Biggins et al. 1999; Tanaka et al. 2002). Since syntelic attachments do not generate tension, this Ipl1 mutant will lack tension. In mitotic rat PtK cells, the kinetochore is attached to an average of 24 microtubules each and in grasshopper spermatocytes, the average is 32 (McEwen et al. 1997; King and Nicklas 2000). The situation in budding yeast is considerably simpler. Each kinetochore attaches to a single microtubule (Winey et al. 1995). Thus, each kinetochore exists in one of two states, attached or unattached, and there can be no partial microtubule occuCorrespondence. E-MAIL [email protected]; FAX (816) 926-2094 Article is online at http://www.genesdev.org/cgi/doi/10.1101/gad.1523107.