Douglas Koshland

Replicated chromosomes (sister chromatids) acquire three specific structural features that are important for chromosome segregation in mitosis. First, each sister chromatid has a centromere that mediates the attachment and the movement of chromosomes on the spindle. Second, sister chromatids are paired. Pairing is needed to establish a stable bipolar attachment of sister chromatids to microtubules of the mitotic spindle, which in turn ensures that sister chromatids segregate from each other during anaphase. Additionally, the dissolution of pairing appears to be a key event in governing the onset of chromosome segregation. Third, sister chromatids are condensed. This condensation helps to minimize entanglement of chromosomes while they move during mitosis, and shortens the chromosomes so that they are always displaced from the cytokinetic furrow at the end of mitosis.

Much of our understanding of mitotic chromosome function and structure has come from the study of vertebrate and invertebrate cells using cytology and established biochemical assays. More recently, studies using the budding yeast, Saccharomyces cerevisiae, have also begun to make significant contributions to the field. This simple eukaryote is amenable to extensive genetic analysis; and researchers have determined the specific DNA sequences required for the function of centromeres, telomeres, and origin of replications. With this knowledge it has been possible to manipulate the structure of endogenous chromosomes, as well as generate novel artificial chromosomes. Genetic and biochemical studies have also identified key genes encoding histones, topoisomerases, centromere proteins, and cell cycle regulators. Mutations in these genes provide important tools to manipulate both chromosomes and the cell cycle in vivo.

The Koshland group has developed three new assays for analyzing mitotic chromosomes in the budding yeast: an in vitro assay for centromere-microtubule interactions, a fluorescent in situ hybridization assay for following sister chromatid pairing and condensation, and a chromatin immunoprecipitation assay for monitoring in vivo binding of proteins to chromosomes at specific sites. Using these assays and yeast genetics, the group has characterized yeast centromere function in vitro, identified and characterized centromere components (Mif2p, Cse4p, and Cep3p), identified a potential nucleosome-based centromere core conserved from yeast to humans, and revealed both centromere movement and regulation during the cell cycle. Koshland's team has also identified and characterized two new classes of proteins important for sister chromatid pairing and condensation. The first is Pds1p: a cell cycle regulator important for the control of the metaphase/anaphase transition, for the exit of mitosis, and for cell cycle arrest in response to DNA and spindle damage. The second class, Smc proteins, is a family of chromosomal proteins, conserved from bacteria to humans that are essential in diverse eukaryotes for processes involving higher-order chromosome structure, including chromosome condensation, sister chromatid cohesion, dosage compensation, and recombination repair. Recently the scientists identified non-Smc subunits required for condensation and cohesion. The study of these proteins provided novel links between cohesion and condensation, and between cohesion and DNA replication. The team has also been able to map cohesion sites on chromosomes, demonstrating that the centromere nucleates the assembly of cohesion factors onto at least several kilobases of DNA flanking the centromere. Further analyses of these proteins and sites will provide insight into higher-order chromosome folding, as well as genome organization and stability.