A-kinase anchoring protein AKAP95 is implicated in mitotic chromosome condensation by recruiting the condensin complex. Here, we report a differential regulation of condensation of maternal and paternal chromosomes mediated by AKAP95 and chromatin composition in mitotic mouse zygotes. AKAP95 is synthesized upon oocyte activation, targeted to the female pronucleus and specifically associates with maternal chromosomes at mitosis. Peptide competition and rescue experiments show that AKAP95 is required for recruitment of the mCAP-D2 condensin subunit to, and condensation of, maternal chromosomes. In contrast, AKAP95 is dispensable for mCAP-D2 targeting and condensation of paternal chromosomes. In vitro nuclear reconstitution and disassembly assays indicate that human hCAP-D2 targets protamine-containing chromatin independently of AKAP95, but requires AKAP95 for association with histone-containing chromosomes. We propose a concept whereby (1) recruitment of condensins to chromatin is affected by chromatin composition and (2) AKAP95 renders histone-containing chromatin permissive to condensin targeting.

The mitotic apparatus is composed of a complex, seemingly chaotic, but wonderfully organized compilation of cytoskeletal elements and motor proteins. Its task is to assure that each daughter cell receives a single copy of the genome. Microtubules are the biological polymers that make up the bipolar mitotic spindle, serving as tracks to separate sister-chromatids to opposite sides of the dividing cell. Understanding the behavior of microtubules during mitotic entry and exit is extremely important for diagnosing possible defects of atypical or failed mitosis. This research uses the advantages of GFP technology to investigate the dynamics of microtubules throughout the cell cycle. Cells permanently expressing GFP-α-tubulin were generated using lipid transfection and drug selection. We show that GFP-tubulin had little effect on the behavior and role of microtubules; therefore, we conclude that cells expressing GFP-α-tubulin are a reliable method for studying microtubule behavior. Live imaging of an epithelial cell line (LLCPK1) stably expressing GFP-α-tubulin revealed a microtubule bundling phenomenon that stabilizes microtubules during prometaphase. Further analysis shows that these bundles are pulled into the forming spindle in a dynein dependent manner. Our final model is an extension of the search and capture model, where we suggest that centrosomal microtubules also capture peripheral microtubule bundles during spindle assembly. During mitotic exit, the process of moving microtubules inward is reversed, as microtubules are subject to outward forces during anaphase. Three main events are documented following anaphase onset: microtubule elongation, microtubule release, and spindle pole fragmentation. These events serve to populate regions of the cell that were devoid of microtubules during metaphase. Evidence is also presented in support of how microtubules might specify the location of the contractile ring. We show that microtubules at the equator target the cortex more frequently than elsewhere in the cell, suggesting a faster rate of delivering furrow-initiation factors. We use numerous GFP fusion proteins that allowed us to tackle our biological question from several angles, which wouldn't have been possible to answer otherwise. In summary, my work provides new insight into the behavior of microtubules during the assembly and disassembly of mitotic spindles in mammalian somatic cells.

Eukaryotic cells contain scores of small nucleolar RNAs (snoRNAs), which are required for maturation of pre-rRNA. Two large snoRNA families exist defined by vital box C/D and box H/ACA motifs. The goal of the present study was to gain new insights into the structure and biogenesis of the box C/D snoRNAs; the knowledge developed from this effort was then recruited for practical applications. The investigation was conducted with the phylogenetically conserved U14 and U3 box C/D snoRNAs, from the yeast Saccharomyces cerevisiae. The specific aims included: (1) identification of cis-elements sufficient for biogenesis of the U14 snoRNA; (2) development of a functional map for the U3 snoRNA, and; (3) development of a U3-based model ribozyme system for in vivo studies. Conclusions derived from the U14 biogenesis studies are: (1) production of U14 involves ordered folding of the precursor RNA, and this step is required for formation of the vital box C/D structure motif, and; (2) the active box C/D motif, which is now predicted to consist solely of the box C and D elements, is necessary and sufficient for both accumulation and targeting RNA to the nucleolus. A general model for box C/D snoRNA biogenesis is proposed. Functional mapping of U3 revealed that: (1) boxes C$\sp\prime$ and D and flanking helices are critical for U3 accumulation; (2) boxes B and C are not essential for U3 production, but are important for function, due most likely to binding of a trans-acting factor(s); (3) the 5$\sp\prime$ portion of U3 is required for function, but not stability, and; (4) the non-conserved hairpins, which account for 50% of the molecule, are not required for accumulation or function. Based on the knowledge obtained with U14 and U3, a model ribozyme system featuring chimeric U3:ribozyme RNAs, or "snorbozymes", was developed and tested in vivo. Remarkably, the cleavage efficiency by a hammerhead ribozyme, both in cis- and in trans-configurations, appears quantitative! Other advantages of the system are: (1) a final product is stable, and; (2) authentic in vivo cleavage can be easily distinguished from artifactual cleavages. Snorbozymes are predicted to be useful for targeting natural transcripts in any eukaryotes, for fundamental research or practical applications.