THREE-DIMENSIONAL STRUCTURE AND FUNCTION OF THE MAMMALIAN KINETOCHORE

点击翻译按钮获取中文摘要

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. ABSTRACT The mammalian kinetochore is a proteineous complexes that perform at least four functions that are vital for accurate chromosome segregation during mitosis: 1) attaching chromosomes to the mitotic spindle; 2) controlling the dynamics of kMTs; 3) generating force for chromosome alignment; and 4) generating a cell cycle checkpoint that delays anaphase onset until all chromosomes are attached to the mitotic spindle and aligned at the spindle equator. Intricate interactions between kinetochores and microtubules (MTs) are essential for all of these vital processes. Despite progress in identifying molecular components of the kinetochore, the underlying mechanisms of kinetochore function and its interactions with MTs remains largely obscure. This is in part due to the paucity of data concerning the 3D ultrastructure of the kinetochore. The only electron tomographic study was performed by the project PI over 10 years ago with chemically fixed specimens (McEwen et al, 1993). In 1998, McEwen and colleagues demonstrated a dramatic improvement in preservation of kinetochore ultrastructure using high-pressure freezing and freeze-substitution, but very little electron tomography was used in that study. Our long-term objective is to establish a high-quality, high-resolution structural model of the mammalian kinetochore in its unbound state and to determine structural changes that occur when key kinetochore components are removed. The next step is to refine our current structural model by reconstructing the kinetochore in the frozen-hydrated state. During the past year we have developed protocol that yield a high density of well-frozen mitotic cells and we are using high pressure-frozen specimens that have been freeze-substituted. This strategy allows a relatively quick and straightforward method of assessing specimen preparations and optimizing protocols before turning to the more technically demanding task of electron tomography of frozen-hydrated sections. The freeze-substituted material will also provide electron tomographic data sets with a higher signal-to-noise ratio and better quality high tilt data. We now have reproducible protocols for obtaining a high percentage of HeLa cells in mitosis using a procedure call mitotic shake off. Consistently good freezing was obtained using 15% BSA as a cryo protectant. We are currently confirming the reproducibility of the result and preparing cells for frozen-hydrated ultramicrotomy. In 2002 we began collaborating with the laboratory of Dr. Qiang Ji at RPI to develop automated tools for segmenting microtubules and their plus-ends from tomographic reconstructions of mammalian kinetochores. Motivation for this project came from our work to classify the plus ends of kinetochore microtubules according to their structural conformations (referred to above as the PtK project). These conformations are indicative of the dynamic state of the microtubules and knowledge of how the kinetochore controls microtubule dynamics is a critical issue in cell biology. Our general approach is to use the high throughput capabilities of modern electron tomography to collect a large enough date base of kinetochore microtubule plus ends to perform statistically meaningful analyses on how the conformations change with stage of mitosis, treatment with pharmacological agents, and knockdowns of key kinetochore molecular components. We are also interested in determining how well coordinated the plus-end conformations are on individual kinetochores. We would prefer to use computer classification methods to sort the conformations, but the reconstructions are too noisy and there is too much background structure to permit classification on kinetochore microtubules extracted directly from the raw tomographic reconstructions. There are also issues of variation due to orientation relative to the missing pyramid. To overcome these problems, we sought to develop an efficient automated segmentation method that would extract the microtubules, including their plus ends, from large numbers of tomographic reconstructions. Dr. Ji is an expert on computer vision. His graduate student, Ming Jing, has developed a multi-step segmentation method that uses the known cylindrical geometry of microtubules, and the range of feasible curvatures found at microtubule plus ends, as constraints on the segmentation process.