UC Davis

The centrosome functions as the main microtubule-organizing center in metazoan cells and orchestrates diverse cellular processes critical for development, including cell division, polarization, and migration. Centrosome dysfunction is associated with cancer and microcephaly in human, warranting a mechanistic understanding of its assembly in vertebrates. The centrosome is a membraneless organelle that consists of a pair of centrioles embedded in a proteinaceous network named pericentriolar material (PCM). The PCM is a complex ensemble of proteins enriched with coiled-coil domains (CCs) and low-complexity regions (LCRs). At the onset of mitosis, centrosomes expand the PCM to maximize their microtubule nucleation capacity. This process, termed centrosome maturation, ensures proper mitotic spindle assembly and chromosome segregation in mitosis. However, as the centrosome expands, how PCM proteins are recruited and held together without membrane enclosure remains elusive. This dissertation work primarily focuses on understanding the functional contribution of pericentrin (PCNT), a conserved PCM component, to centrosome assembly in vertebrates. I found that endogenously expressed PCNT condenses into dynamic aliphatic alcohol-sensitive granules around centrosomes during late G2/early mitosis in human cells. These data suggest that full-length PCNT may undergo liquid-liquid phase separation (LLPS) in physiologically relevant conditions during centrosome maturation. Furthermore, the N-terminal and middle segments of PCNT, enriched with conserved CCs and LCRs, undergo concentration-dependent condensation. Formation of PCNT “condensates” exhibits characteristics of LLPS, including a sharp phase transition at a concentration threshold, as well as coalescence, deformability, and rapid fluorescence recovery after photobleaching. Moreover, these PCNT condensates selectively recruit endogenous PCM components and nucleate microtubules in cells. We propose that CCs and LCRs, two prevalent sequence features in the centrosomal proteome, are preserved under evolutionary pressure in part to mediate LLPS, a process that bestows upon the centrosome distinct properties critical for its assembly and functions. Another part of the dissertation centers on developing computational methods for image analysis to study centrosome assembly in a quantitative manner. To accurately detect and quantify centrosome-localized mRNAs and proteins at single cell resolution, I developed a streamlined workflow for performing, acquiring, and analyzing sequential immunofluorescence and single-molecule RNA fluorescence in situ hybridization data via 3D reconstruction in Imaris, followed by quantifications using MATLAB and R. Moreover, to automatically track the dynamics and spatiotemporal distribution of condensates relative to centrosomes, I developed TRACES (Tracking of Active Cellular Structures), a customizable and open-source pipeline capable of detecting, tracking, and quantifying fluorescently labeled cellular structures in up to three spectral channels simultaneously at single-cell resolution. Overall, my studies provide a molecular and biophysical framework, as well as quantitative image analysis methods, for dissecting mechanisms underlying centrosome assembly in vertebrates.