<p>Many cells spontaneously establish a polarity axis even in the absence of directional cues, a process called symmetry breaking. A central question concerns how cells polarize towards one, and only one, randomly oriented "front". The conserved Rhotype GTPase Cdc42p is an essential factor for both directed and spontaneous polarization in various organisms, whose local activation is thought to define the cell's front. We previously proposed that in yeast cells, a small stochastic cluster of GTP-Cdc42p at a random site on the cortex can grow into a large, dominating cluster via a positive feedback loop involving the scaffold protein Bem1p. As stochastic Cdc42p clusters could presumably arise at many sites, why does only one site become the dominating "front"? We speculated that competition between growing clusters for limiting factors would lead to growth of a single winning "front" at the expense of the others. Utilizing time-lapse imaging with high spatiotemporal resolution, we now document initiation of multiple polarized clusters that competed rapidly to resolve a winning cluster. Such multicluster intermediates are observed in wild-type yeast cells with functional directional cues, but the locations where they are initiated are biased by the spatial cues. In addition, we detected an unexpected oscillatory polarization in a majority of the cells breaking symmetry, in which polarity factors initially concentrated very brightly and then dimmed in an oscillatory manner, dampening down to a final intermediate level after 2-3 peaks. Dampened oscillation suggests that the polarity circuit contains an in-built negative feedback loop. Mathematical modeling predicts that negative feedback would confer robustness to the polarity circuit and make the kinetics of competition between polarity factor clusters relatively insensitive to polarity factor concentration.</p><p>We are trying to understand how competition between clusters occurs. We find that the yeast guanine-nucleotide dissociation inhibitor (GDI), Rdi1p, is needed for rapid competition between clusters. In the absence of Rdi1p the initial clustering of polarity</p><p>factors is slowed, and competition is also much slower: in some cases cells still have two clusters at the time of bud emergence and they form two buds. We suggest that in the absence of Rdi1p, the clusters compete for a limiting pool of Cdc42p, and that slow</p><p>exchange of Cdc42p on and off the membrane in the absence of Rdi1p leads to slow competition.</p> / Dissertation

No description available.

No description available.