QB3 Postdoc Seminar
Seminar | January 18 | 4:30-5:30 p.m. | 177 Stanley Hall
Speaker 1: Lisa Alexander (Bustamante lab)
Kinetic analysis of cotranslational protein folding via single molecule techniques
Multidomain proteins may have altered folding kinetics on the ribosome compared to in solution. Using optical tweezers, we compare the folding of truncated and full-length isolated versions of a multidomain calcium-binding protein with stalled, ribosome-bound nascent chain counterparts and with the polypeptide as it is synthesized in real-time. In solution, a truncation with a partial C-terminal deletion adopts a misfolded state that is not seen in the full-length protein. Formation of this misfolded state is slowed and its rate of unfolding is increased on the ribosome, although it still folds on a timescale similar to the translation rate. However, during active translation, this state is only attained after a long delay, indicating that the growing polypeptide is not equilibrated with its ensemble of accessible conformations during translation, as previously thought. Slow equilibration on the ribosome can delay folding until adequate sequence is available and/or allow time for binding of chaperones, thus promoting productive folding.
Speaker 2: Matthew Akamatsu (Drubin lab)
Principles of self-organization by the branched actin cytoskeleton during mammalian clathrin-mediated endocytosis
During clathrin-mediated endocytosis (CME), a cells plasma membrane is deformed from a flat sheet into a round vesicle to internalize transmembrane proteins and extracellular cargo. The polymerization of actin filaments contributes to mammalian CME, but the molecular mechanism is not understood.
We used mathematical modeling and quantitative fluorescence imaging of genome-edited stem cells to investigate how actin organizes to produce force at sites of endocytosis. Using GFP-tagged nanocages of defined copy number, we developed an intracellular calibration method to relate fluorescence intensity to numbers of molecules in live mammalian cells. This allowed us to count the number of molecules of the branched actin filament nucleator Arp2/3 complex at the endocytic sites of human induced pluripotent stem cells. These measurements, coupled with known rates and positions of endocytic proteins, constrained a mathematical model of actin polymerization coupled to the internalization of an endocytic site. Stochastic simulations of the model revealed that endocytic actin self-organizes at endocytic sites in a dendritic cone that grows toward the base of the pit. Surprisingly, long actin filaments bent between their attachment sites at the coat and the base of the pit, which we confirmed with cryo-electron tomography of actin in mammalian cells. We suggest that self-organization and bending of actin filaments allows the actin network to robustly internalize cellular membranes across a range of physical constraints.