 |
|
|
|
|
Back to 2005 Research
Dyche Mullins
Physiology Course 2005 – Projects
1. Mechanics of motile actin networks: Analysis of speckle fluctuations. What are the mechanical properties of dynamic actin networks generated under load? How do these properties depend on the concentrations and activities of the factors that construct the networks? To address these questions we analyzed thermal fluctuations of fluorescent speckles in motile actin networks generated in vitro using purified components of the actin-based motility machinery. We used low dopings of fluorescent actin derivatives as well as phalloidin-conjugated quantum dots to generate fiducial fluorescent marks throughout the actin network. We then performed high-speed fluorescence microscopy to visualize thermal fluctuations of the fluorescent speckles. We wrote Matlab scripts to analyze the positional fluctuations determine the mechanical properties of the actin network as a function of both space and time. We found that networks generated by the Arp2/3 nucleatin/crosslinking factor in the absence of any other crosslinker were very much more compliant (~10 Pa) than networks generated in cell extracts (~1000 Pa) and that the compliance of a newly-formed piece of the network remains constant until cofilin-mediated depolymerization decreases the actin network density significantly. The mechanical properties of the network appear to fail catastrophically in a manner that may be explained by percolation theory.
2. Bacterial shape and cytoskeletal organization: What is the connection? Rod-shaped bacteria require actin-like filaments, composed of a protein called MreB, to maintain their shape. We investigated the mechanism by which the polymer controls cell shape. We first performed photobleaching experiments on E. coli expressing fluorescently-tagged MreB and found that the cytoskeleton is composed of two filament populations, one that exchanges on a 30 second time scale and another that is significantly more stable. We depolymerized MreB filaments in E. coli using the small-molecule inhibitor A22 and performed time-lapse microscopy on the rounding cells. The change in cell shape occurs only after several cell divisions and is accompanied by: (1) asymmetric cell division, (2) a decrease in cell size and (3) a significant decrease in cell wall rigidity. After the initial decrease in size the spherical cells now become quite large. We also looked at the oscillation of fluorescently-labeled MinD protein and found that cells treated acutely with A22 had no problem determining the proper plane of division. Acute loss of MreB, therefore, appears to have no effect on cell shape or the ability to perform symmetric cell division. The defects in cell shape are due to downstream effects, probably involving cell wall synthesis machinery.
3. Polymer dynamics of a prokaryotic actin. ParM is a plasmid-encoded actin whose assembly segregates low-copy R1 plasmids to opposite poles of rod-shaped cells. A plasmid DNA locus, parC, and a repressor protein that binds to it, ParR, are also required for efficient segregation. To better understand the nature of the interaction between the polymer and the DNA/protein complex we coupled ParR/parC complexes to individual quantum dots and immobilized them on a glass surface. We then used Total Internal Reflection Fluorescence (TIRF) microscopy to monitor dynamics of ParM polymers attached to individual quantum dots. We found that ParM filaments attach end-on to ParR/parC complexes and elongate by insertion of monomers at the attached filament end. Attached ends elongated preferentially compared to unattached ends and were protected from catastrophic depolymerization. These effects combined to produce a population of long, stable filaments not observed in the absence of ParR/parC. Together these results indicate that the ParR/parC complex functions like a kinetechore or like a formin-family protein to enable insertional polymerzation and to harness filament elongation to generate force and directed motion.
4. Mathematical modeling of plasmid segregation by ParM. Polymerization of the plasmid-encoded actin ParM pushes drug-resistance plasmids to opposite poles of rod-shaped bacteria. We previously proposed a model for plasmid segregation in which the ends of spontaneously nucleating ParM filaments search the bacterial cytoplasm until they capture plasmids. Elongation of a ParM filament will generate force to push the attached plasmids to opposite poles of the cell. Together with Jonathan Alberts, Garrett Odell, and Francois Nedelec we constructed numerical simulations of this model and tested their ability to capture and segregate plasmids. To decrease the number of degrees of freedom, we described nucleation and polymerization using measured parameters. The simulations indicate that our proposed mechanism can produce rapid plasmid capture and segregation. In addition they suggest that capture is a multi-step process in which monovalent attachment of ParM filaments can generate multiple rounds of plasmid translocation which ultimately cluster multiple plasmids at one end of the cell. This clustering produces a very high local concentration of plasmids and promotes more efficient bivalent polymer attachment.
|
|
|
 |
|
|
|
