Dr. Anne Bernheim-Groswasser
In- vitro studies of cell cytoskeleton, active polymer networks (actin) and cell motility

Eukaryotic cells move by a complex mechanism of extension and retraction of their actin cytoskeleton network. During locomotion, cells extend protrusions at their leading edge, i.e., Lamellipodia and Filopodia. Endosomes, vesicles and some pathogens move in the cell by a similar molecular mechanism. Despite many advances in identifying the biochemical basis of actin-dependent movement, the precise mechanism of force generation remains unclear. The group of Dr. Bernheim-Groswasser is concerned with elucidating the different aspects of these phenomena especially as related to the role of biopolymers, their self aggregation, polymerization and depolymerization, network formation and effect of surfaces as related to cell motility and deformation. This work is carried out in a unique laboratory dedicated for the study of biophysics of the cell cytoskeleton.

1. In vitro study of the cell cytoskeleton properties
We focused on the bulk properties of actin assembly analyzing the role of the proteins known to control the actin cytoskeleton reorganization. More specifically, we studied the role of Arp2/3 complex, Fascin and capping proteins (CP) by gradually changing their ratio in the motility medium. We show that in vitro bulk polymerization of actin in the presence of Arp2/3 complex and fascin results in spontaneous formation of asters (Fig. 1) and stars (Fig. 2) of actin filaments, similar actin filament stars were recently observed in vivo [M. R. Mejillano et al. Cell 118, 363 (2004).]. Based on these observations we proposed a model for explaining the spontaneous self-assembly of asters and stars, this model is also relevant for understanding the transition from lamellipodia to filopodia (Lior Haviv et al. manuscript in preparation).

Effect of surface proximity. In parallel to the work conducted in the bulk, we started to build the experimental set-up which will be used for studying the role of surface proximity (mimicking the role of the cell membrane) on the system properties. For that purpose we set up the electrofomation technique in the lab to prepare giant unilamellar vesicles. In addition, we purified GST-WASP and GST-cdc42 proteins (plasmids were kindly given by Prof. W. Lim, from the Dept. of the Biochemistry and Biophysics, Univ. of California, San Franscisco).

2. Self-organization properties of active filament-motor systems:
We investigated the role of Myosin II/actin ratio on the reorganization pathway. The role of Fascin (bundling protein) was also elaborated. We were able to show that myosin II can reorganize actin filaments into different microstructures, depending on myosin II concentration. Formation of spiral, branched networks (Fig. 3) and asters (Fig. 26) were observed experimentally. These observations are in accord with a theoretical model published recently by our collaborators [K. Kruse et al. PRL 92, 078101 (2004)] (Prof. F. Julicher and Dr. Karsten Kruse, Max-Planck-Institute for Physics of Complex Systems, Biological Physics Dept., Dresden).

 

Fig 1. Diffuse aster-like structures are formed in bulk in the presence of Arp2/3 actin only

 

Fig 2. The presence of Fascin in addition to Arp2/3 complex and actin results in the and formation of stars (bar = 10?m)

 

Fig. 3. Formation of network of asters by myosin II molecular motors

 

Fig. 4. Large individual asters are also formed

3. Dynamics of active polymer networks
The study of dynamical properties of F-actin (filamentous or polymeric form of actin) dilute solutions was conducted using FCS (fluorescence correlation spectroscopy) technique. Using this technique we performed non-invasive measurements of the stochastic motion of F-actin monomers (determined by us as the elementary unit of the system, i.e., short F-actin element) in dilute solutions [A. Bernheim-Groswasser, R. Shusterman, and O. Krichevsky, Monomer dynamics in F-actin filaments (submitted to Phys. Rev. Lett.)]. Fluorescent labels were placed specifically at the ends of actin filaments. Their motion was then monitored with the help of the FCS technique. This approach allows measuring the monomers' mean-square displacements (MSD) as a function of time in a wide range of timescales, from 20us to 2s (Fig.5). Over four decades in time the monomers' MSD follows ~t3/4 power law dependence in qualitative agreement with the current theories of semiflexible polymer dynamics. Moreover, the experimental results agree quantitatively with the predictions of two theories [R. Granek, J. Phys. II (France) 7, 1761 (1997); K. Kroy and E. Frey, Phys. Rev. E. 55, 3092 (1997)], in the appropriate timescales.

Fig 5. The kinetics of random motion MSD <r2(t)> of actin filaments' ends as function of time.Experimental measurements (blue points) are compared to the theoretical predictions by Kroy and Frey (dashed red line) and Granek (solid green line).