Experimental Soft Condensed Matter Group
Harvard University, Prof. D. A. Weitz

Local Mechanical Properties of Cells and Cytoplasm

Cells are complicated microstructures --- how can we study local mechanics???

Cells are complicated! They can support huge stresses and strains even though they are mostly water, they can change shape and move, and have sophisticated transport mechanisms that allow them to move proteins, DNA and other molecules inside the cell and in and out of the cell membrane. We use novel particle tracking methods to measure the microscopic mechanical and rheological properties of these complex systems.

Model System: XENOPUS EGG EXTRACTS

We study extracted cellular cytoplasm from Xenopus (AKA FROG) Egg Extracts. This work is done in collaboration with Zach Perlman of Tim Mitchison's group at the Harvard Medical School. Frogs' eggs are pretty big -- 1 mm in diameter -- and they contain lots of cytoplasm. Using a high speed centrifugation technique, we can separate the cytoplasm from the lipid structures and organelles. We can then use micro- and macro-rheology techniques to study the mechanics of the cytoplasm. Using drugs and other chemical techniques we can try to isolate contributions of the different filaments to mechanical response. Additionally, the extracts retain a large portion of their biological activity, so we can also probe the impact of rheology and mechanics on mitotic spindle formation, transport mechanisms, and sol/gel transitions in the cell.

Microscopic Dynamics:

The cytoplasm is filled with proteins! Some are filamentous, such as actin, microtubules, and intermediate filaments, while many many others are in globular form, creating a crowded and dense background fluid, called the cytosol. We can investigate both the viscous background and the elastic polymer networks using mechanical probes that span a range of length scales from less than one micron to one millimeter!

In order to measure mechanical response on small length scales, we use microrheology techniques. From the thermal motions of embedded colloids, we calculate the mean-squared displacement (MSD) of the beads and use this to measure the local viscous and elastic modulus of soft materials. On left, we show the mean-squared displacements of 1-micron beads moving in Mitotic Xenopus Extracts. In this stage of the cell cycle, the actin undergoes a myosin-mediated contraction that causes a phase separation between actin-rich and actin-poor regions of the sample. To avoid this, we use cytochalasin to depolymerize the actin. Here we also added colchicine, which depolymerizes microtubules, to investigate their role in the micro-mechanics of this model cytosol. We observe little effect when we remove the microtubules, suggesting they do not play a large role. Also, the MSD increases nearly linearly with time, indicating a viscous solution (with little elastic response). We measure the viscosity to be roughly 10 mPa-sec (ten times that of water).

We can biochemically control the cell cycle stage of our extracts. Here we investigate the role of actin in the microscopic mechanical response of interphase extracts. We observe similar bead dynamics with and with out actin, and our data suggest that at micron length scales, the extracts are viscous, with a viscosity of roughly 10 mPa-sec.

Our data show that at micron length scales, regardless of cell cycle state or the polymerization state of the actin and microtubules, that the extracts are viscous fluids, with viscosities of roughly 10 mPa-sec. The factor of 10 increase over pure water suggests the crowding of globular proteins may be significant. Concentrated colloidal suspensions display an increase in viscosity with increasing volume fractions of spheres. Our data can be mapped onto a colloidal model, assuming a protein concentration in the extracts of roughly 30%, indicating that crowding is a excluded volume effect, and does not depend on the details of the protein-protein interactions.

Macroscopic Dynamics

At larger length scales and at room temperature, the extracts are elastic! Well separated biopolymer filaments do not contribute to the microscopic dynamics of embedded spheres or macromolecules, but do bear stress macroscopically. Here we show the frequency dependence of a native state interphase extract. The elastic modulus is dominant over the entire frequency range and is in the range of 10 Pa -- larger than what is expected for a purely entangled network, suggesting that cross-linking proteins are important.

We can investigate the viscosity of the extracts by keeping them cold (4 degrees Celsius), and measuring viscosity as a function of shear rate. We observe that the fluid shear thins, suggesting that there are large structures that break up at increasing shear rate. At high shear rates, the viscosity is roughly 10 mPa-sec, the same value we measure microscopically. Interestingly, we find that at short times (a few minutes) whatever structures are disrupted at high shear do not re-form. However, by waiting 40 minutes, we can recover the same initial flow curve, suggesting that at longer times the process is reversible. We do not understand the microscopic mechanism of this break-up and recovery.

We are still investigating the role of the actin, microtubules, and intermediate filaments at large length scales using rheology and pharmacological disruption techniques.

Also, check out the CIMS Cell Culture Microscope Facility, and other work by the Weitzlab Biophysics Division.

This page is maintained by:

Megan Valentine
Department of Physics
Division of Engineering and Applied Science
Harvard University
9 & 15 Oxford Street, McKay Laboratory
Cambridge, MA 02138
617-495-3705

valentin@fas.harvard.edu