Breaking the actin treadmill [with video]

A mathematical analysis of two distinct microscopic experiments, which have been thought to contradict each other, provides a quantitative description of actin turnover and recycling mechanisms at the leading edge of crawling cells.

HFSP Young Investigator Grant holders Dimitrios Vavylonis and Naoki Watanabe and colleagues
authored on Tue, 12 March 2013

Animal cells crawl by extending lamellipodia, thin sheet-like structures that protrude from the body of the cell. Lamellipodia are made out of the protein actin that assembles into filaments, which are then cross-linked with one another to form a network structure. The thin geometry of the lamellipodium makes it ideal for microscopy and this is one of the reasons that it has been used as a model system to study how actin dynamics regulate a large variety of cellular processes such as neurite out growth, immune cell response and cancer metastasis. HFSP grantees Naoki Watanabe and Dimitrios Vavylonis study a fascinating aspect of the lamellipodium, namely its dynamic nature: as cells crawl, the actin fibers at the leading edge constantly assemble and disassemble, most of them within just 30 seconds.  

The process of actin filament network assembly and disassembly in the lamellipodium has been described as a treadmill: single actin proteins (called monomers or G-actin) polymerize into the filament network (F-actin) close to the cell membrane. At the back of the lamellipodium, the F-actin network disassembles into monomers. The monomers then recycle back to the leading edge by diffusion through the cytoplasm.  

But is the lamellipodium really analogous to a treadmill? Earlier experimental studies have provided apparently contradictory results.  Studies by Naoki Watanabe and collaborators who imaged single actin molecules had suggested that actin polymerization and depolymerization occurs throughout the lamellipodium and not just at the front and back. Other studies using photobleaching (where a part of the fluorescently-labeled actin in the lamellipodium is bleached to monitor the recovery of fluorescence) suggested a pure treadmilling process. However, a mathematical analysis to compare these two experiments was lacking.

To resolve this important issue, Smith et al. employed mathematical and computational modeling to compare these two types of experiment. Using the data from single-molecule studies as an input to the model, they calculated the predicted photobleaching recovery and compared the model results to photobleaching experiments that they performed under the same controlled setup. Their results showed that there is in fact no disagreement. As indicated by the earlier single-molecule studies, actin turnover is distributed in space throughout the lamellipodium. According to their calculations, actin polymerizes at a fast rate at the very front of the cell but the whole F-actin network is being remodeled throughout the lamellipodium.

A crucial aspect to the remodeling process is the diffusive motion of the disassembling pieces of the network, the result of actin filament severing. The results of the mathematical model of Smith et al. hint at an intriguing possibility: a big part of this diffusive population might be slowly diffusing actin oligomers. The authors called this species of actin “O-actin”.  The distinction between two forms of actin, namely F-actin and G-actin, has been recognized for a long time; for example, Fumio Oosawa, one of the founders of biophysics in Japan, described the critical concentration of the G- to F-actin transition in the late 50's. Oosawa’s experiments demonstrated that actin exists in the form of either G-actin or F-actin, and there is no intermediate state of actin in vitro. O-actin is still a type of polymer. But its different dynamic properties compared to the static F-actin network might represent a separate component of actin in cells. So far very little is known about this transient species. The study by Smith et al. motivates future experiments to visualize these diffusing oligomers that may hold many surprises.

Video: Experiment (left) and Monte Carlo simulation (right) of actin fluorescence recovery after photobleaching in the lamellipodium


Distributed Actin Turnover in the Lamellipodium and FRAP Kinetics. Smith MB, Kiuchi T, Watanabe N, Vavylonis D. Biophysical Journal 104:247-57 (2013)

Pubmed link