How do amoebae swim and crawl?

The surface behaviour of swimming amoebae was followed in cells bearing a cAR1-paGFP (cyclic AMP receptor fused to a photoactivatable-GFP) construct. Sensitized amoebae were placed in a buoyant medium where they could swim toward a chemoattractant cAMP source. paGFP, activated at the cell's front, remained fairly stationary in the cell's frame as the cell advanced; the label was not swept rearwards. Similar experiments with chemotaxing cells attached to a substratum gave the same result. Furthermore, if the region around a lateral projection near a crawling cell's front is marked, the projection and the labelled cAR1 behave differently. The label spreads by diffusion but otherwise remains stationary in the cell's frame; the lateral projection moves rearwards on the cell (remaining stationary with respect to the substrate), so that it ends up outside the labelled region. Furthermore, as cAR1-GFP cells move, they occasionally do so in a remarkably straight line; this suggests they do not need to snake to move on a substratum. Previously, we suggested that the surface membrane of a moving amoeba flows from front to rear as part of a polarised membrane trafficking cycle. This could explain how swimming amoebae are able to exert a force against the medium. Our present results indicate that, in amoebae, the suggested surface flow does not exist: this implies that they swim by shape changes.