Protein pumps are crucial in modern cells. They use the energy from light or food to move ions across a membrane, creating an electrical gradient. These gradients act as a kind of battery, driving cellular processes.

Unlike the sophisticated proteins that make up pumps in modern cells, the pumps of "protocells" must have formed from simple proteins present on ancient Earth. What's more, these pumps must have assembled themselves.

A team from NASA's Center for Computational Astrobiology in California, led by Andrew Pohorille, has now simulated the formation of such pumps, using a protein called M2 from the human flu virus.

Though it's highly unlikely that M2 was used by protocells, similar proteins may have been involved. "We really don't care all that much for the specific identity of the actors," says Pohorille, "we want to know if we can understand the play."

He and his colleagues simulated what happens to M2 molecules when they interact with simple membranes made of a mixture of organic substances such as carboxylic acids. Such membranes can spontaneously form vesicles reminiscent of cells ( New Scientist, 12 September 1998, p 30).

The M2 protein has a water-loving backbone and oily, water-hating side chains. In water, the protein is held open. When it interacts with a membrane, however, the protein folds into an alpha helix.

According to the computer simulation, four of these helices then bond to each other, forming a channel on the inside, and the whole package inserts itself into the membrane to escape the water.

Within the channel, parts of the proteins bond to form a gate that blocks most ions. However, hydrogen ions captured at the outside of the gate are rapidly conducted through the proteins by a series of chain reactions that eventually spews hydrogen ions into the interior. This process causes protons to accumulate inside the vesicle, creating an electrical gradient.

David Deamer, a biophysicist at the University of California at Santa Cruz, says this type of modelling could help researchers create lab versions of protocells in the near future.

"For the first time we are in a position that we can do it," agrees Pohorille, whose team last week presented its findings to an American Physical Society meeting in Minneapolis.

This article appeared in the April 1 issue of New Scientist New Scientist. Copyright 1999 - All rights reserved. The material on this page is provided by New Scientist and may not be published, broadcast, rewritten or redistributed without written authorization from New Scientist.