BioWorld International Correspondent

LONDON - A ring of bacterial proteins that punches a massive hole in human cells could be the target for new vaccines and therapies to treat pneumonia, meningitis and ear infections.

The proteins change from an innocuous soluble form into a ring of lethal spokes that stand on the surface of the cell before suddenly doubling over - rather like a ring of people all crashing their heads into the ground at once.

The effect, said Helen Saibil, professor of structural biology at Birkbeck College in London, is to dent the membrane of the host cell, and ultimately perforate it.

"It is fascinating how these proteins manage to change character so suddenly, and undergo such a gigantic change in shape," she told BioWorld International.

Saibil, together with collaborators at the University of Oxford and the University of Leicester, has been studying how the molecules of the toxin pneumolysin, which is produced by the bacterium Streptococcus pneumoniae, form a ring in order to puncture the host-cell membrane. About 20 other types of bacteria are thought to use the same or a similar strategy to attack host cells, so the findings could help scientists develop vaccines or drugs to treat a range of diseases.

"This toxin is one of the main virulence factors of S. pneumoniae. If we could knock it out, this could greatly reduce the clinical effects of the bacterial infection. In addition, because it is highly conserved and does not vary greatly, this route of therapeutic attack could get around the problems of antibiotic resistance," Saibil said.

The work is reported in the April 22, 2005, issue of Cell in a paper titled "Structural Basis of Pore Formation by the Bacterial Toxin Pneumolysin."

Many bacteria strike their human hosts by releasing toxins that attack human cells while leaving their own untouched. The molecules of pneumolysin released by Streptococcus pneumoniae, for example, bind to cholesterol, so they readily stick to the membranes of human cells, which contain plenty of cholesterol, but not to bacterial cells, which have no cholesterol.

Once on the host-cell membrane, the pneumolysin molecules assemble into a ring of about 40 molecules. Saibil and her collaborators used cryo-electron microscopy to study those proteins on the surface of liposomes.

She said: "Initially the ring sits on the surface of the membrane like a crown. But then these innocuous soluble water-loving proteins suddenly change their character and become hydrophobic membrane proteins capable of piercing the cell membrane."

The researchers found that there was a dramatic change in the shape of the proteins, from long stick-like forms to doubled up hairpin-shaped arches.

The pore that forms is "really gigantic," Saibil said. "If you wanted to kill a cell by making a hole in it, you just need to make the hole big enough for calcium ions to flood in. But this is 10 to 15 times bigger and it's not clear to us why it has to be so huge."

Analysis showed that the change in shape came about when two regions of alpha-helices, which were coiled up in the water-soluble form of pneumolysin, became stretched out straight in beta-sheet form.

The group is investigating how to use the new information to help design vaccines or therapies against S. pneumoniae. Peter Andrew, professor of microbial pathogenesis at the University of Leicester in the UK, and one of the co-authors of the paper, said earlier work by the group had shown that if mutations were introduced in the part of the pneumolysin protein that binds to the cell, its toxicity was much reduced. When the team used the resulting toxoid to immunize animals, they were then protected against subsequent challenge with S. pneumoniae.

Andrew told BioWorld International: "We have now been awarded a European grant to investigate how we might identify small molecules and peptides that will neutralize the activity of the toxin. For example, if we could find a molecule that would bind to part of the toxin and stop the molecule from bending, then you would have a ring of molecules that could not form a hole. Likewise, if you put a mutation in that area, to stop it bending, you would have a molecule that looked like the wild-type molecule, but without being toxic - and that would be a good vaccine candidate."