A mimic of the potent proteins used by white blood cells to punch holes in bacteria is to begin clinical trials in Canada. It could be a valuable new weapon against the growing threat of antibiotic-resistant "superbugs".

Phagocyte immune cells that engulf bacteria use defensins to digest their prey. These small molecules are electrically attracted to the bacterium's outer membrane, and fuse with it to create gaping holes that destroy invaders.

The complexity of bacterial cells's outer membranes means they cannot easily evolve to become resistant to defensins. But attempts to harness them for medicine have struggled. Defensins are difficult and expensive to produce, and are typically destroyed by the host's immune system before they can reach an area of infection.

Local effect

"They are not optimised to circulate around the body," Bill DeGrado, a biochemist at the University of Pennsylvania, Philadelphia, US, told New Scientist. He decided that building his own defensin was the only solution.

His research team stripped down the structure of natural defensins to just the essential membrane-busting components, making one small enough to go undetected by the immune system.

They focussed on the split chemical personality that makes the molecules deadly to bacteria. They have one water-soluble end, and one water-repellent end with a positive charge. This draws the defensin to bacterial membranes like a targeted torpedo.

After settling on a simple ribbon-like structure, they enlisted the help of theoretical chemist Michael Klein at the University of Pennsylvania to check whether the proposed molecule could be stable.

Staying power

The simulated results were good enough to encourage DeGrado to build the molecule and set up a company, Polymedix, to develop it.

In a standard measure of the risk of bacteria evolving resistance to a new treatment, the defensin outstripped conventional antibiotics.

In the test, a sample of bacteria was given enough compound to kill 90% of the culture, with the survivors used to found a new one. The process was repeated, which drives the bacteria to evolve resistance.

"For conventional antibiotics, you generally find it takes 100 times more of the antibiotic to kill the bacteria after 9 repeats," says Nick Landekic, Polymedix's chief executive. "We've done 14 repeats with PMX-30063 and there is no change in its potency."

"Really exciting"

Michael Zasloff at Georgetown University, Washington DC, US, was not involved with the work but plans to follow its progress.

"Their compounds are really exciting and look great, but the value of this class will be based on safety." He says there is a chance that PMX-30063 may punch holes in human cells as well as bacterial ones.

"My concerns are with how it interacts with sites in the body known to be sensitive to damage by positively-charged peptides – areas such as the kidney and middle ear," says Zasloff.

DeGrado and colleagues think the risk is low, saying their molecule is several thousand times more likely to target bacterial cells than it is to attack mammalian ones.

After a programme of animal testing the Canadian government's regulator, Health Canada, this month gave the go-ahead to human clinical trials.