An experimental vaccine against the deadly poison ricin has produced encouraging results in mice, US army experts report. The treatment can be applied directly to the skin, potentially eliminating the need for injections.

The researchers hope that the vaccine can be adminstered through a patch similar to the nicotine patches that smokers use when quitting. "The ultimate goal is something like a Band Aid," says Gary Matyas of the Walter Reed Army Institute of Research in Silver Spring, Maryland, who led the research.

Ricin is a potent toxin that forms as a by-product when castor beans are turned into castor oil. An amount equivalent to a grain of salt is enough to kill an adult. Victims develop fever, nausea and abdominal pain, before dying of multiple organ failure within a few days of exposure.

Because castor beans are readily available, public-health officials are worried that it could be used as a bioterror agent. Last month, several buildings in Washington DC were evacuated when ricin powder was found in mail sent to the office of the senator Bill Frist.

There is no effective vaccine or treatment for ricin poisoning in humans, says Matyas. His team unveiled the mouse study at a meeting of the American Society for Microbiology in Baltimore, Maryland.

The researchers took a harmless fragment of the ricin molecule, called RTA 1-33/44-198, and gave it to mice either as a liquid dropped directly onto the skin or as a patch impregnated with the chemical. After breathing a fine mist of tiny ricin particles, all of the mice given the liquid vaccine survived, whereas untreated animals were poisoned.

What's more, 70% of animals given the patch also survived, leading Matyas's team to hope that the method could be honed for even greater success. Patches will be more practical than skin drops for treating humans, he points out, as the correct dose can easily be delivered.

Deliberate ricin attacks are currently rare, says Catalin Dogariou, a forensic scientist at the University of Huddersfield, UK. But the poison's potency means that a vaccine could still be useful for those most at risk of exposure, he adds.

"We should consider risk factors to identify the people under greatest potential threat," Dogariou says. These might include inhabitants of large cities, he suggests, or more specific groups such as those who work in confined areas. This danger was underlined in Tokyo in 1995, when terrorists killed 12 people on the city's subway system with the synthetic nerve agent sarin.

There is a long way to go until the new vaccine is available to humans, Matyas warns. His team found that mice treated with RTA 1-33/44-198 develop antibodies against the poison, but it is not clear how long the immunity lasts.

So Matyas and his colleagues are planning to run more tests in mice, before graduating to large primates and, ultimately, humans. "The vaccine induces huge numbers of antibodies," he says. "But we need to work out how long they last."