A recently discovered class of immune cell may hold the key to new treatments for asthma - and explain why existing therapies sometimes fail.

Asthma occurs when immune cells go into overdrive and release inflammatory chemicals called cytokines. These cause excess production of mucus, which plugs up the lungs. The disease is generally associated with immune cells called T-helper 2 (TH2) cells and the cytokines they release, but their response alone is not enough to trigger asthma.

Natural killer T (NKT) cells produce some of the same cytokines as TH2s, but release them faster and in greater quantities (see Diagram). NKT cells are hybrids: they kill invading microbes, like natural killer immune cells, but they also bind to antigens - foreign substances that trigger an immune response - like T-cells do.

Last year, Omid Akbari and his colleagues at Children's Hospital Boston, Massachusetts, discovered high numbers of NKT cells in the lungs of people with severe asthma, but virtually none in the lungs of healthy people. Meanwhile, separate studies in mice have shown NKT cell activation alone is enough to trigger asthma, prompting researchers to speculate that NKT cells might be equally, or more, important than TH2 cells in the development of the disease.

Akbari's team has now used a drug called DPPE-PEG to inhibit the action of NKT cells in mice and found that it prevented the type of asthma linked to allergies. DPPE-PEG already has approval from the US Food and Drug Administration, so clinical trials could begin as early as this year, says Akbari, who presented his findings at a meeting of the American Academy of Asthma, Allergy and Immunology in San Diego, California, this week.

The findings may help explain why some people with asthma are resistant to conventional drugs such as corticosteroids. These inhibit many immune cells, but do not work on NKT cells, Akbari says.

However, while DPPE-PEG shows promise, more work is needed to understand the role of NKT cells in asthma. For a start, they seem to bind to glycolipids - molecules made from fats and sugars - rather than the protein antigens recognised by most immune cells. Few glycolipid antigens have been discovered so far, although some have now been identified in bacteria and pollen.

In the case of DPPE-PEG, Akbari believes it works by blocking the antigen receptors on NKT cells and so preventing their immune response. But since the allergen his team used to trigger asthma in the mice was an egg protein, it is unlikely to have been this that stimulated the NKT cells in the first place. It could be that protein allergens trigger the release of naturally occurring glycolipids by some as yet unidentified mechanism, which then stimulate NKT cells, suggests Akbari.

"It is also possible that there's a bacterial component to asthma," says Mitchell Kronenberg, president of the La Jolla Institute of Allergy and Immunology in California. He suggests that bacteria present in the body may produce glycolipids that prime NKT cells to respond when an allergen is introduced.