HIV can modify itself very easily

The key seems to be genetic variation in molecules that trigger the destruction of cells infected by HIV.

Oxford University researchers have found HIV is more likely to scupper this process for common forms of the molecule than for rarer versions.

The study is published in Proceedings of the National Academy of Sciences.

HIV's profound ability to adapt to its environment continues to astound. Dr John Frater

The Oxford team looked at the role of molecules called human leukocyte antigens (HLAs), which are found on the surface of many of the body's cells.

Under normal circumstances when a virus such as HIV infects a cell, HLAs trigger the activation of immune system T cells that move in and destroy the infected cell.

The Oxford researchers examined data from a long-term study of Swiss HIV patients.

They found that patients who showed signs of rapid disease progression often carried an HLA variation that failed to mobilise T cells in the normal way.

The scientists also found that rare HLA types were more likely to elicit an immune response than commonly occurring HLA types.

They believe that HIV, which has an astonishing ability to mutate to suit its circumstances, may be able to change to evade detection by the more commonly occurring HLA types.

However, because the virus has seldom encountered the more rare HLA types, it has not had the opportunity as yet to mutate, and escape detection in the same way.

Researcher Dr John Frater said the findings suggested that potential HIV vaccines based on commonly found HLA versions might be less likely to be effective.

He told BBC News Online: "HIV's profound ability to adapt to its environment, whether it be drug pressure, or the body's own immune system, continues to astound."

Hope for vaccine

Keith Alcorn, senior editor of the HIV information service NAM, told BBC News Online: "The findings help to explain why some people can live with HIV infection for years without illness.

"They also have implications for vaccine design because vaccine developers are trying to identify which viral peptides need to be included in a vaccine."

However, Mr Alcorn said to get a fuller picture it would be necessary to repeat the research on other groups of HIV patients who have been infected with different sub-types of the virus.

Deborah Jack, chief executive of the National Aids Trust said: "This kind of research makes an important contribution to our understanding of HIV and Aids and can contribute to the vital objective of discovering an effective Aids vaccine."