The immune system uses proteins encoded by the human leukocyte antigen system (HLAs) to differentiate self-cells from non-self cells. The proteins encoded by HLAs are effectively unique to that person - any cell not displaying a person's HLA type is an invader, and 'marked' for destruction by the bodies defense system of killer T cells.

It is known that some individuals are better at controlling an HIV infection than others. It is thought that certain types of HLA proteins (for example variant HLA-B*5703) in these individuals allow the HIV virus to be more quickly identified and attacked by the bodies defense system.

However, HIV is proving to be a sneaky competitor. In individuals with this beefed up immune system, the virus has been shown to mutate three amino acids that T cells need to recognize the infected cells - effectively making the virus invisible to the bodies defense system.

In a study published in the April 13 issue of the Journal of Experimental Medicine (http://jem.rupress.org/), an international team of researchers have proven that these mutations made to hide the virus from the immune system actually reduce the virus' ability to replicate. In the report, researchers demonstrated that the mutant virus replicated 20 times slower than normal in cell culture.

The researchers went on to study the transmission patterns of the mutated virus. They looked at Zambian couples in which one HLA-B*5703-expressing person infected with triple-mutant virus passed the infection to a partner who either did or did not have the same HLA variant. When the HIV virus was transmitted to a person without HLA-B*5703, the virus changed its mutated amino acids back to their original sequence, most likely because the benefit of avoiding killer T cells no longer outweighed the cost of reduced replication. However, when transmitted to another HLA-B*5703-expressing person, the triple-mutated virus remained mutated - and in fact came out on top despite its reduced replication. In these individuals, the avoidance of killer T cells allowed the infection to rapidly proceed to clinical illness.

The research suggests that vaccines should be designed to produce a T cell response against a number of different viral peptides - something that experimental human vaccines so far have yet to achieve.