AIDS virus may feast on an unexpected meal.

HIV is picky about its prey. The AIDS virus attacks only immune cells--and only those with a surface protein called CD4. The virus' preference stems from its need to bind to CD4 before infecting cells.

The discriminating palate of the AIDS virus has been thought to render the other major class of immune cells, which displays a protein called CD8, but not CD4, virtually invulnerable to infection. Three research groups now challenge that dogma with evidence showing that under certain conditions, CD8 cells can make the CD4 protein and consequently become infected by the AIDS virus.

In the body, CD8 cells can transform themselves into killers that seek out and destroy other cells infected by HIV. The virus' unexpected access to CD8 cells may impair these killers, thus speeding the development of AIDS, investigators speculate.

"One envisions that these cells get infected [with HIV] as soon as they start to respond [to it]," says Jerome Zack of the University of California, Los Angeles. "That would be a really smart way for the virus to elude the immune system."

While most scientists have focused on the declining number of CD4 cells that follows untreated HIV infection, others have documented that CD8 cell counts also fall over time, especially late in an infection. Moreover, several recent studies of people with AIDS have shown that their CD8 cells frequently contain HIV genes, evidence that those cells were infected.

Since CD8 cells weren't known to make CD4, a prerequisite for infection by the virus, the puzzling results went unexplained and to some extent ignored.

Now, Robert C. Gallo of the University of Maryland's Institute of Human Virology in Baltimore and his colleagues report that when CD8 cells are stimulated out of their quiescent state, as they would be when responding to an infection, they make CD4.

"This gives a mechanism [for CD8 cell infection]. People now have to pay attention," says Gallo, whose group reports its findings in the March 17 Proceedings of the National Academy of Sciences.

"Now that there are three labs finding this, I think people are going to believe it," agrees Carl June of the Naval Medical Research Institute in Bethesda, Md. Zack and June presented data at a recent AIDS meeting showing that stimulated CD8 cells produce CD4 and thus become vulnerable to HIV infection. June's group will publish its findings in the April Journal of Experimental Medicine.

June's and Zack's groups have observed that HIV infects naive CD8 cells--those never before exposed to an infection--more efficiently than CD8 cells that have previously been called into action. Naive cells can make CD4, but the veteran cells cannot, says June.

Since most CD8 cells in infants are naive, the fondness of the AIDS virus for those immune cells "may be one reason why children with an HIV infection progress so much faster than adults," notes Zack.

While all three groups have shown that CD8 cells grown in test tubes or obtained directly from tissues or blood can make CD4 and become infected by HIV, none has yet found that this phenomenon impairs the function of the cells in the body. In test tubes, the HIV-infected CD8 cells die prematurely, says Gallo.

"The key question is whether the infection of CD8 cells causes a decline in cytotoxic function," says Peter Simmonds of the University of Edinburgh, who had previously found that CD8 cells in people can be infected by HIV. "If you can show [the CD8 class] is involved, that clearly makes a big difference in trying to understand the immune suppression that AIDS patients have."

Why would CD8 cells begin to make CD4 when stimulated? Zack speculates that the protein, known to mediate interactions between cells, may help CD8 cells make contact with the class of immune cells that teaches them how to identify HIV-infected cells.