Mummified HIV: it's still dangerous.

In a perverse twist of nature, the body's mechanism for conferring long-term immunity against mumps or measles may also keep HIV infections going, report researchers at Virginia Commonwealth University in Richmond.

Specialized cells in the lymph nodes called follicular dendritic cells (FDCs) constitute a reference library for immune memory. But the new work indicates that FDCs somehow prompt the AIDS-causing virus to become highly infective, even when it's surrounded by antibodies that would normally neutralize it.

The lymph tissues thus become the major site of activity for HIV, especially during the "latent" period of the disease.

Earlier studies had hinted that lymph tissues harbor infection, but this work gives the first clear sign of how HIV infects immune cells in these sequestered sites, the group reports in the Oct. 26 Nature.

With their "octopuslike extensions," explains coauthor Gregory F. Burton, FDCs form a web within the spongy tissues of the lymph nodes, tonsils, and spleen. There, they catch and hold antigens--foreign molecules from bacteria, viruses, and the like--as a mesh screen catches lint. For months or even years, the immune system's memory cells refer back to the antigens for information on how to combat these bacteria or viruses.

FDC surfaces can also trap whole viruses. In the case of HIV, the surfaces bind not to the virus directly but to antibodies coating it. These antibodies remain from earlier attempts by the body to neutralize the virus.

Immunologists have known for years that lymph tissues trap HIV. Researchers at the National Institutes of Health in Bethesda, Md., for example, showed that HIV congregates on FDC "fingers" as the infection grows. Since 1993, they've also known that HIV infects cells there.

No one knew, however, whether the molecularly mummified HIV, filtered from the blood and held in lymph tissues, could still infect. "One wouldn't expect HIV to be infectious here," says Burton, "because it's covered with antibody."

The new work clearly shows that HIV in lymph tissues remains active. In a test-tube experiment, researchers combined antibody-coated HIV with FDCs from the tonsils of uninfected people. They then added CD4+ T cells, the white blood cells that the virus typically attacks. Within a short time, the virus had infected the T cells.

Researchers also injected HIV into mice and then collected FDC cells from the animals' lymph nodes. The cells were dotted with antibody-coated HIV. The group later added human T cells to the mouse lymph cells and found that HIV readily infected the T cells.

In a final test, the researchers added antibody-coated HIV to laboratory-grown T cells either with or without FDCs. In the absence of FDCs, the coated HIV did not infect the T cells, Burton says, "but if we put [FDCs] into the culture, all of a sudden there was tremendous infection." The researchers conclude that, once in the lymph tissue and linked with FDCs, the antibody-coated virus becomes infective.

"We don't know exactly what's happening," he adds. But the researchers suspect that the FDCs somehow push the virus' antibody coat aside, revealing areas that react with T cells.

What does this mean for experimental vaccines that aim to quell HIV by stimulating antibody production? Thailand, for example, is conducting tests of HIV vaccines on people. "The antibody vaccines may have some usefulness," Burton says, "but I don't think they will ever block infection completely because of what goes on at the FDCs." NIH's Anthony S. Fauci adds: "To say from this that the vaccines are no good is too major a leap."