Among drug developers, HIV's main claim to fame is its infection of CD4-positive T cells. Most anti-HIV drugs, whether approved or in the clinic, focus on T cells, since that is where HIV replicates, killing off the very T cells that are meant to fight it in the process.

But the virus does not stop there. Two recent papers explore the mechanisms behind HIV's effects on B cells on the one hand, and its killing of uninfected T cells on the other.

In the July 2006 issue of PloS Pathogens, researchers from the University of Pittsburgh showed the molecular mechanisms by which B cells serve as the viruses' helper in infecting T cells, a process known as trans-infection.

The new research identifies an important first step in a new pathway involving B cells that express a protein called DC-SIGN. While those cells themselves do not become infected, they play a pivotal role as an accomplice in HIV's takeover of T cells.

Giovanna Rappocciolo, associate professor of infectious diseases and microbiology at the University of Pittsburgh and the study's first author, described DC-SIGN as "sort of a bridge HIV uses to reach the surface of T cells."

The researchers found DC-SIGN was expressed on B cells from both healthy subjects and HIV-infected individuals. About 8 percent of blood-borne B cells expressed DC-SIGN in both infected and uninfected individuals. In tonsils, a lymphoid organ where both T cells and B cells normally interact, about a quarter of B cells expressed DC-SIGN.

To test whether DC-SIGN is an accomplice in HIV's infection of T cells, Rappocciolo and her colleagues placed activated DC-SIGN B cells in culture with T cells and HIV. In the presence of B cells, HIV infected the T cells while sparing the B cells. When researchers repeated the experiment without B cells, the HIV had little effect on the T cells alone. Pretreating the B cells with a molecule that blocks DC-SIGN activation also protected the T cells from HIV infection.

While there's no great mystery as to what kills infected T cells, a more surprising feature of HIV is that it also kills "bystander" T cells that it does not infect. In the Aug. 1, 2006, issue of Journal of Clinical Investigation, researchers at the Institut de Biologie in Montpellier and INSERM Institute in Villejuif, both in France, shed some light on the exact mechanisms by which HIV propels uninfected T cells toward their demise, as well as on the more general relationship between two methods of cellular self-destruction: apoptosis and autophagy.

Autophagy, a process of cellular housekeeping, is one way in which cells protect themselves against some infectious invaders, but it was unclear whether HIV was one of them.

Autophagy also is somewhat on the fence in terms of its effects on the cell. It can be a survival mechanism, allowing a cell to rid itself of unwanted guests, or during periods of starvation, cannibalize itself until better times arrive. But other lines of evidence suggest that autophagic processes also can be a prelude to wholesale cell death.

In the JCI paper, the researchers cocultured cells that expressed HIV envelope proteins and uninfected T cells that expressed the chemokine receptor CXCR4, to test whether the interaction contributes to cell death via either autophagic or apoptotic mechanisms, or both.

They showed that when infected and uninfected T cells interacted via envelope proteins and CXCR4, that activated the autophagy pathway, which in turn triggered apoptosis in the uninfected T cells. Blocking autophagy, either via chemical compounds or short interfering RNAs, also blocked the apoptosis.

It is unclear whether autophagy is triggered in the uninfected T cells with the goal of killing them, or whether it is a protective mechanism of the cell that goes awry. But what is clear is that "once activated, the ultimate consequence of autophagy is the triggering of cell death," and that "autophagy thus plays a crucial role in . . . T-cell death that leads to the physiopathology of AIDS."