Afficionados of espionage, whether real-life or fictional, know the most sinister of all enemy agents is the sleeper. He or she was planted in a target country long years before they had to surface, learning the terrain and inspiring local confidence, while waiting to strike when the iron got hot.

So it should come as no surprise that HIV, the AIDS virus, plants sleeper infective particles in the CD4+ T cells that are its targets for replication.

Actually, those target T cells have been activated — at the ready to divide — by the viral invasion. A minority of non-dividing T cells stay in reserve, resting, as they await future assignments.

It is those resting T cells that HIV singles out as hideaways for its sleeper DNA. "Obviously," said immunologist and virologist Tae-Wook Chun, "the resting cells carrying an integrated form of viral DNA can act as a persistent and stable reservoir because those cells don't turn over. They just sit there and do nothing unless they are activated."

It was Chun, with a colleague, who in 1995 discovered this ability of the AIDS virus to go latent while he was a doctoral candidate at The Johns Hopkins University, in Baltimore. Chun is now a staff investigator at NIAID — the National Institute of Allergy and Infectious Diseases, in Bethesda, Md.

"At that time," he recalled, "and this was before the protease inhibitor came along, a lot of people found numbers of virions in patients' blood, even during the asymptomatic phase of HIV-1 infection. They speculated that there is no latency, but just because the active virus always exists during infection doesn't mean that cellular latency does not exist.

"So what we did," Chun recounted, "was speculate that some of the small number of cells, which were productively infected in an activated state, can actually go back to a resting, memory state."

He compared this sleeper strategy with the equal and opposite program of the human immune system's T cells.

"Just as when T cells get activated, if you induce some cell activation in a particular antigen, some of the T cells react to the pathogen [and] some of them go back to resting state. So next time this same antigen comes around, they respond quickly.

"We found," Chun went on, "that this same T cell process happens in HIV infection. And we isolated very pure resting CD4+ T cells from the peripheral blood of patients, and did a PCR-based assay that detects only the integrated form of DNA. Obviously the resting cells carrying integrated DNA can act as a long-term reservoir."

"At that time," he observed, "we looked only at patients treated with a less potent antiretroviral therapy. It was before the protease inhibitor came along."

Viral Eradication: 'A Major Major Obstacle'

Last year, Chun's lab showed that even in patients treated with triple-drug therapy for up to 30 months, the virus could still be isolated from that latently infected portion of resting T cells.

"This suggested," Chun pointed out, "that just because you have undetectably low viral loads in patients who are on triple therapy doesn't mean that the virus has been eradicated. So it seems likely that the resistance of these cells is going to be one of the major major obstacles in our attempts to eradicate the infection. And that's where we are today."

Chun is first author of a paper in today's Proceedings of the National Academy of Sciences (PNAS), dated July 21, 1998, titled "Early establishment of a pool of latently infected, resting CD4+ T cells during primary HIV-1 infection."

It reports testing 10 HIV-infected patients at the University of Washington, in Seattle, with early and vigorous HAART — highly active antiretroviral therapy. That regimen consisted of zivudine (AZT) and lamivudine (3TC), two HIV reverse transcriptase inhibitors, plus the protease inhibitor, indinavir.

Chun described how these three drugs act to prevent HIV-1 from replicating in its target T cells.

"What the two HIV reverse transcriptase inhibitors, AZT and 3TC, do," he explained, "is when HIV enters the cells, it carries RNA, not DNA. So it has to convert itself into DNA, because RNA can't integrate into the host cell's DNA. The drugs block that process of HIV genomic RNA turning into DNA.

"Even after integration of DNA," he continued "the virus has to produce relevant HIV proteins. These must assemble themselves, so they can bud out from infected cells, and enter noninfected cells. What the protease inhibitor does is block the enzyme that HIV brings in to mature its own protein component. Protease inhibitors make it impossible for them to assemble the functioning virions."

The PNAS paper sought to answer two key questions raised by HIV's latent hideaway. How early is this latency established? And can starting HAART early prevent it?

Triple Therapy Kept Viral Load Down, Not Out

All 10 of those Seattle patients had a substantially high load of virus in their blood; nine of them came to the clinic with fever, swollen lymph glands, sore throat, rash and other discomforts — the symptoms of first-stage HIV infection.

"Those patients were put on triple therapy right away," Chun related, "anywhere from ten days after onset of symptoms to four months after. And they continued this treatemt for an average 10 months, up to 17 months.

"In all 10, we were able to detect the latently infected cells, which confirmed that this latency kicks in very early on, during the primary infection."

From these findings, Chun and his co-authors concluded, "As long as we have a detectable viremia in blood at the time of instituting therapy, it would be very difficult to prevent establishment of this latency."

He observed, "We are planning to revisit those 10 individuals pretty soon and find out what really happened to this latent reservoir. We don't know yet if it's gone or still there. But what I can tell you," he added, "is that in this particular group of patients, everybody responded extremely well to triple therapy. As far as I know, all of their viral loads are under detectable level. Which means they didn't form any persistent virus yet."

NIAID director Anthony Fauci is the PNAS paper's senior author. He reported its findings late last month to the World AIDS Conference, in Geneva, Switzerland. "Latently infected CD4+ T cells are potential sources of new viral replication," he told the delegates, "if a patient stops therapy or is unable to adhere to the proper treatment regimen. We now know that these reservoirs are established very early in the course of infection."

Fauci outlined three potential approaches for containing or depleting these reservoirs: "maintaining a strong immune system; enhancing a weakened immune system; and developing methods to deliberately purge the reservoirs by activating the resting cells to produce virus under the cover of HAART." *