New insights into how HIV becomes latent in host cells could lead the way to improved retroviral therapy, according to a report in the August issue of Cell Host and Microbe.

Once HIV infection occurs, the virus can integrate into actively transcribed host genes, where it becomes latent. These reservoirs of latent HIV can reactivate and continue to spread the disease after retroviral therapy -- which can only target active virus -- has removed the majority of the viral load.

Now, researchers led by B. Matija Peterlin and colleagues at the Rosalind Russell Medical Research Center at the University of California, San Francisco have found two mechanisms that, at least experimentally, trigger transcription of latent HIV.

While there are many mechanisms by which a virus can become latent, Peterlin's group focused on transcriptional interference, in which a virus silences itself by integrating into a commonly used gene. The cell's own transcription machinery then plows over the viruses' initiation site and the viral genome is spliced out with the gene's intronic sequences.

Peterlin's group found that transcriptional interference could be reversed by either inactivating the host gene or attaching strong viral activators, such as NF-kB, which "bind DNA tighter than the polymerase can displace it," said Peterlin. "The moment you stop transcribing the active gene," said Peterlin, "the virus can come right out."

Since DNA polymerase can't be completely silenced, said Peterlin, his group has focused on inducing NF-kB in the host cells, which triggers latent virus to begin expressing in the cell model.

"There's a long history of transcriptional interference in the field," said Rick Bushman from the University of Pennsylvania Medical Center, who was not involved in the study. Peterlin's study adds to the current knowledge, he said, by working out some of the proteins involved in the process.

However, transferring the results to human could be a challenge, he added. "It would be great if it works," said Bushman, but studying latency from patient cells has been notoriously difficult. Also, one would need to activate all of the latent virus for complete recovery. "It's kind of a yes-no thing," said Bushman.

However, activating NF-kB also has the affect of loosening chromatin, which HIV recruits in another mechanism of latency. Therefore, upregulating the viral activator would be "good for both of these forms" of latency, said Peterlin.

While there are drugs that could be used in humans to activate NF-kB, toxicity is an issue, said Peterlin. His group plans to test different pharmacological agents to try to purge HIV from its latent stage.