HIV Inserts Into Human Genome Using A DNA-associated Protein

HIV integrates into active transcription units on chromosomes within the nucleus of human cells. These units are sites that lead to efficient expression of the viral genome. Most HIV-infected cells in a patient will have a very short life span, a day or less. "We surmise that this strategy helps the virus make hay while the sun is shining, as it were, producing lots of viral copies during a short time, so that the virus can maximize production of daughter virions," says Frederic Bushman, PhD, Professor of Microbiology at Penn.

This present study demonstrates the first piece of a mechanism that dictates where HIV integration takes place. Previous studies at other institutions showed that LEDGF binds tightly to HIV integrase, the enzyme that's important for the integration reaction. Now, Penn researchers showed in this study that the way LEDGF binds to HIV integrase and to specific sites on chromosomes suggests that HIV targets integration using a molecular tether.

Retroviruses contain RNA in their particles. They enter a cell and convert RNA into DNA by the enzyme reverse transcriptase and then integrate that DNA copy into the DNA of the host, using the integrase enzyme. The new viral particles are made by transcription of the viral genome, as with any cellular genes. If the cell divides, the viral DNA is copied and inherited, along with cellular human genes.

Bushman and his team made cells that were depleted of LEDGF and found that integration was less frequent in transcription units and in genes regulated by LEDGF. "This implies that LEDGF is part of the machinery that helps dictate the placement of retroviral integration sites within chromosomes," says Bushman.

Bushman notes that finding that LEDGF is part of the cellular apparatus necessary for HIV replication is important to the field of gene therapy. Controlling where gene-therapy vehicles insert in the human genome could help make the delivery of new therapeutic sequences safer. The new findings about LEDGF suggest that engineered tethering interactions might some day allow control over integration site selection during gene therapy. According to Bushman, this finding is of particular importance in light of recent cases where integration of gene-therapy vectors near cancer genes contributed to the development of leukemia in gene-therapy patients.

"This is first example of a cellular factor that's a clear player in target site selection," says Bushman. "This isn't engineering yet, but it's a key piece of information on the way."

This research was funded in part by the National Institutes of Health, the J.B. Pendleton Charitable Trust, and the F.B. Burns Foundation. Other co-authors in addition to Bushman are: Angela Ciuffi, Christian Hoffman, and Jeremy Leipzig, all from Penn, as well as Manuel Llano and Eric Poeschla from the Mayo Clinic College of Medicine, Rochester, Minn., and Paul Shinn and Joseph R. Ecker from The Salk Institute, La Jolla, Calif.