Retroviruses insert their DNA into a host genome in order to reproduce, but if they stick around long enough they might undergo a mutation that keeps them from popping back out. Nearly 8 percent of the human genome consists of such captured retroviral DNA sequences, which gradually become garbled over the millennia. A few of those acquired more recently, however, have nearly complete sequences. They belong to an extinct family of retroviruses called HERV-K (for human endogenous retrovirus, K type). Some of these HERV-K elements seem to play a role in placental development and even cause viruslike particles to form in certain tumors. Researchers could not isolate a functioning, infectious HERV-K virus from human samples to study its possible function, though.
Thierry Heidmann at France's Institute Gustave-Roussy in Villejuif and his colleagues made an end run around this obstacle by comparing 30 different HERV-K sequences. For each position in their final sequence they assigned the nucleotide base that was most common among the 30 originals at that position, according to a paper published online October 31 in Genome Research. They called the final virus product "Phoenix." When they exposed Phoenix to cultured human and mammalian cells, they observed spiky virus particles pinching off from the cells and floating in between them. The genomes of the cells also contained new HERV-K sequences, indicating the viruses had infected them.
The group also found they could reconstruct an infectious Phoenix-like virus by stitching together elements from three known sequences, a process that could in principle occur in living people, they say. Luckily, Phoenix itself infected cells weakly, and the stitched-together version was even milder, perhaps because of cellular defenses against retroviruses, the researchers report. They note that investigators could now use Phoenix as a type of reference in studying the possible role of spontaneous HERV-K activity in cancer.
The team states that Phoenix was handled according to French regulations and would only be sent to other labs that agree to follow biosafety level 3 precautions--the second most stringent. American researchers used the same level of safety in reconstructing the strain of flu responsible for the 1918 pandemic, as they reported last year. Some researchers take issue, however, with bringing the retrovirus back to life this way. The group could not have known or predicted the low infectivity of the virus beforehand, stresses molecular biologist Richard Ebright of Rutgers University, and should therefore have performed the work at the highest safety level after seeking national or international review.