Despite much effort, no one has managed to develop an effective form of gene therapy to treat the lung problems of people with cystic fibrosis. The condition is caused by a mutation in the CFTR gene, and one of the problems in developing a treatment is getting a working version of CFTR into a high enough proportion of cells to restore normal lung function.
While HIV is good at adding genes to cells' genomes, it does not target lung epithelial cells. Ebola binds strongly to these cells, so Gary Kobinger at the University of Pennsylvania in Philadelphia tried adding a surface protein from Ebola to HIV. Initial tests on mice showed the hybrid virus was very efficient: the gene it carried was active in 24 per cent of airway cells after two months, a far better proportion than achieved by other delivery methods (New Scientist, 10 March 2001, p 19).
But will it work in humans? The team has now done the next best thing and used the hybrid virus to deliver a test gene, lacZ, to the lungs of monkeys. After two months the gene was active in 21 per cent of cells, Kobinger told New Scientist. He is preparing the results for publication.
"The gap between monkeys and humans is much smaller than from mice to monkeys," says Ray Pickles, a gene therapy researcher at the University of North Carolina, Chapel Hill. "This could be in human trials pretty soon."
Kobinger now plans a more extensive monkey trial to convince the US Food and Drug Administration that human trials will be safe. The hybrid virus is created by adding HIV RNA, a stripped-down version of the Ebola surface prot ein RNA and the therapeutic gene to cells. The hybrid virus then self-assembles. Crucially, only the therapeutic gene is incorporated into the hybrid virus, so it cannot replicate.
All attempts in the lab to force such viruses to regain the ability to replicate have failed. Even if this did happen in the body, the resulting virus would resemble HIV, as the hybrid contains no Ebola genes - and HIV cannot infect lung cells. If it did get into the blood, where normal HIV thrives, such a virus would not replicate as efficiently as wild-type HIV. While the patient might suffer, the virus could never reach high enough levels in the body to infect others, Pickles says.
The biggest risk is that if a patient becomes infected with wild-type HIV, the wild-type virus and the hybrid could somehow merge, but such recombination has never been observed in HIV.
Despite his confidence that the hybrid will be safe, Pickles has abandoned similar research as he fears opposition from a fearful public will prevent the method ever being used. Another problem with Kobinger's approach is that patients will require repeated treatments, because the added gene ends up in surface cells that die off after a few months. Other teams are trying to find ways of targeting the progenitor cells that give rise to the surface cells.