By David N. Leff

In the sex life of bacteria, the "F" word stands for "factor."

It's the factor that controls, by conjugation, the exchange of genetic material between bacteria of different mating types.

"The F factors of bacteria," observed virologist Jean-Michel Vos, at the University of North Carolina, in Chapel Hill, "are huge — the size of one megabase."

That dimension served Vos well in constructing a large, circular DNA molecule, on the order of 200 kilobases in length. So did the genome of the Epstein-Barr virus (EBV).

"The trick," he recounted, "was to take some of the genetic components coming out of EBV, which, like bacteria, has the ability to maintain itself as a circular episome, or plasmid, in certain human cells."

Vos and his co-workers combined these viral sequences with a bacterial artificial chromosome (BAC) derived from the F factor. "What we did," he went on, "was merge those two elements, viral and bacterial, with the ability to maintain a very large episome in bacteria as a circular plasmid — five to 20 times the size [that] gene therapists work with — and we wanted to do it in human cells."

Vos is senior author of a paper reporting this feat, in the September issue of the journal Human Gene Therapy. Its title is, "A system for shuttling 200-kb BAC/PAC clones into human cells: Stable extrachromosomal persistence and long-term ectopic gene activation."

They constructed a chimeric system they call BAC-HAEC — the latter initials standing for "human artificial episomal chromosome."

"This kind of vector," Vos observed, "is very sophisticated. It carries a lot of genetic elements from both bacteria and human viruses. To be able to work with a defined human gene, we had to develop a novel trick — screening a BAC-based library for human beta-globin." That's the protein that carries the oxygen for red blood cells.

"We had to find a library in which we can pull out this human beta-globin gene of interest as a very large piece of DNA," he said. "But there are some regulatory elements in the human beta-globin locus that are far away from the gene itself — in our case, 50 kilobases away.

"To be sure that we could span that region, we had to look for an unusual type of database. The human genome project has generated this kind of library now, between 100 and 300 kilobases, which is very helpful. If you go to a regular library, the maximum you can find is 50 kilobases, which is far too small."

In the phage-based library the co-authors tracked down, they "identified a clone that carried the whole human beta-globin locus, including its control region. That's required to have the gene expressed as it should be. We took out this very large human insert of the BAC clone and retrofitted it with the BAC and HAEC sequences together."

An In Vitro Element Of Surprise

Next, they put that construct into human cells, identified the clone, "and the whole thing," Vos pointed out, "is still there and functioning more than a year later."

His university has filed for a patent, and Vos is cooperating with a biotech company, Chromos Molecular Systems, in Vancouver, B.C.

He recalled "a little element of surprise. We didn't expect that the beta-globin gene would be expressed in this cell type, because it's a non-erythroid [non-blood-forming] cell. It was a lucky break, because a lot of people would have said, 'You won't get any gene expression,' and actually we did.

"We don't think it's a lot," Vos allowed. "I don't know how helpful this would be in terms of gene therapy. What we are doing now," he added, "is putting the episome with the beta-globin gene into globin stem-cell precursors, a cell type where we know it should be expressed at a very high level.

"But for some reason," he went on, "when we put the gene on this episome, we get some expression. That's very interesting. We call it 'ectopic genomic activation,' which is a novel phenomenon that could be helpful for some gene therapy protocols. It means having the ability to put a very large gene with its own promoter into a different cell type, where it's usually not expressed, and getting it expressed.

"The good news," he said, "is that it should work in many different cell types besides the erythroid ones. We haven't tested human cells systematically in the lab, but we have done it in human skin and liver cells, and it does work very well."

Vos and his co-authors have begun experiments in transgenic mice with thalassemia, which is caused by a mutation in a hemoglobin gene. "Over the past year," he said, "we have been working with these rodents, carrying our artificial episome.

"It appears to work in this whole-animal system, which to me is very promising," he observed. At least for single-gene hereditary diseases, such as Lesch-Nyhan [syndrome] and cystic fibrosis, it looks like the way to go. Now we're about to start experiments with the beta-globin gene," Vos concluded. "I hope to have the data within six months, and then it's on to primate studies." *