Researcher Robert Debs has long considered liposomes moreattractive than viruses for gene delivery. But a large questionhas remained: Would they work?

In a research paper published in today's issue of Science, hepresents evidence that intravenous injection of liposome-DNAcomplexes into adult mice resulted in gene expression for up tonine weeks in lung, spleen, lymph nodes and bone marrow.

Debs, an associate research physician at the Cancer ResearchInstitute at the University of California, San Francisco, co-authored the paper with UCSF post-doctoral scholars Ning Zhuand Yong Liu, and Denny Liggitt, professor of comparativemedicine at the Department of Comparative Medicine at theUniversity of Washington School of Medicine in Seattle.

"We have the option of expressing virtually any gene in anytissue," Debs told BioWorld.

Several factors enabled the complexes to work, he said,including use of a cationic liposome, a DNA to liposome ratio of1 to 8, and high doses of DNA (100 micrograms for a 20-grammouse).

Expression levels were increased by using a plasmid containingthe immediate-early 1 promoter-enhancer element of humancytomegalovirus and an intron on the 5-prime side of thecoding region.

The plasmid apparently was taken into cells across endothelialbarriers and appeared to stay separate in the cell rather thanjoin the chromosome, where it might have risked alteringnormal functions.

Debs suggests targeting specific cell types, for instance, byusing specific promoter-enhancers, introducing the complexinto selected areas of the body, and placing targeting ligands onthe surface of the liposome, whose positive charge enablesnegatively charged DNA to attach more tightly to the lipidsphere.

His discoveries are subject to a licensing agreement betweenUCSF and Megabios Corp., a private, 18-month-old companybased in San Francisco, which was formed to develop lipid-based in vivo gene therapy.

Started by Frank Caufield, a venture capitalist retired fromKleiner Perkins Caufield & Byers, and chemical engineerWilliam Brown, Megabios concentrates on vector construction,lipid construction and complexing lipids with DNA, said Brown,the company's vice president.

"We thought there were all sorts of problems with viralvectors," he added.

For instance, Debs said, viruses might provoke immuneresponses and can only deliver small pieces of DNA. He saidliposomes, by contrast, are cheaper, non-toxic, more versatileand compatible with repeated human use.

Debs has also administered DNA via an aerosol directly to thecentral nervous system. Besides introducing a marker genecalled chloramphenicol acetyltransferase (CAT) into mice, healso induced mice cells to incorporate the cystic fibrosistransmembrane regulator (CFTR) gene that can correct cysticfibrosis. The gene was expressed in large numbers of lung cellsat least 150 days after final injection.

Since the paper was submitted, the team and collaboratorshave explored delivery of anti-cancer genes such assuppressers or the cytokine granulocyte macrophage colonystimulating factor (GM-CSF) in rats, mice and goats, and anti-infective properties of GM-CSF gene delivery in bacterialpneumonia (with other cytokines and anti-viral and anti-fungalproperties also on the agenda). The team also introduced theCFTR gene into sheep, monkeys and mice.

"It works with different animals and different genes," Debssaid. On the therapeutic front, Debs is collaborating withRichard Boucher, director of the Cystic Fibrosis Center at theUniversity of North Carolina, to design a human trial for cysticfibrosis, and is also looking at genes that might treat geneticdiseases such as Gaucher's.

-- Nancy Garcia Associate Editor

(c) 1997 American Health Consultants. All rights reserved.