Inserting a gene in vivo to create a transgenic mouse, or treat agenetically challenged human, is at best a crap-shoot _ a riskyenterprise.

"Pretty well all of the efforts to make a transgenic animal," observedmolecular geneticist Oliver Smithies, "have been done by injectingthe DNA into a male pronucleus, and allowing it to go randomly intothe genome." Such standard gene transfer methods, he pointed out,"result in randomly inserted genes and unpredictable copy numbersand expression levels."

Smithies, a professor of pathology at the University of NorthCarolina in Chapel Hill, is senior author (and post-doc SarahBronson, first author) of a paper in today's Proceedings of theNational Academy of Sciences (PNAS) titled: "Single-copytransgenic mice with chosen-site integration."

"What we've done," Smithies told BioWorld Today, "is demonstratethe usefulness of combining transgenesis technology withhomologous recombination," a sequence-matching method heintroduced in 1985. Combining both approaches, he said, "allows onereproducibly to make transgenic animals in a chosen place on thegenome." In other words, he went on, "Any time you make atransgenic animal by this method, all of the animals will be the same.Whereas by the standard technique, they are all essentially different."

As reported in PNAS, the targeted gene transfer produced singlecopies of the inserted sequence, with site-specific integration andexpression in a wide variety of tissues.

The method used homologous recombination in embryonic stem cellsto generate mice carrying a single copy of the bcl-2 gene, integratedinto a preselected genomic location. Smithies explained that "bcl-2 isa protooncogene that helps prevent programmed cell death,particularly in the developing immune system. We hoped to use it tomake bone marrow stem cells survive longer."

Dead Aim, Not Scatter Shot

In this initial test case, he continued, "we wanted to see if the genewould be expressed in all tissues, so we picked the b-actin promoterin two versions, human and avian [chicken]." Both in vitro and invivo, the human promoter directed a higher expression level of thetransgene than did the avian. They both did so in a wide variety oftissues, including bone marrow, brain, heart, thymus, kidney, spleenand reproductive organs.

Smithies and his co-authors positioned their bcl-2/b-actin expressionvector into murine embryonic stem cells adjacent to the genomiclocus of a housekeeping gene, Hprt (hypoxanthinephosphoribosyltransferase). It resides on the sex-linked Xchromosome, and is expressed ubiquitously in the body.

"Hprt," Smithies observed, "is an interesting gene in its own right.Mild defects cause gout, but only in males. Severe defects cause thelethal Lesch-Nyan syndrome, also in males only."

The team made Hprt its bull's eye, he explained, "because we alreadyhave DNA constructs that can hit this locus with high efficiency."

Having shown the ability of the double-threat expression system tohit a specific gene locus, one of Smithies' co-authors moleculargeneticist Nobuyo Maeda, now is aiming a gene at a single tissuetarget in which it is normally active, namely the liver.

This is the gene that encodes human haptoglobin, a protein that bindshemoglobin. At first, Smithies recounted, "she introduced it into themouse Hprt locus, and found that it was not expressed there, eventhough everything about the gene is functional. But when sheintroduced it in the murine liver," he added, "the transgene wasactive."

The haptoglobin promoter, he explained, "already has liverspecificity, but that was not sufficient. It had to be not only a liver-specific promoter, but a liver-specific place to which the gene went."

That demonstrates, he pointed out, that "Where one puts thetransgene makes a difference to how it works. That's been known fora long time, but it's not been done in a way that allows one to choosethe optimal genomic site."

Molecular geneticist Raju Kucherlapati, who chairs moleculargenetics at Albert Einstein College of Medicine, in Bronx, N.Y., is ascientific founder and board member of Cell Genesys Inc., of FosterCity, Calif., and Millennium Pharmaceuticals Inc., of Cambridge,Mass. He is well acquainted with Smithies' latest experiment, andtold BioWorld Today:

"What Oliver shows in this paper is, first, that the site at which theDNA sequences become incorporated into the genome plays a veryimportant role in terms of their expression. The second feature,"Kucherlapati continued, "is that the technology he describes really forthe first time shows that you can introduce a gene into these cells, andbe able to predict a priori exactly what the levels of expression aregoing to be."

Besides such transgenics, he pointed out, "In most of the currenttechnologies that are used for gene therapy, where you actuallyintroduce a gene that is missing, one of the big difficulties is reallybeing able to predict what the levels of expression of these genes aregoing to be. If you could modify primary cells, using the technologythat Oliver's talking about, this would have very significant value fora lot of biotechnology people who are interested in trying to getpredictable and high-level and regulatable expression of genes."

Smithies concluded: "Now that it's been published, I'm sure otherlaboratories will ask about this approach. I don't think it's abreakthrough, just a nice extension of many other things." n

-- David N. Leff Science Editor

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