Transgenic ewes in Scotland, cows in Holland, sows inVirginia, nanny-goats in Massachusetts, and mice inMaryland all are tooling up to express high-value proteinsin their mammary glands. At least one of the companiesengaged in this form of animal pharming, GenzymeTransgenics Corp., of Framingham, Mass., expects tostart Phase I trials by 1996 of its human anti-thrombin IIIprotein, produced in goats' milk. (See BioWorld Today,Feb. 13, 1995, p. 10.)
But none of these ovine, bovine, porcine, caprine andmurine transgenic ventures is home free. Turning dairyanimals and their mammary tissues into live bioreactorshas its limitations.
"One of the limiting steps," said biochemist andmolecular biologist Henryk Lubon, "is to express proteinsat very high levels, which are never expressed in thehuman body. We are going toward the point where wewill find limitations with cell systems."
Such an obstacle "is the processing step in proteinmodification. Most proteins," he explained, "aresynthesized from precursors. They are not synthesized asfully mature molecules."
Lubon is project leader of the transgenic researchprogram at the American Red Cross Holland Laboratoryin Rockville, Md. There he is currently engaged inoptimizing synthesis of protein C, a key player in theblood's anti-clotting cascade. (See BioWorld Today, Oct.31, 1995, p. 1.)
The current issue of the Proceedings of the NationalAcademy of Sciences (PNAS), dated Nov. 7, 1995,carries a paper reporting what Lubon calls "a unique andpatentable" strategy in expanding the capacity ofmammary cells to synthesize foreign proteins. Its title:"Proteolytic maturation of protein C upon engineering themouse mammary gland to express furin."
Stripping Proteins For Action
Lubon describes furin, (alias paired basic amino acidcleaving enzyme), as "the knife for changing transgenicanimals."
In PNAS, he described how he created hybrid, biogenicmice, by outfitting them with two transgenes, one toexpress protein C, the other encoding human proteolyticfurin, to process that product of interest. "This enzyme,"Lubon pointed out, "can specifically recognize thesequences in the protein C molecule, which have to beprocessed to make it biologically active."
He continued, "Protein C needs at least two enzymaticcuttings to make it fully processed. One clips itspropeptide, the second cuts the precursor protein to get itstwo-chain functional form."
To pilot this two-gene construct to its site of synthesis inthe mouse mammary gland, Lubon hitched the DNAcoding sequences to a whey acidic protein regulatorysequence, which is genomically programmed to act onlyin milk-producing tissues.
"Normally in the mammary gland," he explained, "asimilar furin enzyme exists. It's present in nearly everycell of every mammal. But that enzyme cannot process anunlimited amount of foreign protein. Introducing a genethat over-expresses the enzyme solved the problem ofproteolytically matured protein C in the mouse mammarygland."
This approach, the Red Cross researcher observed, "isnothing new from the viewpoint of technical users,because everybody has been doing transgenic animals formany years.
"For example, my colleagues at Genzyme Transgenics areworking on anti-thrombin III, and many other proteins.Some of these have another modification, likeglycosylation for example. And they have to consider thatin maybe a thousand bioreactor animals, thisglycosylation will not be perfect. So they can change it byadding the correct glycosylation processing enzyme.
"What is new," he went on, "is the concept of engineeringspecific tissues in transgenic animals in the direction ofbecoming bioreactors.
"Nobody could predict," Lubon observed "the effect ofexpressing this type of processing enzyme on our second-generation mice. We could get situations where theanimal would die, or where we might never obtain anyhealthy ones." He added, "The pleasant part for us, as weare reporting in PNAS, is that our biogenic mice aredoing surprisingly well."
Lubon sees a broader future for the furin family ofproteolytic enzymes than just clipping peptides offmarketable proteins in animal mammaries. "They processmany different proteins," he said. "Some expressed byHIV, others by bacterial toxins. Some people believe thatthe enzyme helps viruses to infect cells by cutting veryspecific proteins.
"Also, furin occurs in brain tissue, and some types oflung-cancer cells at high levels. So maybe in the future,"he suggested, "furin will have something to do withdiagnostics."
He concluded, "A significant implication of this hybridtransgenic strategy is that it's not just a situation ofexpressing something to produce something. But you canengineer some metabolic pathways in transgenicanimals."
Molecular biologist Harry Meade is GenzymeTransgenics' vice-president of research. "Anything youcan do to improve on nature," he commented, "is whatthat PNAS paper is all about. Over-expressing an enzymethat helps in the clipping is certainly within reason. Ithink it's certainly viable."
Almost paraphrasing Lubon, Meade added, "It's the sametype of think we've been thinking about as well. If theglycosylation is a little different, you may want to put ingenes like glycosyl transferases." n
-- David N. Leff Science Editor
(c) 1997 American Health Consultants. All rights reserved.