For 400 years, pet canary birds (Sirenus canaria) have filled countlesshomes with song. And coal miners carried caged canaries down intothe pits to warn them against toxic methane gas.

The little yellow birds are sensitive to another lethal danger as well _canarypox virus (CPV), which is species-specific to them. CPV is anavipox virus, related to variola virus, the once-dreaded vector ofsmallpox, and to variola's weaker cousin, cowpox virus, a.k.a.vaccinia.

It was precisely two centuries ago this month, on May 14, 1796, thatan English country doctor, Edward Jenner, inoculated pus from acowpox sore into the skin of eight-year-old James Phipps. Six weekslater, Jenner challenged the lad with smallpox pus; he resisted thathyper-contagious infection.

Vaccinia virus thus gave its name to smallpox vaccination, which theWorld Health Organization used to wipe out the disease in 1977.

More recently, virologists and immunologists at Virogenetics Corp.,in Troy, N.Y., recruited vaccinia virus as a viral vector for genetherapy experiments. Though quite successful, it had somedrawbacks, so they began employing canarypox virus as well.

CPV turned out to be a gene therapist's dream come true. Its roomygenome delivered gene constructs into animal models, which dulyexpressed their protein, but the virus itself could not replicate inmammals; it is strictly for the birds.

The company, now owned 80 percent by Pasteur-Merieux Serums etVaccins of Lyon, France, has teamed up with Arnold Levine,chairman of molecular biology at Princeton University. Their jointproject is to develop a multipurpose vaccine against cancer, based onthe p53 protein.

p53 is perhaps at once the most famous and infamous molecule of ourdecade.

The good-cop, wild-type protein is ubiquitous in mammalian cells,which it patrols as a tumor suppressor. But when p53 mutates, itsbad-cop nature turns it into a tumor-promoting oncogen. Mutations inthe p53 tumor suppressor gene are the most commonly found geneticalterations in human malignancies.

At Princeton, Levine and his post-doc, Judith Roth, set out to devisean anti-tumor vaccine based on putative antigenic targets in p53mutants.

They report their unexpected preclinical results to date in theProceedings of the National Academy of Sciences (PNAS), datedMay 14, 1996, under the title: "p53 as a target for cancer vaccines:Recombinant canarypox virus vectors expressing p53 protect miceagainst lethal tumor cell challenge."

The paper's first author, physician and oncologist Roth, toldBioWorld Today: "Virogenetics constructed the recombinant virusesfor us, which I then used for the animal experiments andimmunization studies."

Besides the wild-type human p53 gene, which the researchersregarded as a non-immunogenic control, they commissioned tworecombinant mutant versions from Virogenetics, involving amino-acid substitutions, arginine to histidine, at two sites on the genome.The canarypox viral vector constructs used vaccinia virus promoters.

Groups of young female mice vaccinated with one or another of thesegene vaccines were subsequently inoculated with challenge doses ofthe aggressive murine malignancy.

Roth and her co-authors asked themselves: Which mice will developtumors; which not?

"Our first observation," she said, "was that all 25 of the control mice_ those not vaccinated, or which received the wild-type canarypoxviral vector _ developed tumors. On the other hand, when we usedthe virus that expresses p53, about 75 percent of the animals wereprotected. They didn't develop any tumors at all during the 130-dayobservation period reported in our PNAS paper. I eventually endedtheir lives at 250 days."

Then came the kicker:

"What was interesting for us," Roth continued, "was the fact that wesaw no difference between the virus that expressed the wild-type,tumor-suppressive p53, and the virus that expressed the two differentmutant p53 forms." She added that this outcome "was not surprisingfor us. It complemented our theoretical predictions that immunizationwith a wild-type virus was as efficient as with a mutant virus."

Confirmation of this phenomenon, Roth observed, "makes the wholeconcept more promising. It would mean that we wouldn't have toidentify a precise p53 mutation in an individual patient in advance.

"Otherwise, we would have to go into every tumor first, look at whatkind of mutation was there _ as people have done before when theywanted to do peptide vaccination. First they define the mutationprecisely, then design a peptide that makes it probably immunogenic.

"But," she continued, "that is much more work-intensive. And wedon't think it necessary, because we think it is the over-expression ofthis mutant p53 protein that is responsible for eliciting an effectiveimmune response.

"Therefore, if we could use a virus that expressed wild-type p53, thenwe could immunize against any type of tumor that makes mutantp53."

The Princeton and Virogenetics co-authors now are pursuing "asecond generation of viral vectors," Roth said, "designed to expressanti-tumor cytokines _ for example, interleukins, granulocyte-macrophage colony-stimulating factor and tumor necrosis factor. It'shard to predict in advance which of these will win finally," sheobserved.

Negotiations with the company also are going on, she added, foreventual human trials of their CPV-based, anti-cancer vaccines.Meanwhile, Virogenetics has provided Genentech Inc., of South SanFrancisco, with canarypox virus genetically engineered to expressantigenic HIV proteins. (See BioWorld Today, Feb. 21, 1996, p. 3.)n

-- David N. Leff Science Editor

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