By David N. Leff

A stroke of serendipity in France has cancer researchers in Boston swivelling their big guns onto a new target -- p73.

No, that's not a typographical error: p73 is a just-discovered gene that looks a lot like the well-known tumor suppressor gene, p53. The differences, as well as the similarities, between these two molecules and their expressed proteins are the subject of papers in two recent journals, Cell, dated Aug. 22, 1997, and today's issue of Nature, dated Sept. 11, 1997.

The article in Nature reports that "p73 is a human p53-related protein that can induce apoptosis." The paper in last month's Cell, which broke the p73 discovery story, bears the title: "Monoallelically expressed gene related to p53 at the neuroblastoma suppressor-1 locus." Its principal author is molecular geneticist Daniel Caput, assistant scientific director of Sanofi Recherche, in Labége, France.

"p73 was just fortuitously discovered," Caput told BioWorld Today, "through an unrelated type of screening. In fact, we were looking for a cytokine receptor. It's probably because we were doing systematic sequencing, using oligonucleotide degeneration," he added, "that we were lucky enough to come across this sequence."

Mouse Studies Examining How p73 Works

Since that happy happenstance in 1995, Caput and his team have been "deeply working on the biology of p73, which means its relationship with p53's function. And we are doing a lot of other things, obviously, because we also have mouse studies."

Cell biologist Frank McKeon and his colleagues at Harvard University Medical School, in Boston, are doing the hands-on murine side of the French research, in collaboration with Caput. They have generated knockout mice, lacking the p73 gene, "and we're now analyzing these critters," McKeon told BioWorld Today, "but that's another story."

He's also enrolled other mice. "We're going with all due speed," McKeon said, "but to really understand p73 you want to be crossing these p73-minus knockouts into p53-minus ones, as well as mice lacking other models for tumor development -- and development in general. So we're striving to make mouse models that will really mean something, not just be freaks of nature."

In recent years, the p53 gene has gained fame as a tumor suppressor, which, when mutated, expresses proteins that contribute to malignancy. A good half of all cancer cells carry the aberrant p53 protein, or lack the normal p53 altogether. Hence the interest in p73 as a putative tumor suppressor on its own, or as some sort of back-up to p53.

The p53 gene resides on the short arm of human chromosome 17, while p73 sits on chromosome 1's short arm. This chromosomal region is a notorious hot spot for neuroblastomas, plus carcinomas of breast, colon, melanoma, and other tumors too numerous to mention.

Molecular biologist William Kaelin, principal author of the paper in Nature, observed: "Whether p73 turns out to be the neuroblastoma gene, or whether in fact it's even a tumor suppressor gene, what I would say at this point is that it's an extremely strong candidate to be one of the tumor suppressor genes suspected of inhabiting that region of chromosome 1."

Kaelin's lab at the Dana-Farber Cancer Institute, in Boston, concentrates on tumor suppression.

"Our paper in Nature," he told BioWorld Today, "focused on the question: Could p73 act like p53? What we reported was that, as predicted, at least when overproduced, the protein products of p73 could in fact activate transcription of genes normally regulated by p53 and itself act as a tumor suppressor in certain standard assays.

"This," he added, "appeared to be due, at least in part, to programmed cell death -- apoptosis -- which is also how p53 works."

But Kaelin went on to say that he detected this apparently parallel behavior between p73 and p53 under special conditions.

"We were conservative in our paper to point out that we don't know that this protein, under normal circumstances, activates transcription, and/or induces apoptosis. But we've shown, at least in principle, that it could be made to do so. In other words," he went on, "if you had a drug, which for example induced the expression of this protein specifically, you might imagine, based on our work, that that would then lead to programmed cell death in a tumor cell.

"It's been a long-held dogma," he observed, "that there were no other p53-like molecules in the cell, and this says that that's not correct. That there's likely to be at least one other protein in the cell that could do the job, or some of the job, performed by p53."

Research Agenda Seeks Answers

Kaelin and his group are now tackling several basic questions about the still-shadowy p73 gene:

Is it really a tumor suppresor gene?

Are there tumors that do harbor mutations of this gene?

If one disrupts the functions of this gene in an animal model, such as a mouse, will it develop tumors?

A second set of questions concerns the p73 protein that p73 expresses:

What signal or signals normally induce the expression of this protein?

Whether or not it suppresses tumors, what are the protein's normal, non-mutated functions?

Are there specific tissues where it performs some of the functions normally performed by p53?

Or is everything in our paper simply an aberration related to overexpression?

As a tumor suppressor gene gone wrong when mutated, p53 has generated clinical interest in the prognosis of cancer progression, Kaelin observed.

"Secondly," he suggested, "it's provided a conceptual foundation for both the discovery of novel genes implicated in cancer and cell growth and so on, but also for the development of new drugs.

"And the mere fact that p53 plays a pivotal role in inducing apoptosis," Kaelin said, "has prompted many researchers to move from the cell-division aspect of the cancer equation into the cell-death part."

He concluded, "Our results raise two possibilities: first, that there may be additional p53 gene family members; second, that p53 function in carcinomas lacking that tumor suppressor might, in principle, be restored by activating p73."