By David N. Leff

Do fruit flies get tumors? Yes, but who cares?

Molecular oncologists should, but many do not. There seems to be a politically incorrect species put-down of Drosophila melanogaster in the search for genes of possible use in fighting cancer. Molecular oncologist Christopher Kemp, at the Fred Hutchinson Cancer Research Center, in Seattle, put the snub this way:

"Some 50 tumor-suppressor genes have been discovered in Drosophila, two dozen have been cloned, and at least nine show clear homology to mammalian genes," he said. "Due in part to anthropocentrism, [i.e., Homo sapiens' snobbery], these studies have not gained broad acceptance or notice in the mammalian cancer genetics community."

Kemp was commenting on a two-part paper in the February 1999 issue of Nature Genetics. One bears the title "Human homologue of the Drosophila melanogaster lats tumour suppressor modulates CDC2 activity." The second article's title reads "Mice deficient of Lats1 develop soft-tissue sarcomas, ovarian tumours and pituitary dysfunction." The senior author of both papers is cancer geneticist Tian Xu, of the Howard Hughes Medical Institute, at Yale University, in New Haven, Conn.

Xu and his co-authors don't spurn fruit flies as a potential source of anticancer agents. On the contrary, among the insects' genes they have identified one they named lats - standing for "large tumor suppressor." It encodes a putative serine/threonine kinase, which, when mutationally deleted, forms large, aberrantly proliferating cancers on many fruit-fly tissues. "It was not clear previously," Xu told BioWorld Today, "whether these fly genes are directly relevant to tumorigenesis in mammals."

He and his group went on to clone the human equivalent of the insect Lats1 gene, and showed that its product binds the crucial cell-cycle regulator CDC2.

Fly, Humans Shared Ancestor 800 Million Years Ago

When they introduced the unmutated product of the intact human gene into the genome of flies lacking the Drosophila version of Lats, it prevented tumor formation in the insects. This suggested a considerable degree of genomic resemblance between the gene sequence in fly and man. "Human LATS1 and fly LATS," they report, "share 74 percent sequence identity (86 percent similarity) in the kinase-containing carboxy-terminal domain."

Furthermore, a database search "revealed LATS-like molecules from Caenorhabditis elegans [the nematode round worm] and Saccharomyces cerevisiae [yeast], suggesting that members of this protein family might exist in all eukaryotes."

This congruence, Kemp observed, "in addition to the interesting philosophical issues, further indicates the conservation of the tumor-suppressing function of LATS through the 800 million years since these phylogenetic lines diverged." He added that Xu and his co-authors show, "apparently for the first time, that a gene functions as a tumor suppressor in both invertebrates and vertebrates."

The human LATS1 gene, the Yale group found, was expressed in all fly fetal and adult tissues they examined, with highest expression in the adult ovary. The co-authors generated knockout mice lacking mammalian LATS by disrupting murine Lats gene sequences expressing a total of 374 amino acids. In these animals, the ovary was only one of several organ systems that developed cancers. However, the non-malignant ovarian tumors, occurring when the mice were three months of age, grew in stromal cells, rather than the epithelial cells that mark most human cancers of the ovary.

Besides the double-minus homozygotic knockouts, the mouse progeny also included plus/minus heterozygotes, harboring only one Lats allele [parental gene]. When the team mated two of these plus/minus mice, they expected the resulting offspring to reflect Mendelian inheritance - 50 percent plus/plus wild types and 25 percent each, plus/minus heterozygotes and minus/minus homozygotes. But only eight percent of the pups assessed at three weeks of age were minus/minus, rather than the 25 percent expected. This anomaly cleared up when the co-authors looked at fetuses just before their birth, and found live homozygous embryos at the normal Mendelian frequency.

LATS-Lacking Mice Suffer Tumors, Deformities

More than half of these - 60 of 101 - died of undetermined causes. The survivors were growth-retarded, attaining weights averaging only 70 percent that of wild-type litter mates. Both sexes had diminished fertility, reaching 60 percent in females. Besides deformed and tumorous ovaries, they also suffered mammary gland dysfunction, such as absent nipples, and widespread endocrine aberration.

No tumors arose in 80 wild-type control mice or, to date, in 152 heterozygous animals.

But the LATS1-minus homozygotes proved highly sensitive to artificial tumors induced by strongly carcinogenic chemicals, such as dimethylbenzanthracene, or DMBA. In response to this treatment, they developed large soft-tissue sarcomas not often occurring in humans. This indicates to Xu and his co-authors that the Lats1 gene is a potent tumor suppressor in mammals.

Moreover, he pointed out, it's a previously unknown kind of tumor suppressor, which blocks a specific stage of cell proliferation. "These LATS molecules," Xu said, "are a new type of 'negative' regulator for the enzymes that drive the cell cycle." He hinted that further study of this mechanism could lead to new cancer therapies.

"Most importantly," he pointed out, "if you look at what kills cancer patients, it's either metastases or tumor cells resistant to radiation or chemotherapy, owing to defects in their cell cycle. CDC2 is the major enzyme promoting progression of the G2/M stage of the cycle. We don't know as much about G2/M as we do about the G1/S stage. So, our findings should provide new insight into the mechanism of cancer biology."

Their paper points out that "[just] as mice deficient for homologues of known human tumor suppressors develop tumors, our data suggest that inactivating LATS1 will result in tumor development in humans."

Xu concluded: "Our work adds to the evidence indicating that organisms such as Drosophila can serve as excellent models for many aspects of cancer biology."

Kemp agreed. His "News & Views" editorial is headed "You don't need a backbone to carry a tumour suppressor gene."