P&S Journal: Winter 1995, Vol.15, No.1
Wilms' Tumor Gene Research
New findings about a gene implicated in Wilms' tumor could lead to less toxic treatment for the disease and provide further evidence for a completely new mechanism of how cells can become cancerous. The work was published in the July issue of Nature Genetics.
Wilms' tumor is found most often in children between birth and 6 years of age and is one of the most common childhood cancers, with approximately 600 new cases diagnosed each year in the United States. While most children can be treated successfully with surgery and chemotherapy, the treatment produces negative side effects, such as immunosuppression, low blood counts, nausea, bleeding, and baldness.
"The treatment of Wilms' tumor is one of the successes of modern oncology," says Dr. Benjamin Tycko, assistant professor of pathology and senior author of the paper. "But a thorough understanding of the biological basis underlying the formation of the tumor might allow ultimately for the development of new treatments with less severe side effects."
In addition to having potential clinical application for Wilms' tumor patients, Dr. Tycko's research is important because it provides novel insights into the process by which cells become cancerous. The new Wilms' tumor gene, H19, is a member of the hot, new class of cancer-causing genes called tumor suppressors, which means the gene's absence leads to cancer. p53, a tumor suppressor gene recently isolated, for example, is being used to diagnose colorectal cancer.
H19 may define a new class of tumor suppressor gene for two reasons. First, the mechanism for the deletion or inactivation of the normally active copy of the gene involves an unusual phenomenon called genomic imprinting. Second, the H19 gene seems to operate at the level of ribonucleic acid (RNA), rather than as a protein, a cellular strategy not encountered before in other tumor suppressor genes.
Genomic or parental imprinting is a violation of Mendelian inheritance in which each person normally inherits two copies of a given gene, one from each parent. For most genes both are equally functional inside a cell. However, recent studies reveal that in a small subset of genes subject to parental imprinting, only one copy of the pair works. Depending on the particular imprinted gene, the functional one can be maternal or paternal.
Building on earlier work in mice from Dr. Shirley Tilghman's laboratory at Princeton University, Dr. Tycko's group showed that the human H19 gene was subject to imprinting-the paternal gene was silent and the normally active maternal gene is frequently deleted in Wilms' tumor patient DNA. For most tumor suppressor genes, the cancer usually will not form until both copies of the gene are lost or inactivated, the so called "two-hit pathway." In the case of H19, however, only one copy-the active maternal one-needs to be lost, since the paternal copy is already inactive. According to Dr. Tycko, H19 may be an "Achilles' heel" for the cell in terms of cancerous transformation, since it can be inactivated completely in one hit.
In the Nature Genetics paper, Dr. Tycko shows that in addition to actual genetic loss, the gene can be inactivated in Wilms' tumor. He found that the maternal copy of the H19 gene was lost in eight out of 25 tumors taken from Wilms' patients and that in 12 of the remaining 17 tumors, the maternal gene was not lost, but inactivated. The inactivation occurs by a chemical modification called DNA methylation, in which a methyl group consisting of carbon and three hydrogen atoms is added to cytosine, one of DNA's building blocks.
Besides showing that maternal gene loss or inactivation is associated with the cancer, Dr. Tycko's group looked for evidence that H19 loss or inactivation may precede cancer formation and not be one of the many alterations in the cell that can occur after cancerous transformation. In the Nature Genetics paper, his group reports finding that both copies of the H19 genes are sometimes methylated in normal kidney cells near the tumor of Wilms' patients. Rather than methylation occurring after the cells turned into tumors, these results in normal tissues suggest methylation occurred early in the patient's kidney development and hint that methylation actually contributed to the formation of the cancer.
Drugs that remove the abnormal methylation of the maternal copy of H19, in theory, could be used to treat Wilms' tumor, says Dr. Tycko. Azacytidine, a drug used in the past to treat sickle cell disease, works by demethylating DNA and has limited side effects.
The other noteworthy aspect of H19 is that it seems to operate at a level of RNA, rather than coding for protein. RNA normally takes information from a gene and puts it into a form, mRNA, that allows cells to make proteins, the workhorses of cells.
"Throughout its nucleotide sequence there are stop signals that prevent it from being made into a protein," says Dr. Tycko. In the Nature Genetics study, Dr. Tycko found a low level expression of H19 RNA in Wilms' tumor patients, an expected result for a tumor suppressor but never shown before. If suppression is found to act at the level of RNA it would define a new class of tumor suppressor genes. Since H19 RNA does not give rise to any protein, the way in which it affects cell growth is mysterious but may involve interference with function of other RNAs.