VHL Tumor Suppressor Gene Found!

In the May 28, 1993, issue of Science magazine, an international genetic research team reports that it has located the von Hippel-Lindau Disease Tumor Suppressor Gene.

Congratulations go to Farida Latif, Masahiro Yao, Mary Lou Orcutt, Igor Kuzmin, Fangwei Zhou, Berton Zbar, Michael I. Lerman. and Michael Dean of the National Cancer Institute, Frederick, Maryland; Kalman Tory, Fuh-Mei Duh, Thomas Stackhouse, William Modi, Laura Geil, Laura Schmidt, Hua Li, Ming Hui Wei, Fan Chen, and Damjan Glavac of Program Resources Inc./DynCorp., Frederick, Maryland; James Gnarra, McClellan M. Walther, Yongkai Weng, Dah-Shuhn R. Duan, W. Marston Linehan, Gladys Glenn, and Peter Choyke of the National Cancer Institute, Bethesda, Maryland; Frances M. Richards, Paul A. Crossey, Malcolm A. Ferguson-Smith, and Eamonn R. Maher of Cambridge University, Cambridge, England; Denis Le Paslier, Ilya Chumakov, and Daniel Cohen of the Centre d'Étude du Polymorphisme Humain, Paris, France; and A. Craig Chinault of the Institute for Molecular Genetics, Baylor College of Medicine, Houston, Texas.

"They definitely have it and it's very exciting. It means real help for the von Hippel-Lindau families," says Wayne State University molecular geneticist David I. Smith, who was also racing to find the gene.¹ This achievement builds on a great deal of work done not only by this team but by many others whose published works along the way provided pieces of the puzzle. Credit is also due to the many families affected with VHL who have contributed the blood and tissue samples used in this research.

The VHL gene is "an important one to know about. It's really nice to have it join the club," says Alfred Knudson of Fox Chase Cancer Research Center in Philadelphia, whose work more than two decades ago laid the foundation for current tumor suppressor gene research.¹

Using a number of different techniques this team was able to map almost completely the relevant area of chromosome 3. They found "nested constitutional deletions" in three unrelated VHL patients. This finding and the availability of cloned DNA accelerated the speed of their search for the VHL gene. "We reasoned," says Dr. Latif, "that the smallest of these three deletions should either encompass or interrupt the gene." Like a Russian doll that gets smaller and smaller, explains Dr. Zbar,¹ the study of these nested deletions narrowed the location of the gene.

They used probes³ to determine what genes were in this area, and began to work primarily with two genes, referred to as g6 and g7. They then tested blood samples from VHL patients to see which of these were present. They found 120 VHL patients in which g6 was not involved.

"By contrast," says Dr. Latif, "g7 proved to be a strong candidate for the VHL gene." They found that g7 was expressed in a variety of VHL tumor tissues, including brain and kidney tissues.

"Furthermore Southern (DNA) analysis showed that the g7 sequence is highly conserved across species as diverse as mammals, drosophila [fruit fly], and sea urchin." Evidence that this genetic material has changed very little in the course of evolution suggests that it performs a very basic cellular function.

Finally they tested for deletions in the g7 gene in the blood of 221 unrelated VHL patients, including eight patients classified as "new mutations." They found deletions in at least 18%. The entire gene is 6 to 6.5 kilobases in length; the identified sequence is 1.8 kilobases in length.

They were able to confirm that the VHL gene behaves as a typical tumor suppressor gene, as defined in Knudson's theory of human carcinogenesis². According to that theory, a tumor suppressor gene normally keep tumors from forming. It is the loss or inactivation of one or more of these tumor suppressors that leads to cancer. If this is the case, they reasoned, then this altered form of the gene would be present in VHL tumor tissue.

They decided to look and see if there were deletions or mutations in the g7 genes in random or sporadic renal cell carcinoma (RCC) in the general population.

"We chose to study RCC because Knudson's model predicts that sporadic cancers should be associated with mutations in the same loci [locations] affected in the corresponding hereditary cancer."

A team under Dr. Linehan looked at a number of samples of tissue from random RCC tumors. They identified small mutations that created frameshift errors, and presumably result in the building of the wrong protein in the body. In another set of patterns there was a "nonsense mutation" where two components were reversed (instead of CA, it was AC).

In VHL patients they also saw deletions or mutations among the samples, which supported the conclusion that g7 is the VHL tumor suppressor gene and that it has a role in some sporadic RCC.

They compared this string to genes and proteins already registered in the databases of genetic information, and found no immediate matches. "It already looks like it will be a new type of tumor suppressor gene," says Dr. Lerman.¹

The sequence obtained so far in the g7 gene shows that the gene encodes 284 amino acids, the building blocks of proteins. There may yet be more. There is one known protein which contains the pattern seen so far, one which is involved in the passing of signals among cells, telling them, for example, when to multiply or not. Evidence indicates, though, that they may be looking at a new protein, not previously identified. "The presence of this acidic [pattern] suggests that the VHL protein may be localized in the cell membrane and may be involved in signal transduction or cell adhesion. Further clues to the function of the VHL protein may emerge as more sequence information is obtained."

What does this mean to us now? The DNA testing which we reported in March uses indirect testing, analyzing multiple samples in a family to determine a repeated pattern associated with VHL which is unique to each family. This new information allows us to use a method of direct testing, which is easier to do and less prone to errors. In the families for which this testing works, it is much more reliable, and it can be used for individuals without other affected family members; but as explained above, it will only apply immediately to about 18% of families. When the rest of the VHL gene has been sequenced, the percentage of people for whom this test works will increase.

Researchers are hoping to have the gene's full sequence within the year, and a 100% accurate diagnostic test for VHL should soon follow. Dr. Zbar also hopes to be able to correlate specific mutations on the gene with the distinct forms of cancer that strike different VHL families. This could greatly improve patient monitoring by suggesting which tumors to look for most carefully.¹

The DNA testing laboratories are gearing up to do this new testing. Requests for testing may be submitted by a genetic counseling center to one of the testing centers listed on the VHLFA website. [Editor's note: This paragraph updated 12/2000.]

This article is based on Latif et al, "Identification of the von Hippel-Lindau Disease Tumor Suppressor Gene," Science 260, 1317-1320, 28 May 1993. Prepared with the kind assistance of Dr. Frances Richards, Ms. Corinne Boehm, Dr. Berton Zbar, and Dr. Gladys Glenn.

1. As quoted in John Travis, "New Tumor Suppressor Gene Captured," Science 260, 1235, 28 May 1993.

2. Knudson's theory is presented in A. Knudson, Proc. Natl. Acad. Sci. U.S.A. 68, 820 (1971); Annu. Rev. Genet. 20, 231 (1986).

3. To understand the terms in this article you will want to read our articles "Tracking Down the VHL Gene" and "Why so Many Errors in our DNA?" in the March 1993 issue of VHL Family Forum. q