Recent Insights into the Functions

of the von Hippel-Lindau Protein

-- William G. Kaelin, Jr., M.D., Ph.D., Dana-Farber Cancer Institute, Boston, Massachusetts

Our group is specifically interested in how the normal von Hippel-Lindau protein (called pVHL for short) prevents tumor development. Every protein is made up of building blocks, called amino acids, hooked together in a particular sequence. The sequence of amino acids found in pVHL does not closely resemble the sequence of any known protein. Thus, the sequence of pVHL does not provide any immediate clues as to what job(s) it might perform.

We have already shown that reintroducing a normal version of pVHL into kidney cancer cells will prevent them from forming tumors in mice. To try to understand how pVHL might operate, we first immunized mice and rabbits with pVHL. These animals then generated antibodies which would bind specifically to pVHL. Using these antibodies we were able to identify pVHL in normal cells. Furthermore, we determined that pVHL "lives" primarily in a compartment of the cell known as the cytoplasm.

Just as you can tell a lot about a person from the company he or she keeps, so too you can learn a lot about a protein from the other proteins it bind with, or "talks to". Using our antibodies, we have determined that normal pVHL binds to about five other cellular proteins. Our group, as well as a group led by Dr. Richard Klausner, director of the U.S. National Cancer Institute, have shown that two of these proteins are proteins called elongin B and elongin C. When these two proteins bind to a third protein, called elongin A, they generate an enzyme which can "turn on" certain genes. pVHL competes with elongin A for binding to B and C, thereby inhibiting this enzyme activity.

These observations suggested that at least one pVHL job is to regulate which genes are "on" and "off" in a cell. Tumors which arise in VHL families contain many blood vessels and secrete factors, such as vascular endothelial growth factor (VEGF for short), which actually stimulate blood vessel formation. The genes which are the "blueprints" for these factors are normally "off" unless cells are deprived of oxygen. We have now shown that cells which lack pVHL are unable to turn off these genes. Similar observations have also been made by Marston Linehan's group an Dieter Marme's group. Thus, cells which have VHL mutations and lack a normal copy of pVHL behave as though they are deprived of oxygen even when they are not.

In the simplest view, inhibition of elongin, at least indirectly, inhibits the activation of these oxygen-regulated genes. Our laboratory is currently doing experiments to see if this is the case. Specifically, we are engineering cells so that we can selectively inhibit the elongin enzyme. We can then ask whether this is sufficient to turn off genes such as VEGF. If so, we can think about designing or developing drugs which will likewise inhibit the elongin enzyme. If we are lucky, these drugs might mimic the ability of pVHL to suppress tumor growth. In the meantime, several companies are developing drugs which will inhibit the ability of VEGF to induce blood vessel formation. It is expected that these drugs will go into clinical trials next year. The obvious hope is that these drugs will have an effect on the development and proliferation of tumors in VHL patients.

As published in the VHL Family Forum, 4:4, December 1996. For permission to reprint, please contact the VHL Family Alliance, editor@vhl.org. Further information is available from the VHL Family Alliance, info@vhl.org.