NFRP 2012 Investigator Vignette

Title: Identifying Modifier Genes in the Drosophila NF1 Orthologue

Investigator: Andre Bernards, PhD; MGH Center for Cancer Research, Harvard Medical School

Since 1991 I have been actively involved in Neurofibromatosis Type 1 research. The NFRP has been instrumental in allowing me to focus on this-on this one question-what are modifier genes, which I think is the critical question in NF1 research and which is-it's been a long slog to-to successfully-to bring to a successful conclusion.

Neurofibromatosis Type 1 is a very common autosomal dominant genetic disease. One in 3,000 patients are affected by this disease so about 2 million patients worldwide. Its hallmarks are benign tumors of the peripheral nervous system mostly including malignant peripheral nerve sheath tumors. Patients often have skeletal abnormalities, vascular abnormalities, skin pigmentation defects. At least 50% of children with NF1 have learning difficulties. It's what clinicians call a multisystem disorder. There is an enormous variability in the expressivity of the disease, so you can either have just a few tumors, benign tumors, or you can have thousands of these benign tumors. And it's clear that modifier genes play an important role in determining how severely affected you will be. And that actually has been the underlying theme of virtually all of my research; we have been trying to identify what these modifier genes are. The reason why we are interested in these modifier genes is that by definition they control rate-limiting steps during symptom development. And genes that control rate-limiting steps are very good therapeutic targets. If you manage to inhibit those genes, you can slow down the development of the disease.

Now back in the 1990s I identified a fruit fly, Drosophila melanogaster NF1 orthologue because in fruit flies you can easily identify modifier genes. Well, I say easily...it's taken us almost 20 years to come up with a complete list of modifier genes in flies, but it's doable. There are tools to specifically knock out the expression of any gene in the Drosophila genome, tissues specifically in a temporarily controlled manner. So the tools that are available for Drosophila genetic research are absolutely second to none.

If you knock out NF1 in flies, you get a number of phenotypes that are recognizable but they're subtle. There are circadian rhythm defects where flies are arrhythmic in a day/night cycle. There is a learning difficulty in the NF1 mutant flies similar to what you see in patients. There is an overall growth deficiency so the flies are a bit smaller than wild types. But the flies don't get tumors. And the flies don't get anything really immediately obvious that you could easily base a modifier screen on.

All that we have to go on is the reduced size of the flies and reduced size is a phenotype that is usable in a genetic modifier screen but it's not ideal because size is a sexually dimorphic phenotype in flies. Females are much bigger than males. So it's something that you have to very carefully use in the screen otherwise you'll start barking up the wrong tree fairly rapidly.

So we've learned how to do this; we have done a very comprehensive screen of 60% of the fly genome and we have identified dozens of modifiers. Some of them look very, very intriguing, including a neuronal receptor tyrosine kinase called ALK, which if we pharmacologically or genetically interfere with the function of this kinase we can rescue the Drosophila melanogaster size defect but we can also completely restore the Drosophila on the learning defect. So then of course the next obvious question becomes is this relevant to the human disease if we inhibit the human ALK kinase. Does that do the same thing? And that's of course what we want to answer next. We haven't answered it yet but we do have preliminary evidence that the same ALK NF1 pathway exists in human cells. So that looks very hopeful.

Now I like to work on projects where I have some hope that I can actually solve them. So that's always attracted me to NF1 and I think in the end maybe we'll get to a point where we can slow down tumorigenesis; we can do something about the learning difficulties, and we can make life for these patients a lot better.