Neurofibromas, present in nearly all neurofibromatosis type 1 (NF1) patients, can cause disfigurement or disability and can develop into malignant tumors that are lethal. The factors causing neurofibroma formation are presently unknown. To understand how tumors form and ultimately to halt their formation, several laboratories have developed unique methods to study neurofibroma cells and tissues. Here we are bringing together these investigators to compare and contrast mouse and human NF1 tumor data. Research into the role of specific genes in tumors, as in the case of NF1, previously has been restricted to identification of changes one by one. It is simplistic to believe that complex processes such as tumor formation are mediated through single changes. We expect that the regulation of hundreds of genes and proteins will be involved in neurofibroma formation. Technologic advances have made possible a comprehensive understanding of complex biological processes at the cellular level. First, detailed DNA databases have resulted from genome sequencing. Second, high density "arrays" of individual DNA segments can be placed onto solid supports and used to compare gene expression among many samples. Here, we describe experiments to study gene expression in mouse and human NF1 model systems, using the new method of high throughput functional genomics, microarray analysis of mRNA. Using this approach, thousands of genes can be compared. Genes that are differentially expressed will become targets for further study.

The huge amount of data generated in comparisons of normal and NF1 mutant cells is far beyond the capacity of any single lab to study in detail. Many NF1 labs have begun small-scale experiments to compare slightly different cell populations using functional genomics approaches. Therefore, it is crucial to develop a system that allows sophisticated data analysis and data sharing across laboratories. An important aim of this proposal is to bring together NF1 laboratories to generate and share data and to make the data available to the NF1 community at large. To accomplish this, we are working with a functional genomics core facility and with a biostatistician, both new to the NF1 community, to mine the data. Data mining uses sophisticated computer programs to compare large data sets, so that gene expression patterns unique to neurofibroma cells can be identified.

We anticipate gaining novel and important insights into the undoubtedly complex molecular and cellular mechanisms by which neurofibromas develop. These insights will provide molecular targets for therapeutic approaches.