Variation in Gene Expression in Neurofibromatosis Type 1

National Human Genome Research Institute

Investigators

Linked publications & trials

Abstract

Neurofibromatosis type 1 (NF1) is a common (approximately 100,000 Americans) genetic disorder of dysregulated cell growth. Affected people can develop multiple (tens, hundreds, or thousands) of soft fleshy tumors called neurofibromas. People with NF1 also develop malignant cancers. There are limited therapies and no cures for NF1. At this time it is essentially impossible to predict the severity of the disorder. With few exceptions, neither mutation testing of the involved gene (NF1) nor the severity of other affected family members help in estimating the clinical course of the condition. In this study we seek to identify the genes that influence the wide variability in severity seen in patients with NF1. We apply the principles of translational medicine to find genetic modifiers of severity in NF1. To do so, we quantitatively evaluate families with NF1 using a variety of methods (physical exam, magnetic resonance imaging (MRI), skin and eye photography, echocardiography, dental evaluation). In our study, we have correlated differences in these measurements with differences in the expression of genes, as determined by a microarray. We have identified numerous genes whose expression correlates with phenotypic severity (e.g. number of caf-au-lait spots, a clinical feature of NF1). We call these correlates putative genetic modifiers. These candidate genes are being sequenced to identify variants. We expect that the sequenced variants (seen across the population) may be useful in predicting the severity of the disorder. It is our hope that our approach to the identification of genetic modifiers in NF1 will serve as a paradigm to identify novel therapeutic targets and to serve as the basis for personalized medicine. Dr. Stewart joined the National Human Genome Research Institute at the end of November 2004. The IRB approved our project in April 2005 and recruitment started shortly thereafter. By the end of fiscal 2010, we had phenotyped approximately 135 individuals affected with NF1 and collected DNA from many of the parents. In a novel screen, we cultured 79 lymphoblastoid cell lines (LCLs) from our phenotyped group of individuals affected with NF1, harvested RNA and hybridized it to microarrays (expression profiling). (This was also done on 25 controls.) We then examined the correlation between a variety of phenotypes (height, number of caf-au-lait spots, tumor burden, etc) and each of the approximately 24,000 genes expressed on the array. We subjected approximately 30 genes whose expression value correlated with phenotypic severity to quantitative PCR (qPCR). We identified several genes whose phenotype/gene expression correlate was still significant after qPCR. From this group, there were numerous biologically plausible candidates. We are sequenced (exons, introns in part, conserved non-coding sequence, plus 800 bp 5 and 3 to the gene) five of the qPCR-verified candidate genes. Our goal is to identify specific alleles (or haplotypes) in each gene that correlates with a specific degree of severity. As a complementary approach to identifying genetic modifiers, we are collaborating with the NF1 group at the Massachusetts General Hospital in Boston. We provided 50 samples to be genotyped using the Illumina 1M Quad platform (>1 million SNPs). We are genotyping 100 phenotyped samples at NHGRI with the same platform. Our goal is to use association to identify genetic modifiers. We have also developed a collaborative project (using funds from the Bench-to-Bedside program) with Dr. Brigitte Widemann and Dr. Thomas Hornyak of the National Cancer Institute to investigate the growth and biology of neurofibromas. The project, titled Natural history and biology of dermal neurofibromas in neurofibromatosis type 1 was approved by the NHGRI IRB in May 2006. It is designed to investigate the growth rate and rate of appearance of dermal neurofibromas using several imaging modalities. We will also biopsy a dermal neurofibroma and normal skin. To streamline recruiting, the eligibility criteria overlap with that of our primary project, Variation in Gene Expression in Neurofibromatosis Type 1. By the end of fiscal 2010, we finished the evaluation, longitudinally, of the growth rate of dermal neurofibromas from 12 individuals affected with NF1 and were analyzing the data. Since Dr. Stewart is trained as an internist, he has a special interest in manifestations of NF1 affecting adults. To this end, we have been evaluating individuals with NF1 with unique and under-recognized disease features in the adult. By the end of fiscal 2010 we had developed a multi-disciplinary group of clinicians at the NIH Clinical Center with expertise in the evaluation and resection of glomus tumors. These tumors are uncommon but painful benign tumors in the fingertips. A paper describing the clinical manifestations of glomus tumors was published in the Journal of Medical Genetics in FY 2010. A paper describing the pain complications (and their management) from glomus tumors was submitted in FY 2010. A paper describing the genome-wide changes in gene expression secondary to NF1 haploinsufficiency was published in BMC Genomics. We have submitted a paper on the cognitive profile of patients with Legius syndrome, a novel disorder similar to NF1. We presented the results of our novel screen to identify genetic modifiers as a platform presentation at the 2009 annual meeting of the American Society of Human Genetics, and the 2010 annual meeting of the Children's Tumor Foundation. Dr. Stewart was also invited to give presentations at the National Cancer Institute, Johns Hopkins University, and Virginia Commonwealth University.