The identification of a disease gene is often only the first of many steps toward understanding the pathophysiology of a disorder, particularly when a cascade of events leads to the disease phenotype. In Neimann-Pick types A and B there is a primary genetic defect involving sphingomyelin, which results directly in neuralgic dysfunction. This is not the case in Niemann-Pick Type C (NPC), where the gene NPCl, which regulates lysosomal cholesterol metabolism, is mutated1. Here the link between the primary genetic defect and the severe demyelination, which is a hallmark of the disease, is not apparent. We have used cDNA microarray technology2 to identify genes which could be influenced by the primary gene defect and whose altered expression could result in the phenotype. Two arrays were used, an early 1,400 human EST set, and a sequence-verified 5,000 EST set. Fluorescently labelled NPC human fibroblast cDNA (Cy5) and unaffected human fibroblast cDNA (Cy3) were hybridized to the array and ratios of the transcript level of each EST between normal and affected were determined. Transcripts that were outside of the 99.0% confidence interval were chosen on the basis of interesting biology for further consideration. Among these was the gene CD9, which encodes a transmembrane protein which is a major component of the myelin sheath and which is confirmed by northern analysis as significantly downregulated in NPC (20% of normal). FACS analysis confirms the abnormality at the protein level as well. The mouse model of NPC was used to verify the abnormality at the RNA level in vivo. This data suggests a major component of the myelin sheath is directly regulated during cholesterol metabolism and may contribute to the NPC phenotype. Another gene, encoding Galactocerebrosidase, was upregulated in NPC and may indicate a similarity in mechanism between NPC and the two other forms of the disease (A and B). Finally, the entire cholesterol biosynthetic pathway can be seen to be upregulated by looking specifically at enzymes in this pathway in the outlier data set. We are attempting to facilitate elucidation of the link between the primary genetic defect and the molecular pathophysiology of Niemann-Pick type C disease.