Use of gene expression profiling to identify a novel glucocorticoid sensitivity determining gene, BMPRII.
Donn R., Berry A., Stevens A., Farrow S., Betts J., Stevens R., Clayton C., Wang J., Warnock L., Worthington J., Scott L., Graham S., Ray D.
Wide variation in glucocorticoid (Gc) sensitivity exists between individuals which may influence susceptibility to, and treatment response of, inflammatory diseases. To determine a genetic fingerprint of Gc sensitivity 100 healthy human volunteers were polarized into the 10% most Gc-sensitive and 10% most Gc-resistant following a low dose dexamethasone (0.25 mg) suppression test. Gene expression profiling of primary lymphocytes identified the 98 most significantly Gc regulated genes. These genes were used to build a subnetwork of Gc signaling, with 54 genes mapping as nodes, and 6 non-Gc regulated genes inferred as signaling nodes. Twenty four of the 98 genes showed a difference in Gc response in vitro dependent on the Gc sensitivity of their donor individuals in vivo. A predictive model was built using both partial least squares discriminate analysis and support vector machines that predicted donor glucocorticoid sensitivity with 87% accuracy. Discriminating genes included bone morphogenetic protein receptor, type II (BMPRII). Transfection studies showed that BMPRII modulated Gc action. These studies reveal a broad base of gene expression that predicts Gc sensitivity and determine a Gc signaling network in human primary T lymphocytes. Furthermore, this combined gene profiling, and functional analysis approach has identified BMPRII as a modulator of Gc signaling.