Integrated multi-omics profiling of nonfunctioning pituitary adenomas.

PURPOSE Genetic and epigenetic alterations are involved in pituitary adenoma pathogenesis, however the molecular basis of proliferative nonfunctioning pituitary adenomas (NFPAs) remains unclear. Here, we analyzed integrated multi-omics profiling including copy number variation (CNV), DNA methylation and gene expression of 8 NFPAs. METHODS We collected 4 highly proliferative (hpNFPA, Ki-67 ≥ 3) and 4 lowly proliferative (Ki-67 ≤ 1) NFPAs, and comprehensively assessed CNV, DNA methylation, and gene expression by Illumina HumanMethylation450 BeadChip and Affymetrix GeneChip PrimeView Human Gene Expression Array. We performed Ingenuity Pathway Analysis (IPA) for differentially expressed genes to illustrate aberrant pathways and delineated protein-protein networks of selected key genes in dysregulated pathways. RESULTS Aberrant arm level CNV, dysregulated DNA methylation, and associated impacts on gene expressions were observed in 2 early occurring hpNFPAs. Chromosomal losses were associated with attenuated expression of DNA methyltransferases, further altering global methylation in these 2 samples. Correlation analysis between DNA methylation and gene expression in 8 NFPAs indicates methylation in promoter and gene body regions are both involved in gene regulation. IPA showed PPARα/RXRα, dopamine receptor signaling, cAMP-mediated signaling, and calcium signaling were all activated, while p38 MAPK and ERK5 signaling were inhibited in hpNFPAs. Moreover, selected key gene networks in hpNFPAs exhibited concurrent methylation status and expression levels of adenylate cyclase genes, G protein subunits, HLA genes, CXCL12, and CCL2. CONCLUSION This study presents comprehensive multi-omics views of CNV, DNA methylation, and gene expression in 8 NFPAs. Pathway analysis and network maps of key genes provide clues to elucidate the molecular basis of hpNFPA.