Identification of key genes in invasive clinically non-functioning pituitary adenoma by integrating analysis of DNA methylation and mRNA expression profiles.

backgroundTumor surrounding the internal carotid artery or invading to the cavernous sinus is an important characteristic of invasive pituitary adenoma, and a pivotal factor of tumor residue and regrowth. Without specific changes in serum hormone related to the adenohypophyseal cell of origin, clinically non-functioning pituitary adenoma is more likely to be diagnosed at invasive stages compared with functioning pituitary adenoma. The underlying mechanism of tumor invasion remains unknown. In this study, we aimed to identify key genes in tumor invasion by integrating analyses of DNA methylation and gene expression profiles.

methodGenome-wide DNA methylation and mRNA microarray analysis were performed for tumor samples from 68 patients at the Beijing Tiantan Hospital. Differentially expressed genes and methylated probes were identified based on an invasive vs non-invasive grouping. Differentially methylated probes in the promoter region of targeted genes were assessed. Pearson correlation analysis was used to identify genes with a strong association between DNA methylation status and expression levels. Pyrosequencing and RT-PCR were used to validate the methylation status and expression levels of candidate genes, respectively.

resultsA total of 8842 differentially methylated probes, located on 4582 genes, and 661 differentially expressed genes were identified. Both promoter methylation and expression alterations were observed for 115 genes with 58 genes showing a negative correlation between DNA methylation status and expression level. Nineteen genes that exhibited notably negative correlations between DNA methylation and gene expression levels, are involved in various gene ontologies and pathways, or played an important role in different diseases, were regarded as candidate genes. We found an increased methylation with a decreased expression of PHYHD1, LTBR, C22orf42, PRR5, ANKDD1A, RAB13, CAMKV, KIFC3, WNT4 and STAT6, and a decreased methylation with an increased expression of MYBPHL. The methylation status and expression levels of these genes were validated by pyrosequencing and RT-PCR.

conclusionsThe DNA methylation and expression levels of PHYHD1, LTBR, MYBPHL, C22orf42, PRR5, ANKDD1A, RAB13, CAMKV, KIFC3, WNT4 and STAT6 are associated with tumor invasion, and these genes may become the potential genes for targeted therapy.