Demethylation status of somatic DNA extracted from pituitary neuroendocrine tumors indicates proliferative behavior.

Cytosine intermediers 5-methylcytosine (5mC) and 5-hydroxymethylcytosine (5hmC) as epigenetic hallmarks have never been investigated in pituitary neuroendocrine tumors (PitNET).To examine methylation-demethylation status of global DNA in PitNET tissues and to assess its correlation with clinical and biological parameters.Alltogether, 57 PitNET and 25 corresponding plasma samples were collected. 5mC and 5hmC were investigated using liquid chromatography-tandem mass spectrometry. Expression of DNA Methyltransferase 1 (DNMT1), Tet Methylcytosine Dioxygenase 1-3 (TET1-3) and Ubiquitin-like with containing PHD and RING Finger domains 1-2 (UHRF1-2) was measured by RT-qPCR. Level of 5hmC and UHRF1-2 was explored by immunohistochemistry too. Effect of demethylating agent decitabine was tested on pituitary cell lines.5hmC/5mC ratio was higher in less differentiated PitNET samples. A negative correlation between Ki-67 proliferation index and 5hmC, 5hmC/5mC ratio were revealed. Higher 5mC was observed in SF-1+ gonadotroph adenomas with higher Ki-67 index. Expressions of TET2 and TET3 were significantly higher in adenomas with higher proliferation rate. UHRF1 showed gradually increased expression in higher proliferative adenoma samples and a significant positive correlation was detected between UHRF2 expression and 5hmC level. Decitabine treatment significantly decreased 5mC and increased 5hmC levels in both cell lines accompanied with decreased cell viability and proliferation.Demethylation process negatively correlated with proliferation rate and 5hmC/5mC was higher in less differentiated adenomas. Hence epigenetic markers can be potential biomarkers for PitNET behaviour. Altering epigenome in adenoma cells by decitabine decreased proliferation suggesting that this treatment might be a novel medical treatment for PitNET.

Click here to read the full article @ The Journal of clinical endocrinology and metabolism