Mitochondrial STAT3 contributes to transformation of barrett’s epithelial cells that express oncogenic ras in a p53-independent fashion

Metaplastic epithelial cells of Barrett’s esophagus transformed by the combination of p53-knockdown and oncogenic Ras expression are known to activate signal transducer and activator of transcription 3 (STAT3). When phosphorylated at tyrosine 705 (Tyr705), STAT3 functions as a nuclear transcription factor that can contribute to oncogenesis. STAT3 phosphorylated at serine 727 (Ser727) localizes in mitochondria, but little is known about mitochondrial STAT3’s contribution to carcinogenesis in Barrett’s esophagus, which is the focus of this study. We introduced a constitutively active variant of human STAT3 (STAT3CA) into the following: 1) non-neoplastic Barrett’s (BAR-T) cells; 2) BAR-T cells with p53 knockdown; and 3) BAR-T cells that express oncogenic H-Ras^G12V. STAT3CA transformed only the H-Ras^G12V-expressing BAR-T cells (evidenced by loss of contact inhibition, formation of colonies in soft agar, and generation of tumors in immunodeficient mice), and did so in a p53-independent fashion. The transformed cells had elevated levels of both mitochondrial (Ser727) and nuclear (Tyr705) phospho-STAT3. Introduction of a STAT3CA construct with a mutated tyrosine phosphorylation site into H-Ras^G12V-expressing Barrett’s cells resulted in high levels of mitochondrial phospho-STAT3 (Ser727) with little or no nuclear phospho-STAT3 (Tyr705), and the cells still formed tumors in immunodeficient mice. Thus tyrosine phosphorylation of STAT3 is not required for tumor formation in Ras-expressing Barrett’s cells. We conclude that mitochondrial STAT3 (Ser727) can contribute to oncogenesis in Barrett’s cells that express oncogenic Ras. These findings suggest that agents targeting STAT3 might be useful for chemoprevention in patients with Barrett’s esophagus.