Interleukin-6 mediated upregulation of CYP1B1 and CYP2E1 in colorectal cancer involves DNA methylation, miR27b and STAT3.

BACKGROUND: The pro-inflammatory cytokine interleukin-6 (IL6) promotes colorectal cancer (CRC) development. It is also known to regulate cytochrome P450 (CYP450) enzymes, which are involved in CRC tumour initiation and promotion via activation of chemical carcinogens. Here, IL6 regulation of CYP450 expression was investigated in CRC. METHODS: The effect of IL6 on CYP 1A1, 1B1 and 2E1 expression was determined in vitro using CRC cell lines HCT116 and SW480, and CYP450 expression was determined by immunohistochemistry in CRC tissues previously shown to have increased levels of IL6. RESULTS: In mechanistic studies, IL6 treatment significantly induced CYP1B1 and CYP2E1, but not CYP1A1, gene expression in HCT116 and SW480 cells. CYP2E1 expression regulation occurred via a transcriptional mechanism involving STAT3. For CYP1B1 regulation, IL6 downregulated the CYP1B1-targeting microRNA miR27b through a mechanism involving DNA methylation. In clinical samples, the expression of CYP1B1 and CYP2E1, but not CYP1A1, was significantly increased in malignant tissue overexpressing IL6 compared with matched adjacent normal tissue. CONCLUSIONS: Colonic inflammation with the presence of IL6 associated with neoplastic tissue can alter metabolic competency of epithelial cells by manipulating CYP2E1 and CYP1B1 expression through transcriptional and epigenetic mechanisms. This can lead to increased activation of dietary carcinogens and DNA damage, thus promoting colorectal carcinogenesis.

Upregulation of cyclooxygenase-2 is accompanied by increased expression of nuclear factor-kappa B and I kappa B kinase-alpha in human colorectal cancer epithelial cells.

Cyclooxygenase-2 (COX-2) is selectively overexpressed in colorectal tumours. The mechanism of COX-2 induction is not fully understood, but requires de novo messenger RNA and protein synthesis, indicating regulation at the transcriptional level. Sequence analysis of the 5'-flanking region of the COX-2 gene shows two nuclear factor-kappa B (NF-kappa B) sites. Inhibition of this protein in model cell culture systems attenuates COX-2 expression and implies that NF-kappa B plays an important role in COX-2 induction. We measured COX-2, NF-kappa B and I kappa B kinase alpha (IKK alpha) protein expression in matched colonic biopsy samples comprising both nontumour and adjacent tumour tissue from 32 colorectal cancer patients using immunohistochemistry. There was none or very little expression of COX-2, NF-kappa B and IKK alpha in non-neoplastic colon epithelial cells, while the expression of all three of these proteins was significantly increased (P<0.05, Wilcoxon's signed rank test) in adjacent cancerous cells. Moreover, all three proteins were found to be coexpressed in the neoplastic epithelium, with the expression of COX-2 and NF-kappa B highly correlated (Pearson's correlation, P<0.005). There was no apparent correlation between enhanced COX-2, NF-kappa B or IKK alpha expression and tumour Dukes' stages. Our results are compatible with the hypothesis that IKK alpha and NF-kappa B are involved in COX-2 induction in these tumours and the lack of association between COX-2 expression and severity of disease as measured by Dukes' stage is consistent with the proposal that COX-2 expression is an early postinitiation event.