PURPOSE: Cyclooxygenase-2 (COX-2) promotes carcinogenesis and its expression associates with clinicopathologic characteristics in gastric cancer. HuR is an mRNA binding protein that controls the stability of certain transcripts including COX-2. We evaluated the prognostic significance of COX-2 and HuR expressions in gastric cancer and whether there exists a link between HuR and COX-2 expressions. EXPERIMENTAL DESIGN: The study included 342 consecutive patients with histologically confirmed gastric adenocarcinoma, of whom 321 patients had tissue specimens available for COX-2 and 316 for HuR immunohistochemistry. Specimens were stained by COX-2- and HuR-specific monoclonal antibodies and scored by two independent observers. Correlation to clinical data and survival was assessed. TMK-1 gastric adenocarcinoma cells were treated with small interfering RNA against HuR and expressions of HuR and COX-2 were detected by immunofluorescence and Western blot analysis. RESULTS: Patients with low COX-2 expression had a cumulative 5-year survival of 53% and those with high COX-2 expression had 16% (P < 0.0001). In multivariate analysis, COX-2 was an independent prognostic factor (P = 0.003). Cytoplasmic HuR expression was associated with high COX-2 expression (P < 0.0001) and with reduced survival (P = 0.004) whereas nuclear positivity for HuR was not. When TMK-1 cells were treated with HuR small interfering RNA, expressions of HuR and COX-2 were reduced. CONCLUSIONS: High COX-2 is an independent prognostic factor in gastric cancer. Cytoplasmic expression of HuR associates with high COX-2 expression and with reduced survival, and tissue culture experiments show that HuR can regulate expression of COX-2 in gastric cancer cells.
PURPOSE:Cyclooxygenase-2 (COX-2) promotes carcinogenesis and its expression associates with clinicopathologic characteristics in gastric cancer. HuR is an mRNA binding protein that controls the stability of certain transcripts including COX-2. We evaluated the prognostic significance of COX-2 and HuR expressions in gastric cancer and whether there exists a link between HuR and COX-2 expressions. EXPERIMENTAL DESIGN: The study included 342 consecutive patients with histologically confirmed gastric adenocarcinoma, of whom 321 patients had tissue specimens available for COX-2 and 316 for HuR immunohistochemistry. Specimens were stained by COX-2- and HuR-specific monoclonal antibodies and scored by two independent observers. Correlation to clinical data and survival was assessed. TMK-1 gastric adenocarcinoma cells were treated with small interfering RNA against HuR and expressions of HuR and COX-2 were detected by immunofluorescence and Western blot analysis. RESULTS:Patients with low COX-2 expression had a cumulative 5-year survival of 53% and those with high COX-2 expression had 16% (P < 0.0001). In multivariate analysis, COX-2 was an independent prognostic factor (P = 0.003). Cytoplasmic HuR expression was associated with high COX-2 expression (P < 0.0001) and with reduced survival (P = 0.004) whereas nuclear positivity for HuR was not. When TMK-1 cells were treated with HuR small interfering RNA, expressions of HuR and COX-2 were reduced. CONCLUSIONS: High COX-2 is an independent prognostic factor in gastric cancer. Cytoplasmic expression of HuR associates with high COX-2 expression and with reduced survival, and tissue culture experiments show that HuR can regulate expression of COX-2 in gastric cancer cells.
Authors: Anke Doller; El-Sayed Akool; Andrea Huwiler; Roswitha Müller; Heinfried H Radeke; Josef Pfeilschifter; Wolfgang Eberhardt Journal: Mol Cell Biol Date: 2008-02-19 Impact factor: 4.272