For comparisons among three groups of quantitative variables, the Kruskal Wallis test was used. In cases where there GSI-IX was a significant dif ference between the groups, a pairwise comparison was carried out by adjusting the alpha risk by the method of Bonferroni. Inhibitors,Modulators,Libraries Students t tests were performed when indi Inhibitors,Modulators,Libraries cated in figures legends. All statistical analyses were car ried out by the Department of biostatistics, epidemiology, public Health and medical information of the N?mes University Hospital using the SAS software. Results Expression of hPXR in colon tissues and colon cancer cells To examine whether hPXR is expressed in human colon in vivo, we analyzed its mRNA expression in both nor mal and neoplastic human colon tissues. Human liver tissues, which are known to have high PXR expression, were used as positives controls.
As shown in figure 1A, PXR mRNA was detected in both normal and cancerous human colon tissues. In our panel Inhibitors,Modulators,Libraries of 14 patients, PXR mRNA was more abundant in some colon tissues than in liver biopsies, and displayed a greater variability in colon tissues than in liver tissues. Note that, although we did observe some quantitative differences between the expression of PXR in normal tissues and their matched colon tumour samples, we found no clear trend of tendency. In addition, we observed a very low expression of PXR mRNA in both colon can cer cells lines LS174T, SW480, SW620, or HT29 and hepatic cell lines HepG2 and HuH7. This result on cell lines is in accordance with the previous observation that cultured cells often loose metabolic abilities and PXR expression.
Representative exam ples of PXR immunostaining are Inhibitors,Modulators,Libraries displayed in figure 1D, liver were used as positive control for PXR expression and negative controls without primary antibody are shown. We observed a strong immu nostaining of PXR in both normal colon and tumors consistent with mRNA results. Functional characterization of PXR transfected LS174T Inhibitors,Modulators,Libraries colorectal cancer cells To decipher the role of PXR in irinotecan mediated toxicity in colorectal cancer cells, we first established stable clones overexpressing hPXR in LS174T. These cells were chosen because of their low, but detectable, PXR mRNA expression, and their well known sensitive ness toward both irinotecan and SN38. In addition these cells express a functional p53, which is partly involved in irinotecan response.
As shown in figure 2A, we selected three clones harbouring different PXR expression levels at both mRNA and protein levels. Basal expression levels of CYP3A4 mRNA were higher in PXR expressing cells compared to controls and were further increased by rifampicin. In agreement with the very low level of endogenous thorough PXR detected in these cells, we observed a very weak increase of CYP3A4 mRNA levels upon treatment of parent and pcDNA3 transfected cells with rifampicin.