Irinotecan induces steroid and xenobiotic receptor (SXR) signaling to detoxification pathway in colon cancer cells

Abstract

Background: Resistance to chemotherapy remains one of the principle obstacles to the treatment of colon cancer. In order to identify the molecular mechanism of this resistance, we investigated the role of the steroid and xenobiotic receptor (SXR) in the induction of drug resistance. Indeed, this nuclear receptor plays an important role in response to xenobiotics through the upregulation of detoxification genes. Following drug treatments, SXR is activated and interacts with the retinoid X receptor (RXR) to induce expression of some genes involved in drug metabolism such as phase I enzyme (like CYP), phase II enzymes (like UGT) and transporters (e.g. MDR1).

Results: In this study, we have shown that endogenous SXR is activated in response to SN-38, the active metabolite of the anticancer drug irinotecan, in human colon cancer cell lines. We have found that endogenous SXR translocates into the nucleus and associates with RXR upon SN-38 treatment. Using ChIP, we have demonstrated that endogenous SXR, following its activation, binds to the native promoter of the CYP3A4 gene to induce its expression. RNA interference experiments confirmed SXR involvement in CYP3A4 overexpression and permitted us to identify CYP3A5 and MRP2 transporter as SXR target genes. As a consequence, cells overexpressing SXR were found to be less sensitive to irinotecan treatment.

Conclusions: Altogether, these results suggest that the SXR pathway is involved in colon cancer irinotecan resistance in colon cancer cell line via the upregulation of select detoxification genes.

Figure 1

SXR accumulates in the nucleus after SN-38 treatment. A. LS180 and HepG2 cells were treated with 10 μM rifampicin (rif) or 10 ng/ml SN-38 for the indicated times, then nuclear and cytoplasmic extracts were prepared and subjected to WB with the indicated antibodies. B. After for 4 h exposure to 10 ng/ml SN-38, immunofluorescence experiment were performed in LS180 cells, showing SXR (green) and nucleus (red). Yellow colour indicates colocalization. C. Analysis of the cell cycle was performed by flow cytometry in LS180 cells after SN-38 exposure. D. LS180 and HepG2 cells were treated with 1 μM CPT-11 then nuclear and cytoplasmic extracts were prepared and subjected to WB as above.

Figure 2

SXR interacts with RXR in cells in a SN-38-dependent manner. LS180 and HepG2 cells were treated with 10 μM rifampicin (RIF), 10 ng/ml SN-38 (SN) or 1 μg/ml CPT-11 (CPT) for 4 h. Cell extracts were immunoprecipitated (IP) with either IgG (IP control) or anti-SXR (IP SXR) in the presence of vehicle (ctl), rifampicin, SN-38 or CPT-11. Association of the endogenous RXR with the anti-SXR precipitate was detected by WB using anti-RXR antibody. Input indicates endogenous RXR present in 5% of total cell lysates used in each IP. To ensure equal amounts were precipitated, WB using anti-SXR was also performed.

Figure 3

SN-38 treatment induces binding of endogenous SXR to the native CYP3A4 promoter. A. ChIP assays were performed and DNA was further analyzed either by classical PCR or by semi-quantitative PCR using a primer set specific for the promoter or control region as indicated by the arrows on the schematic diagram of the CYP3A4 promoter. B. LS180 were treated with 10 μM of rifampicin (rif), 1 μg/ml CPT-11 or 10 ng/ml SN-38 for 4 h and subjected to formaldehyde cross-linking. Soluble chromatin was prepared by sonication. Precleared chromatin solution was immunoprecipitated by antibodies anti-SXR or Ig control, and precipitated. PCR was performed with the precipitated DNA (IP SXR) or the DNA present in 10% of total cell lysates used for each IP (input). C. Cells were treated with SN-38 for 4 h and subjected to ChIP as described above with anti-SXR, anti-RNA phospho-ARN polymerase II and anti-RNA polymerase II. Enrichment factors were determined by qPCR using GAPDH as threshold indicator (internal control for each IP).

Figure 4

Induction of irinotecan metabolism genes after CPT-11 and SN-38 treatment. LS180 and HepG2 cells were treated with 1 μg/ml CPT-11 or 10 ng/ml SN-38 for 24 h and quantitative RT-PCR were performed on irinotecan metabolism genes. Fold inductions were calculated relative to the mean expression of GAPDH and experiments were performed in triplicate. (^a ND for not detected).

Figure 5

SXR involvement in induction of irinotecan metabolism genes. A. Detection of SXR by western blot in LS180 and HepG2 cells after 48 h siRNA transfection. B. After 48 h siRNA transfection, cells were cultivated for 24 h in new medium and mRNA levels of the irinotecan metabolism genes were determined. Data represent percentage of mRNA levels after siRNA-SXR transfection compared to siRNA-control transfection. C. After 48 h siRNA transfection, cells were stimulated with 10 ng/ml SN-38 for 24 h and mRNA levels were determined. Data represent fold induction of mRNA levels compared to untreated cells.

Figure 6

Effect of SXR overexpression on LS180 and HCT116 survival after SN-38 treatment. A. Detection of carboxylesterase 2 (CE2) by western blot in LS180, HepG2 and HCT116 cells. B. and C. LS180 cells (B) and HCT116 cells (C) were transfected with control or SXR vector for 24 h, then treated for 72 h with different CPT-11 concentrations. The SXR overexpression was confirmed by western blot after 24 h transfection. Cell viability following irinotecan exposure was determined by SRB assay by comparison with untreated cells. Each point represents the mean of 3 replicates and the experiment has been repeated 3 times. All results are expressed as the mean ± SE and obtained data were analysed for statistical differences by Student's t test. A p value of less than 0.05 was considered statistically significant (star indicates p < 0.05).

Figure 7

Proposed model for the SXR-mediated response to irinotecan in colon cancer cells. In LS180 cells, CPT-11 is metabolized to SN-38 by the carboxylesterases (CE). SN-38 induces the activation of SXR. Consequently, the transcription factor is recruited to the CYP3A4 and CYP3A5 promoter and upregulates the expression of these genes. SXR also constitutively activates MDR1 and MRP1 expression. As a feedback loop, CYP3A4, CYP3A5 and the transporters reduce the concentration of active irinotecan inside the cell, which favors cell proliferation. As a consequence, we propose that the combined detection of SXR together with a high expression of CYP3A4 should help to define in advance the subsets of tumors that will fail to respond to chemotherapy based on irinotecan treatment.