Background
Colorectal cancer (CRC), a malignancy of the gastrointestinal tract, is the third most common tumor worldwide [
1]. CRC patient outcomes have significantly improved due to the use of novel treatment modalities and complex treatment strategies, including optimized surgery, adjuvant chemotherapy and novel targeted biological agents. At an early stage, CRC is potentially curable by surgery followed by chemotherapy [
2]. However, approximately 25% of CRC patients present with metastatic disease at diagnosis and are unlikely to undergo curative surgical resection, and many patients with metastatic disease will relapse after potentially curative resections [
3,
4]. For these patients, systemic chemotherapy is most often the treatment of choice for increasing survival and improving quality of life [
5,
6]. Thus, chemotherapy is one of the most important treatments for malignancies in humans [
7]. Previous studies have revealed that the response rate after initial chemotherapy treatment is approximately 50% for CRC patients [
8,
9]. However, only 10% of CRC patients respond to secondary treatments [
8]. These data suggest that there is a high degree of resistance to current chemotherapies.
The pregnane X receptor (PXR/NR1I2) is a ligand-activated transcription factor belonging to the nuclear hormone receptor (NR) superfamily [
10]. PXR is expressed in many normal human tissues, including ovary, small intestine, colon, liver, breast, heart, and prostate [
11]. Furthermore, increased PXR expression has been reported in various tumors, such as breast, ovarian, colon and endometrial cancers [
12‐
17]. Further studies have shown that PXR plays an important role in the development of resistance to certain chemotherapeutic cancer treatments [
18‐
22]. PXR activates many drug-metabolizing enzymes and drug transporters, resulting in altered drug clearance [
23]. Our previous study showed that the mRNA expression of MRP3 significantly correlates with PXR expression in CRCs, suggesting that MRP3 might be involved in PXR-mediated chemotherapeutic resistance [
16]. However, the specific role and precise molecular mechanism of PXR- and MRP3-mediated chemotherapeutic resistance remain unclear.
In the present study, we demonstrated that PXR significantly reduced the amount of oxaliplatin (L-OHP) in tumor cells, and prevented L-OHP-mediated inhibition of cellular proliferation and induction of apoptosis. Our nude mouse model showed that PXR reduced the efficacy of L-OHP treatment on tumor growth. Further analysis revealed a positive correlation between MRP3 and PXR expression (P = 0.0023), which was confirmed in two independent cohorts from The Cancer Genome Atlas (TCGA) database. We also detected increased MRP3 expression in PXR over-expressing cell lines. Moreover, luciferase reporter, ChIP and site-directed mutagenesis assays demonstrated that PXR could directly bind to the MRP3 promoter, and that the key binding site were located at -796 to -782bp on the MRP3 promoter. Further investigations showed that PXR expression was associated with poor prognosis and was an independent prognostic factor in CRC. Our data reveal that PXR can transcriptionally activate MRP3 expression by directly binding to its promoter, which increases drug efflux and results in CRC resistance to L-OHP or platinum drugs.
Methods
Patient samples
A total of 93 colorectal cancerous and 87 corresponding noncancerous tissues were obtained from the Southwest Hospital in Chongqing, China. This study was approved by the ethics committee of Southwest Hospital. Informed consent was obtained from all recruited patients.
Tissue microarray generation and immunohistochemical analysis
Noncancerous adjacent tissues were compared with normal tissue, stained with haematoxylin-eosin and reviewed by at least 2 pathologists. The tissue microarray (TMA) containing tissues from 93 tumors with 87 pairs of tumor and matched peritumoural tissues, was constructed (in collaboration with Shanghai Biochip Company Ltd, Shanghai, China) as previously described [
24].
Immunohistochemistry (IHC) was performed using PXR and MRP3 antibodies (both 1:200; both from Santa Cruz Biotechnology; sc-48403 and sc-5774). Tumor cell staining was considered positive when immunoreactivity was greater than or equal to 10%. Positive staining was divided into 5 categories, and staining intensity was graded according to 4 levels as previously described [
25]. The expression levels were defined by the sum of positive staining and intensity.
The construction and transfection of expression and siRNA vectors for PXR
Expression and siRNA vectors for PXR were constructed and transfected as previously described [
26]. The stably PXR-transfected cells were selected using G418 (Calbiochem, La Jolla, CA, USA). Cell clones were obtained using the cylinder method.
RT-PCR and western blotting (WB) analysis
The mRNA expression analysis was performed using RT-PCR. Series PCR assays with different cycle numbers were performed to determine the linear phase of amplification. Based on the pilot experiments, the appropriate cycles were chosen. The primers are listed in Additional file
1: Table S1.
Sixty micrograms of protein was resolved by 10–15% sodium dodecyl sulphate- polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride membranes (Millipore Corporation, Bedford, MA, USA). After blocking, the membranes were incubated with primary antibodies overnight at 4 °C. The proteins were detected by chemiluminescence (Pierce, Rockford, IL, USA) after incubation with secondary antibodies. The probed membranes were stripped and incubated with β-actin monoclonal antibody (1:1200; Sigma). The primary antibodies used were PXR mouse monoclonal antibody (1:1000, Santa Cruz Biotechnology, sc-48403) and MRP3 goat polyclonal antibody (1:800; Santa Cruz Biotechnology, sc-5774). The secondary antibodies used were horseradish peroxidase-conjugated anti-mouse and anti-goat antibodies (both 1:2000, Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA).
Oxaliplatin drug treatment
For L-OHP content measurement, tumor cells were treated for 2 h at a concentration of 20mmol/L L-OHP (Hengrui Medicine, Lianyungang, Jiangsu Province, China); For the functional experiments in vitro, tumor cells were exposed for 24 h to 20 mmol/L L-OHP. The culture medium was replaced on a daily. For the tumor formation in vivo, the nude mice were injected once every 3 days with 5 mg of L-OHP (dissolved in normal saline)/kg body weight. Controls were treated with the same volume of normal saline.
Measurement of L-OHP content in tumour cells by mass cytometry
The transfected, L-OHP-treated tumor cells were analyzed using a CyTOF1 mass cytometer (DVS Sciences, Richmond Hill, Ontario, Canada). The instrument settings and initial post-processing parameters were set as previously described [
27‐
29]. Cells were measured at approximately 600 cells per second. Noise reduction was activated, and cell extraction parameters were as follows: cell length range was set from 15- to- 65 pushes, and the lower convolution threshold was set to 15.
HCT116 and LOVO cells were plated in 12-well plates at 3 × 105 cells per well. After culturing for 24 h, the cells were transfected with PXR or vector plasmids. After 48h of transfection, the cells were collected, diluted 1:5, plated in 12-well plates and selected with 0.8mg/ml of G418 for 14 days to establish stable clones and for subsequent treatment with L-OHP. Surviving colonies were stained by using Giemsa’s azur eosin methylene blue solution (Merck, Darmstadt, Germany) and counted. These experiments were performed in triplicate.
HCT116 and LOVO cells were plated in 96-well plates at 3.5 × 103 cells per well, transfected with PXR or empty vector plasmids, and treated with L-OHP. For knockdown, SW480 cells were plated in 96-well plates at 3 × 103 cells per well, transfected with PXR siRNA or negative control plasmids, and treated with L-OHP. Cell proliferation was evaluated using the Cell Proliferation Reagent MTS (CellTiter 96® Aqueous One Solution Cell Proliferation Assay, Promega) on days 1, 2, and 3. The assay was carried out in triplicate in three independent experiments.
Flow cytometry assay
PXR or empty vector-transfected cells (3.5 × 105 cells per well) were harvested at 48 h post-transfection and L-OHP treatment, and fixed in 70% ethanol overnight at 4 °C. The cells were stained with propidium iodide (BD Pharmingen, San Jose, CA, USA). Fifty thousand cells were sorted by FACSCalibur System (BD Biosciences, Franklin Lakes, NJ, USA), and cell cycle profiles were analysed using the ModFit software (Verity Software House, Topsham, ME, USA). Apoptosis was also assessed by Annexin V-APC/7-amino-actinomycin D staining (KeyGEN, Nanjing, China). Briefly, cells were harvested, washed with phosphate-buffered saline, resuspended in 500 μl of binding buffer, mixed with 5 μl of Annexin V-APC and 5μl of 7-ADD, incubated for 5–15 min in the dark. Fifty thousand cells were analyzed using a FACSCalibur cytometer. The Annexin V-positive cells were considered apoptotic cells and analysed using the ModFit software. The assays were carried out in triplicate in three experiments.
In vivo tumourigenicity
To assess the effects of PXR expression and L-OHP treatment on tumour formation in vivo, the PXR- or empty vector-expression stable HCT116 cells were injected subcutaneously into the right flanks of 6-week-old male Balb/c nude mice that were subsequently subjected to L-OHP treatment. Tumor volume was calculated as 0.5236 L1 (L2)
2 [
30]. The developing tumors were observed during the following 5 weeks, after which the mice were sacrificed. All mouse experiments were approved by the Institutional Animal Care and Use Committee of Third Military Medical University, China.
Site-directed mutagenesis of binding site assay
The site-directed mutagenesis assays were performed as previous study (26). The binding sites in MRP3 promoter constructs were mutated or deleted using a Quik- Change Lightning Multi Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA, USA) according to the instructions, and the mutations or deletions were confirmed by direct DNA sequencing.
Luciferase reporter and chromatin-immunoprecipitation (ChIP) assay
Cells were seeded in 24-well plates at 3.5 × 104 cells per well. After an overnight incubation, the cells were transfected with a DNA mixture including pGL3-MRP3 promoter-luciferase, pIRES2-EGFP-PXR or vector, as well as the internal pRL-TK plasmids. Luciferase activities were measured at 36 h post-transfection using a dual-luciferase reporter kit (Promega). Each experiment was performed in triplicate and repeated three times.
Chromatin-immunoprecipitation (ChIP) analysis was performed using a ChIP Assay Kit (Cell Signaling Technology) according to the manufacturer’s protocols. The immunoprecipitated and input DNA were used as templates for RT-PCR analysis, and the primers are listed in Additional file
1: Table S1.
Statistical analyses
Statistical analyses were performed using SPSS 13.0 software (SPSS, Inc., Chicago, IL, USA). The results are expressed as the mean ± standard deviation (s.d.). The data were analysed by Student’s t and chi-square (2-sided) tests. Patient clinical and pathological characteristics were compared by Pearson χ
2 test. Overall survival (OS) was calculated according to Kaplan-Meier and Cox regression. The p values less than 0.05 were considered significant.
Discussion
In this study, we investigated the molecular mechanisms of PXR-mediated MRP3 over-expression, which we demonstrate for the first time contributes to chemotherapeutic resistance in CRC. Our data demonstrated that PXR increased the L-OHP efflux capacity, reducing drug concentrations in tumor cells. Functional analysis revealed that PXR blunted L-OHP-mediated inhibition of cellular proliferation and inducing of apoptosis. Our nude mouse experiments showed that PXR reduced the efficacy of L-OHP treatment on tumor growth. TAM analysis revealed increased expression of PXR and MRP3 in cancer tissues and a positive correlation between MRP3 and PXR expression (P = 0.0023), which was further confirmed in two independent cohorts from the TCGA database. To investigate the mechanism of PXR-induced L-OHP resistance, MRP3 expression was assessed in cell models that differentially expressed PXR. The results revealed increased MRP3 expression in PXR over-expressing cells. Luciferase reporter and ChIP assays indicated that PXR can bind to the promoter of MRP3 to activate transcription and furthermore, that the PXR-binding sites are located at between -796 to -782bp of the MRP3 promoter. Moreover, we found that PXR expression is an independent predictor for poor prognosis in CRC patients. These results suggest that PXR can transcriptionally activate MRP3 by directly binding to its promoter, increasing drug efflux capacity and resulting in resistance to platinum drugs.
Chemotherapy is the treatment of choice in patients with advanced or metastatic CRC. Chemotherapeutic resistance of tumor cells is crucial problem and a complicated process that involves multiple genes and steps. PXR is a member of the ligand-activated transcription factor superfamily, and its downstream target genes are involved in drug metabolism and transportation [
20,
31]. Increasing evidence suggests that high PXR expression is associated with decreased treatment efficacy and increased chemo-resistance against drug-based cancer treatments, including irinotecan, tamoxifen, paclitaxel, doxorubicin and vinblastine [
32‐
39]. In this study, PXR increased the drug efflux capacity and reduced the amount of drug in tumor cells, attenuating the efficacy of L-OHP. PXR expression significantly decreased L-OHP.-mediated tumour cell growth inhibition and inducing of apoptosis. In our nude mouse experiments, PXR decreased L-OHP-mediated suppression of tumor growth. These results demonstrate that PXR enhances the resistance of CRC cells to the chemotherapeutic agent L-OHP.
Previous studies have indicated that PXR regulates a number of genes associated with drug resistance, such as cytochrome P450, multidrug resistance 1 and multidrug resistance‑associated protein 2 [
20,
31,
40]. Mounting evidence has shown that PXR is a master regulator of chemotherapeutic drug resistance in cancer treatment. However, the specific mechanism underlying this resistance remains unclear. MRP3 (also called ABCC3) belongs to the ABCC subfamily, which consists of 13 members in mammals and has been divided into the following 3 classes: the multi-drug resistance proteins (MRPs), the sulfonylurea receptors and the cystic fibrosis transmembrane conductance regulator (CFTR/ABCC7). MRP3 has not been extensively studied and its function and regulation remain to be completely established. Moreover, MRP3 is known to transport various bile salts and clinical drugs [
41‐
44]. In this study, our results revealed a positive correlation between MRP3 and PXR expression (
P = 0.0023), and in three independent cohort, we found increased MRP3 expression in samples expressing high levels of exogenous and endogenous PXR. As PXR is a transcription factor, we used luciferase reporter assays to assess whether PXR binds to the MRP3 promoter to activate its transcription. Our results revealed that PXR can directly bind to the MRP3 promoter to activate its transcription. To determine the precise PXR-binding site within the MRP3 promoter, we constructed luciferase constructs containing different regions of the MRP3 promoter. Luciferase reporter and ChIP assays determined that the PXR-binding sites are located between -884 to -620bp of the MRP3 promoter. Further studies reveal that the key binding sites are “---------
AGGGA-” (-787 to -783bp) between -884 to -620bp of the MRP3 promoter. These data showed that PXR activates MRP3 transcription by directly binding to its promoter, which increases the drug efflux capacity and results in resistance to platinum drugs. Unfortunately, the crucial sites within the PXR to bind MRP3 promoter are still unknown, and further studies employing site-directed mutagenesis of putative binding sites in PXR are required to clarify this issue. In addition, we also tested other known PXR target genes, and found that the MRP5, MDR1 and OATP-A expression were slightly up-regulated in the PXR over-expression cells (Additional file
1: Figure S1). Although the other target genes also appeared increasing, we still think that PXR increases the drug efflux capacity at least in part by directly activating MRP3 transcription according to our data of function, mechanism and restoring experiments.
To further determine the importance of PXR in CRC, a correlation analysis between PXR expression and the OS of patients was performed. Kaplan-Meier and Cox-regression analysis revealed significantly reduced OS of patients with high PXR expression. Further studies demonstrate that PXR is an unfavorable and independent prognostic factor for male or clinical stage I + II CRC patients. These results indicate that PXR is a potential biomarker for predicting outcome in CRC patients.
Acknowledgements
The authors would like to thank all the patients involved in this study.