Oxaliplatin is a third-generation platinum agent used in colorectal cancer treatment. Oxaliplatin resistance acquisition is a complex process mainly based on alteration of genes and pathways involved in its mechanism of action. Therefore, our purpose was to perform a gene expression screening in an in vitro model to identify genes that could play a role in oxaliplatin resistance acquisition processes. Four colorectal cancer cell lines and their oxaliplatin-resistant derived sublines were compared. Microarray analysis was done using Human 19K Oligo Array Slides. RNA from cells were hybridized with a commercial RNA reference sample and labeled with both fluorochromes Cy3 and Cy5. Data were analyzed by hierarchical clustering method. Subsequently, quantitative real-time PCR (qRT-PCR) was used to corroborate microarray data, considering as positively validated those genes that showed significant differences in expression levels between groups and a correlation between microarray and qRT-PCR data. By microarray analysis, 32 candidate genes were identified. After validation process by qRT-PCR, the genes AKT1, CDK5, TRIP, GARP, RGS11, and UGCGL1 were positively validated. The 3 first genes proved to be involved in regulation of nuclear factor-κβ antiapoptotic transcription factor previously related to drug resistance, and the other 3 genes are novel finds. We have identified 6 genes related to oxaliplatin resistance acquisition. These findings are of paramount importance to understand these processes better and open new lines of study to elucidate the relevance of this pharmacogenomic approach into the clinic. [Mol Cancer Ther 2009;8(1):194–202]

Colorectal cancer is the second leading cause of cancer related mortality in the western world. Approximately a quarter of patients have incurable cancer at diagnosis and half of the patients who undergo potentially curative surgery will ultimately develop metastatic disease (1). The standard advanced colorectal cancer treatment is based on the administration of fluoropyrimidines (5-fluorouracil or capecitabine) combined with oxaliplatin, the topoisomerase I inhibitor CPT-11, and the monoclonal antibodies cetuximab, bevacizumab, or panitunumab. These newer combinations have significantly improved response rates (>50%) and prolonged overall survival (2). However, drug resistance remains a major clinical challenge for cancer treatment. Although patients suffering from advanced colorectal cancer are in most cases initially responsive to the combined chemotherapy treatment, they later experience disease relapse due to eventual tumor recurrence and emergence of drug-resistant tumor cells.

Oxaliplatin (1,2-diaminocyclohexane-oxalate platinum) is a member of the family of platinum-containing chemotherapeutic agents that also include cisplatin and carboplatin (3). Oxaliplatin is distinguished from these two other drugs by its different spectrum of activity both in preclinical models (4) and in clinical setting (5). It is the only platinum analogue that has activity in colon cancer, a disease for which this drug has now become a mainstay of therapy (6, 7). It mainly forms intrastrand adducts between two adjacent guanine residues or guanine and adenine disrupting DNA replication and transcription (3, 8). Oxaliplatin has been reported to be involved in p38 kinase activation, phosphatidylinositol 3-kinase/AKT pathway, and caspase cascade activation mainly through the apoptotic intrinsic pathway (911). Damage induced by oxaliplatin has been reported to be repaired by the nucleotide excision repair pathway (12). However, the downstream molecular events underlying the cytotoxic effects of this chemotherapeutic agent have not been well characterized.

The mechanisms involved in oxaliplatin resistance are still poorly understood. It is a multifactor process that has been related to decreased drug accumulation, increased detoxification and repair, enhanced tolerance to damage, alteration in pathways involved in cell cycle kinetics, and apoptosis inactivation (3, 13). Platinum-resistant cells are characterized by reduced cellular drug accumulation and decreased DNA adducts formation. Increased levels of intracellular glutathione, abnormalities in apoptotic pathways, and alterations in DNA repair genes, such as ERCC1 and XRCC1, may also play a role (1416). Levels of copper transporters ATP7B, ATP7A, and hCTR1 have been also related to platinum drug resistance (17, 18). However, most of the previous studies have been focused on resistance to cisplatin (3).

In the last decade, gene expression profiling of human cancer has proved valuable in cancer research, providing valuable insight into mechanisms and targets involved in oncogenesis in several neoplasms (19, 20). DNA microarray technology allows the simultaneous assessment of thousands of genes and this approach provides valuable means to identify novel molecular targets for therapeutic intervention.

To clarify the multifactor process of resistance mechanisms and to determine genes involved in several pathways potentially related to oxaliplatin resistance acquisition, the aim of this work is to establish an oxaliplatin acquired resistance in vitro model and perform a gene expression screening subsequently validated by quantitative real-time PCR (qRT-PCR). These genes could potentially be used as oxaliplatin predictive markers in colorectal cancer patients in future in vivo studies.

Cell Lines and Resistance Acquisition Processes

Human tumor-derived colorectal adenocarcinoma HT29, LoVo, DLD1, and LS513 cell lines (American Type Culture Collection) were used as the parental cells to obtain oxaliplatin-resistant sublines. Derivative cell lines HTOXAR3, LoVOXAR3, DLDOXAR3, and LSOXAR3 were established by weekly 24 h exposure to increase concentrations of oxaliplatin, starting on drug inhibitory concentration 50 (IC50) during a 10-month period (18). Cell growth was assessed after 7 days, with daily medium changes, by using the methylene blue technique comparing the sensitive parental and its corresponding oxaliplatin-resistant derivative cell line. Oxaliplatin dose increase was not carried out until oxaliplatin-resistant cells reached growing rates near 70% compared with their corresponding parental cells.

Cell viability testing was done periodically by MTT test (Roche Diagnostics). The resistance degree was proven to be stable after 4 weeks without oxaliplatin exposure. Cell lines were grown as monolayer in RPMI 1640 (Invitrogen) medium (DLD1, LS513, and their resistant sublines) and DMEM (Invitrogen) medium (HT29 and HTOXAR3) supplemented with 10% of heat-inactivated FCS (Biological Industries), 400 units/mL penicillin, 40 μg/mL gentamicin, 2 mmol/L l-glutamine (Sigma), and 10 mmol/L HEPES (Sigma). LoVo and LoVOXAR3 were grown in Ham's F-12 (Invitrogen) with 20% FBS, 200 units/mL penicillin, and 20 μg/mL gentamicin. All cell lines were cultured at 37°C in a humidified atmosphere of 5% CO2. Cells were routinely checked for Mycoplasma contamination by using a PCR-based assay (Stratagene).

Cell Viability Assay

Oxaliplatin IC50 determination and the subsequent analysis of resistance degree and cross resistance to other drugs were assessed by using MTT colorimetric assay. Cancer cells were seeded at a density of 2,000 per well in 96-well microtiter plates and allowed to attach. Cells were treated with a drug delivered for 24 h and fresh medium was added to each well. Cells were allowed to grow for 72 h. Subsequently, MTT was added for 4 h and absorbance was read at 590 nm the following day. Doses for each fraction of survival (ranging from 10% to 90% of cell viability) were determined in each cell line by using the median effect line method described in ref. 21. The data reported in Table 1 represent the oxaliplatin IC50 mean ± SD of a minimum of three independent experiments with 8 wells at each drug concentration tested.

Table 1.

Cytotoxicity and resistance patterns for oxaliplatin, cisplatin, and SN38

Cell linesOxaliplatin (μmol/L)
Cisplatin (μmol/L)
SN38 (nmol/L)
IC50* ± SDRI ± SDIC50* ± SDRI ± SDIC50* ± SDRI ± SD
HT29 6.46 ± 0.13 4.61 ± 0.27 4.70 ± 0.60 0.56 ± 0.04 9.68 ± 0.07 1.06 ± 0.04 
HTOXAR3 29.86 ± 2.33  2.70 ± 0.23  10.31 ± 0.52  
DLD1 5.05 ± 0.22 3.72 ± 0.26 2.87 ± 0.09 1.14 ± 0.06 6.62 ± 0.26 1.23 ± 0.20 
DLDOXAR3 18.79 ± 0.49  3.27 ± 0.06  8.15 ± 1.67  
LoVo 0.25 ± 0.09 9.40 ± 0.31 2.01 ± 0.09 1.37 ± 0.03 4.17 ± 0.21 1.22 ± 0.13 
LoVOXAR3 2.45 ± 0.27  2.78 ± 0.18  5.09 ± 0.82  
LS513 0.46 ± 0.03 3.64 ± 0.42 1.46 ± 0.36 0.57 ± 0.04 5.87 ± 0.16 0.58 ± 0.04 
LSOXAR3 1.69 ± 0.10  0.82 ± 0.14  3.19 ± 0.46  
Cell linesOxaliplatin (μmol/L)
Cisplatin (μmol/L)
SN38 (nmol/L)
IC50* ± SDRI ± SDIC50* ± SDRI ± SDIC50* ± SDRI ± SD
HT29 6.46 ± 0.13 4.61 ± 0.27 4.70 ± 0.60 0.56 ± 0.04 9.68 ± 0.07 1.06 ± 0.04 
HTOXAR3 29.86 ± 2.33  2.70 ± 0.23  10.31 ± 0.52  
DLD1 5.05 ± 0.22 3.72 ± 0.26 2.87 ± 0.09 1.14 ± 0.06 6.62 ± 0.26 1.23 ± 0.20 
DLDOXAR3 18.79 ± 0.49  3.27 ± 0.06  8.15 ± 1.67  
LoVo 0.25 ± 0.09 9.40 ± 0.31 2.01 ± 0.09 1.37 ± 0.03 4.17 ± 0.21 1.22 ± 0.13 
LoVOXAR3 2.45 ± 0.27  2.78 ± 0.18  5.09 ± 0.82  
LS513 0.46 ± 0.03 3.64 ± 0.42 1.46 ± 0.36 0.57 ± 0.04 5.87 ± 0.16 0.58 ± 0.04 
LSOXAR3 1.69 ± 0.10  0.82 ± 0.14  3.19 ± 0.46  
*

IC50, drug concentration necessary to obtain a mortality of 50%. (n = 3).

SD (n = 3).

Resistance index, calculated as the ratio between IC50 of resistant sublines and its corresponding sensitive cell lines.

Drugs

Oxaliplatin, cisplatin, and SN38 were prepared in water at a concentration of 10 mmol/L stock solution, in physiologic serum at a concentration of 3 mg/mL, and in DMSO at a concentration of 10 mmol/L stock solution, respectively, and stored at −20°C. Further dilutions of each drug were made in culture medium to final concentrations before use.

cDNA Microarray Hybridization

For determination of basal levels, resistant cells were maintained in oxaliplatin-free medium for 1 month before the evaluation of the gene expression. Total RNA from cells was isolated using RNeasy mini kit (Qiagen) according to the instructions of the manufacturer. RNA quantification was done using a NanoDrop spectrophotometer (Nucliber). Total RNA (5 μg) for each cell line was used to synthesize double-stranded cDNA using Superscript II reverse transcriptase (Invitrogen) and an oligo(dT) primer (Invitrogen). Gene expression analysis was carried out using the Human 19K Oligo Array slides (Center of Applied Genomics, University of Medicine of New Jersey).4

Probes came from the Compugene collection and were spotted with the GeneMachines OmniriGrid Microarrayers from Genomic Solution in poly-l-lysine slides.

Resistant and sensitive cell lines were hybridized with a commercial RNA (Human Reference RNA; Stratagene). This approach was used to perform an analysis consisting of comparing two groups of cell lines (each cell line was considered as biological replicate in each group) and to obtain oxaliplatin resistance-related common genes (22, 23).

Each sample was labeled with both fluorochromes Cy3/Alexa Fluor 546 and Cy5/Alexa Fluor 647 (Dye Swap) in independent experiments to get a technical replica and a hybridization control for each sample. Labeling and hybridization processes were made by using Genisphere 3DNA Array-350 kit according to manufacturer's protocol (Genisphere). The data reported here have been deposited in the National Center for Biotechnology Information Gene Expression Omnibus5

and is available through Gene Expression Omnibus Series availability no. GSE10405.

Imaging and Data Analysis

All slides were scanned with an Affymetrix Genetic Microsystems GMS418 Array scanner using the appropriate gains on the photomultiplier tube to obtain the highest intensity without saturation. A 16-bit TIFF image was generated for each channel, Cy3 and Cy5. The image analysis was done with ImaGene 4.1 software (Biodiscovery) to quantify the signal intensities. Data from spots not recognized by the ImaGene analysis software were excluded from further considerations (empty, poor, and negative spots). The data sets were imported into Microsoft Excel to analyze the quality of each feature. Normalization data were done with ArrayNorm 1.7.2 software by using Lowess and Dye Swap Sim Fix Filter methods. Visualization of results was carried out with the normalized data through average linkage and Euclidean distance as a measurement of similarity using Genesis 1.5.0 Software (Alexander Sturn, Institute for Genomics and Bioinformatics, Graz University of Technology). ANOVA was used to determine gene expression differences between resistant and sensitive groups. Every cell line was considered as a biological replicate in each group and we calculated ANOVA by using the mean of these replicates for each of the groups. Results obtained from microarray analysis were validated by using qRT-PCR. Because this is a very specific and precise methodology, we used (instead of procedures such false discovery rate) another high restrictive criteria (two-sided P values < 0.01) to assess the statistical significant differences between two groups and determine the set of genes that would be subsequently validated.

Identification of Genes of Interest

Identification of genes that showed expression changes between resistant and sensitive cells groups was made by using the Web searchers,6

only considering well-annotated transcripts (not partial cds for hypothetical proteins, hypothetical insert cDNA clones, etc.).

BLAST hybridization7

between identified genes sequence and corresponding spot oligonucleotide sequence was done to avoid false-positive hybridizations. TaqMan Gene Expression Assays (Applied Biosystems) for each selected gene were searched by using Applied Biosystems Web site searching tools.8

Gene Expression Analysis by qRT-PCR

Expression levels of genes that significantly correlated with oxaliplatin resistance were selected for further confirmation using qRT-PCR. In addition to significant correlation with oxaliplatin resistance, genes selected were those which showed the greatest expression range across the panel of 8 cell lines and the greatest expression differences between resistant and sensitive cell groups.

Reverse transcription was carried out with 3 μg RNA and random hexadeoxynucleotide primers (Invitrogen) in 20 μL solution containing Moloney murine leukemia virus retrotranscriptase enzyme (Invitrogen). Template cDNA was added to TaqMan Universal Master Mix (Applied Biosystems) in a 12.5 μL reaction with specific primers and probe for each gene by using TaqMan Gene Expression Assays (Applied Biosystems) as follows: CDP (Hs00154566_m1), RGS11 (Hs00358686_m1), GUCY1B2 (Hs00188041_m1), GOLGB1 (Hs00189566_m1), CDK5 (Hs00358991_g1), AKT1 (Hs00178289_m1), GARP (Hs00194136_m1), TRIP (HS00183394_m1), SOX5 (Hs00753050_s1), RAB4B (Hs00535053_m1), SEMA4F (Hs00188642_m1), and UGCGL1 (Hs00219941_m1). Analysis of qRT-PCR was carried out using the ABI Prism 7900HT Sequence Detection System (Applied Biosystems). Three replicate wells per gene per cell line were included in each plate and a minimum of three independent experiments were done.

Relative gene expression quantification was calculated according to the comparative threshold cycle method (2-ΔΔCt; ref. 24) using β-actin and 18s rRNA as endogenous control genes, because it has been reported that the use of two housekeeping genes reduces variability data among samples. It is considered that a pair of housekeeping genes are good endogenous controls if the ΔCt between both is similar in all samples analyzed (25). Human reference RNA (Stratagene) was used as calibrator sample (exogenous control). Normalized expression values were determined as follows: 2-(ΔCt sample - ΔCt calibrator), where ΔCt values were calculated by subtracting the Ct value (the PCR cycle at which the DNA amount reaches the detection threshold) of the target gene from the value of the geometric expression mean between endogenous control genes as described previously (25). Fold change between resistant derivative and its parent sensitive cell line was calculated for each gene.

SPSS 12.0 for Windows software was used to perform qRT-PCR statistical analysis. Correlation between qRT-PCR results and microarray data was calculated by using Spearman ρ test, considering a coefficient of correlation (R) over 0.5 and a significant P < 0.05 as positive validated data. Changes in gene expression comparing resistant cell lines group versus sensitive group by using qRT-PCR were calculated with Wilcoxon signed-rank test, considering as positively validated those genes with significant expression changes. Genes analyzed by qRT-PCR that were positively validated according to at least one of the two statistical tests were considered as acquired oxaliplatin resistance-related genes. P ≤ 0.05 was considered significant.

Oxaliplatin-Resistant Cell Lines Isolation

The isolation of HTOXAR3, LoVOXAR3, DLDOXAR3, and LSOXAR3 was started with an initial continuous exposure of their sensitive colorectal cancer parental cells HT29, LoVo, DLD1, and LS513 to their respective oxaliplatin IC50 as described previously in ref. 18. In the following passages, the cell number increased in the presence of the drug until the cells showed a stable growth curve and a survival rate near 70%. A similar sequence of events continued with the next increasing drug concentrations. After 10 months under oxaliplatin continuous exposure, resistant sublines acquired levels of resistance between 3- and 10-fold versus their sensitive parental cells (Table 1). Moreover, oxaliplatin-resistant sublines were stable as they kept their resistance degree after 1 month in drug-free medium. To check for possible cross-resistance with other agents, sensitive and resistant cells were treated with cisplatin or SN38 (irinotecan active metabolite) for 24 h. As shown in Table 1, oxaliplatin-resistant cells did not show cross-resistance with none of these drugs. Thereby, these cells were used at basal conditions as an oxaliplatin resistance in vitro model in the present study.

cDNA Microarray Analysis

To analyze the basal levels of oxaliplatin resistance-related gene expression, cell lines were classified into two groups: sensitive parental cells (S group) and resistant derived sublines (R group). In both groups, two technical replicas of each cell line were made for each hybridization experiment (Dye Swap). After quantification, filtration, and normalization analyses, 14,244 genes from the 19,200 contained in the slide were analyzed by Genesis 1.5.0 software. Gene expression differences were determined by using ANOVA function and 49 genes showed significant changes (P < 0.01) between both groups. According to hierarchical clustering analysis, this set of genes were classified in four groups or clusters that showed expression changes between R and S groups of cells, leading to a gene expression profile related to our in vitro oxaliplatin resistance acquisition model (Fig. 1).

Figure 1.

Gene expression profile obtained from clustering analysis. This set of genes showed significant expression changes between resistant (HT29R3, DLD1R3, LOVOR3, and LS513R3) and sensitive (DLD1S, HT29S, LoVoS, and LS513S) cell groups (ANOVA function; P < 0.01; 8 expression values = no missing values).

Figure 1.

Gene expression profile obtained from clustering analysis. This set of genes showed significant expression changes between resistant (HT29R3, DLD1R3, LOVOR3, and LS513R3) and sensitive (DLD1S, HT29S, LoVoS, and LS513S) cell groups (ANOVA function; P < 0.01; 8 expression values = no missing values).

Close modal

Identification and Selection of Genes from Microarray Analysis

Identification of genes process was done by using several gene Web searchers. Thirty-two of the 49 genes obtained in the clustering analysis were selected, whereas 17 genes that could not be identified were ruled out (Fig. 1). From these, 12 were down-regulated and 14 up-regulated in R versus S group (Fig. 2). Most of the oxaliplatin resistance-related genes could be classified according to gene ontology (Fig. 3) and were mainly involved in physiologic (55.8%) and cellular integrity (41.9%) processes. The sequences corresponding to the 32 genes identified were localized in the slides by hybridization experiments. To avoid false-positive gene identifications, BLAST hybridization between these sequences and sequences corresponding to typified genes was done. As a result, 26 genes were validated and 6 sequences were recognized as false-positive identifications; consequently, they were ruled out.

Figure 2.

List of 26 genes selected after gene identification and validation through BLAST hybridization. Twelve of these genes were up-regulated and 14 were down-regulated in oxaliplatin-resistant versus parental sensitive group of cells.

Figure 2.

List of 26 genes selected after gene identification and validation through BLAST hybridization. Twelve of these genes were up-regulated and 14 were down-regulated in oxaliplatin-resistant versus parental sensitive group of cells.

Close modal
Figure 3.

Classification of gene sets obtained from microarray analysis according to gene ontology by using the Web tool http://apps1.niaid.nih.gov/david/. Each gene was grouped in one or more categories defined in the figure. Most of these oxaliplatin resistance genes are related to cell integrity (41.9%) and physiologic processes (55.8%).

Figure 3.

Classification of gene sets obtained from microarray analysis according to gene ontology by using the Web tool http://apps1.niaid.nih.gov/david/. Each gene was grouped in one or more categories defined in the figure. Most of these oxaliplatin resistance genes are related to cell integrity (41.9%) and physiologic processes (55.8%).

Close modal

Validation of Selected Genes by qRT-PCR

A panel of 12 genes (AKT1, CDK5, CPD, GARP, GOLGB1, GUCY1B2, RAB4B, RGS11, SEMA4F, SOX5, TRIP, and UGCGL1) was selected as they showed the greatest expression range across cell lines and the greatest expression differences between both groups (see Materials and Methods). Basal levels of their corresponding mRNAs were evaluated in the eight cell lines and the calibrator sample to corroborate microarray data. The expression of 11 of 12 genes could be determined. Low expression values for SOX5 made its analysis impossible by qRT-PCR; thus, it was ruled out.

As shown in Table 2, 6 of 11 genes were positively validated after qRT-PCR analysis (see Materials and Methods). According to Spearman ρ test, 5 genes (AKT1, CDK5, RGS11, TRIP, and UGCGL1) showed good correlation between qRT-PCR and microarray data (R > 0.5; P < 0.05). GARP also showed positive correlation (R > 0.5), but its associated P value was not significant. According to Wilcoxon signed-rank test, 4 genes (AKT1, CDK5, RGS11, and GARP) showed significant differences between R and S groups. TRIP and UGCGL1 genes also showed expression changes, but these did not reach a statistical significance. Although CPD, GOLGB1, GUCY1B2, RAB4B, and SEMA4F had been identified as oxaliplatin resistance-related genes by microarray analysis, these results could not be positively validated according to qRT-PCR data.

Table 2.

qRT-PCR validation analysis

GenesWilcoxon signed-rank test
Spearman ρ correlation
PR*P
AKT1 0.01 0.83 0.04 
CDK5 0.02 0.89 0.007 
CPD 0.43 −0.33 0.42 
GARP 0.01 0.68 0.14 
GOLGB1 0.76 −0.38 0.35 
GUCY1B2 0.23 0.44 0.27 
RAB4B 0.10 −0.47 0.24 
RGS11 0.006 0.73 0.04 
SEMA4F 0.23 0.29 0.47 
SOX5 No data No data No data 
TRIP* 0.09 0.89 0.02 
UGCGL1 0.23 0.89 0.02 
GenesWilcoxon signed-rank test
Spearman ρ correlation
PR*P
AKT1 0.01 0.83 0.04 
CDK5 0.02 0.89 0.007 
CPD 0.43 −0.33 0.42 
GARP 0.01 0.68 0.14 
GOLGB1 0.76 −0.38 0.35 
GUCY1B2 0.23 0.44 0.27 
RAB4B 0.10 −0.47 0.24 
RGS11 0.006 0.73 0.04 
SEMA4F 0.23 0.29 0.47 
SOX5 No data No data No data 
TRIP* 0.09 0.89 0.02 
UGCGL1 0.23 0.89 0.02 
*

Coefficient of correlation of Spearman ρ test.

Positive validated genes by qRT-PCR. They were positively validated by at least one of the two statistical tests.

Due to methodologic problems, this gene could not be analyzed by qRT-PCR.

To summarize results, down-regulation of GARP, RGS11, and TRIP and up-regulation of AKT1, CDK5, and UGCGL1 were proven to be related to oxaliplatin resistance acquisition in our in vitro model.

Resistance acquisition to chemotherapeutic agents is one of the main problems that come up during the treatment of all tumor types. The present work is focused on the study of oxaliplatin resistance in colorectal tumors. Moreover, mechanisms of oxaliplatin resistance are poorly understood, as most studies about resistance to platinum agents have been focused on cisplatin (3). We used four colorectal cancer cell lines and we were able to induce resistance to oxaliplatin, a drug widely used in the standard treatment of this tumor. The resulting oxaliplatin-resistant cells were not subcloned, similarly to most of the published works based on microarray analysis by using in vitro models (10, 13, 26). The four colorectal cancer cell lines, which showed genotypic differences such as p53 status and RER phenotype, were analyzed jointly. Thereby, our approach allows emphasizing the involvement of selected genes in general mechanisms of oxaliplatin resistance acquisition and avoids processes due to individual characteristics of a particular cell line. It is important to point out that this is one of the few studies on drug resistance that include more than one resistant cell line (26) and all of them are of the same tumor type (13).

Our study was based on the determination of oxaliplatin resistance-related genes detected by qRT-PCR starting at the results obtained from a previous genetic screening by using microarray technology. The set of genes obtained in this analysis was not considered when their expression was not validated by qRT-PCR technique. DNA microarrays provide an unprecedented capacity for whole genome profiling. However, the quality of gene expression data obtained from microarrays can vary greatly in platform and procedures used. Quantitative RT-PCR is a commonly used validation tool for confirming gene expression results obtained from microarray analysis and is often referred to as the “gold standard” for gene expression measurements due to its advantages in detection sensitivity, sequence specificity, large dynamic range, as well as its high precision and reproducible quantification compared with other techniques (26, 27).

To begin characterizing our data functionally, 6 target genes were selected after validation analysis in our model: AKT1, CDK5, RGS11, GARP, TRIP, and UGCGL1.

Serine/threonine kinase AKT1 (v-AKT murine thymoma viral oncogene homologue 1) plays a critical role in cell survival. It is activated by phosphatidylinositol 3-kinase and transmits survival signals through phosphorylation-dependent suppression of intracellular proapoptotic factors (28). It is well known its implication in the activation of antiapoptotic mechanisms such as nuclear factor-κβ (NF-κβ) pathway (28). Moreover, it plays an important role in early events of the colorectal carcinogenesis interacting with Wnt/β-catenin pathway (29). In fact, the AKT1 signaling pathway is up-regulated in cancers, critical for tumorigenesis and chemoresistance, and therefore a critical target for cancer intervention and for overcoming drug resistance (30, 31). The association between AKT1 activity and resistance to cisplatin has been reported in ovarian, lung, and breast cancer cell lines (32). In lung cancer cisplatin-resistant cells, an increase in AKT1 mRNA levels was found compared with their sensitive parental cells (33). Interestingly, our data corroborate these results in our oxaliplatin resistance in vitro model; indeed, it is important taking into account the low number of published studies about this issue.

CDK5 is a unique member of a small serine/threonine cyclin-dependent kinase family. Different from the other cyclin-dependent kinases, which are involved in cell cycle regulation, CDK5 is essential in neuronal development and is mainly expressed in the nervous system. Nevertheless, the presence of CDK5 in several tumor types has been reported (34, 35). Kim et al. found higher expression levels of this gene in HT29 cells and colon tumor tissues in comparison with colon normal mucosa and suggested that CDK5 expression could play a role in colon cancer cell proliferation (34). Interestingly, they suggested an antiapoptotic effect of this gene by indirectly inhibiting the peroxisome proliferator-activated receptor γ (PPARγ). It seems to be related to the inhibitory action of CDK5 over MEK1 (36), protein involved in PPARγ activation (37). Moreover, the PPARγ proapoptotic activity has been shown, as it reduces proliferation of colorectal tumors in vitro and in vivo (38). Besides, Eichele et al. studied the effect of PPARγ inhibition in chemoresistance (39). They related PPARγ down-regulation with resistance to cisplatin showing the involvement of this receptor in apoptosis elicited by this drug. Moreover, the inhibitory effect of PPARγ activation has been described over NF-κB transcription factor (40), previously related to platinum resistance (9, 41). Thereby, our results would agree with these approaches because an up-regulation of CDK5 expression would be translated into a decrease in PPARγ activation and consequently in a drug resistance acquisition.

TRIP is a component of the tumor necrosis factor receptor superfamily/TRAF signaling complex. Although its function in vivo is still unclear, some studies in vitro have shown the involvement of TRIP in the inhibition of the TRAF2-mediated NF-κβ activation (4244). It is known that NF-κβ is responsible for the expression induction of several antiapoptotic genes (45) and its inhibition has been related to oxaliplatin activity enhancement (9). Indeed, our results support these data because lower levels of TRIP would promote NF-κβ activity and consequently resistance to the drug.

The other three genes characterized in this work are UGCGL1, GARP, and RGS11. The first gene is involved in metabolism of proteins (46), the second is involved in a variety of functions such as cell-cell interaction (47), and the third one belongs to a family of proteins that act as regulators of G-protein activity (48). Besides the presence of DNA amplifications at GARP locus (11q13.5-14) in breast cancer (49), to our knowledge, the possible involvement of these genes in metabolism or resistance to chemotherapeutic agents has not been described. Actually, the little information existing about these genes set them as new oxaliplatin-related targets of study.

Recently, Cui et al. (50) constructed a cancer signaling map. Interestingly, several of the genes we have validated were mapped near a Ras subnetwork and were directly or indirectly related to p38 kinase, a protein described previously to be involved in oxaliplatin-related apoptotic mechanisms (9).

It is worth highlighting the involvement of 3 of these genes (AKT1, CDK5 and TRIP) in NF-κβ pathway regulation. Taken together, we could postulate a model focused on NF-κB activity in which, in concordance with our results, low levels of TRIP and high levels of AKT1 and CDK5 would contribute to NF-κβ activation and consequently cell antiapoptotic activity and oxaliplatin acquired resistance (Fig. 4). According to this, NF-κβ signaling pathway could be considered a good starting point to following studies focused on mechanisms of oxaliplatin resistance acquisition.

Figure 4.

AKT1, CDK5, and TRIP involvement in the regulation of NF-κB. Antiapoptotic activity of this transcription factor was induced by the activation of the first 2 genes and the inhibition of the third. Gray, target genes; dotted arrows, expression status of these genes in oxaliplatin-resistant cells versus sensitive cells.

Figure 4.

AKT1, CDK5, and TRIP involvement in the regulation of NF-κB. Antiapoptotic activity of this transcription factor was induced by the activation of the first 2 genes and the inhibition of the third. Gray, target genes; dotted arrows, expression status of these genes in oxaliplatin-resistant cells versus sensitive cells.

Close modal

Summarizing the results, we identified a set of genes whose expression is related to oxaliplatin resistance acquisition by using an in vitro cellular model. This information may be useful as a starting point to future in vivo studies to validate their role as oxaliplatin resistance markers in colorectal cancer patients.

No potential conflicts of interest were disclosed.

Grant support: Fondo de Investigación Sanitaria, Instituto de Salud Carlos III (Spain), project PI031068.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1
Weitz J, Koch M, Debus J, Hohler T, Galle PR, Buchler MW. Colorectal cancer.
Lancet
2005
;
365
:
153
–65.
2
Pfeiffer P, Qvortrup C, Eriksen JG. Current role of antibody therapy in patients with metastatic colorectal cancer.
Oncogene
2007
;
26
:
3661
–78.
3
Rabik CA, Dolan ME. Molecular mechanisms of resistance and toxicity associated with platinating agents.
Cancer Treat Rev
2007
;
33
:
9
–23.
4
Pendyala L, Kidani Y, Perez R, Wilkes J, Bernacki RJ, Creaven PJ. Cytotoxicity, cellular accumulation and DNA binding of oxaliplatin isomers.
Cancer Lett
1995
;
97
:
177
–84.
5
Misset JL, Bleiberg H, Sutherland W, Bekradda M, Cvitkovic E. Oxaliplatin clinical activity: a review.
Crit Rev Oncol Hematol
2000
;
35
:
75
–93.
6
Grothey A, Goetz MP. Oxaliplatin plus oral fluoropyrimidines in colorectal cancer.
Clin Colorectal Cancer
2004
;
4
Suppl 1:
S37
–42.
7
Andre T, Boni C, Mounedji-Boudiaf L, et al. Oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment for colon cancer.
N Engl J Med
2004
;
350
:
2343
–51.
8
Raymond E, Faivre S, Chaney S, Woynarowski J, Cvitkovic E. Cellular and molecular pharmacology of oxaliplatin.
Mol Cancer Ther
2002
;
1
:
227
–35.
9
Rakitina TV, Vasilevskaya IA, O'Dwyer PJ. Additive interaction of oxaliplatin and 17-allylamino-17-demethoxygeldanamycin in colon cancer cell lines results from inhibition of nuclear factor κB signaling.
Cancer Res
2003
;
63
:
8600
–5.
10
Arango D, Wilson AJ, Shi Q, et al. Molecular mechanisms of action and prediction of response to oxaliplatin in colorectal cancer cells.
Br J Cancer
2004
;
91
:
1931
–46.
11
Griffiths GJ, Koh MY, Brunton VG, et al. Expression of kinase-defective mutants of c-Src in human metastatic colon cancer cells decreases Bcl-xL and increases oxaliplatin- and Fas-induced apoptosis.
J Biol Chem
2004
;
279
:
46113
–21.
12
Arnould S, Hennebelle I, Canal P, Bugat R, Guichard S. Cellular determinants of oxaliplatin sensitivity in colon cancer cell lines.
Eur J Cancer
2003
;
39
:
112
–9.
13
Samimi G, Manorek G, Castel R, et al. cDNA microarray-based identification of genes and pathways associated with oxaliplatin resistance.
Cancer Chemother Pharmacol
2005
;
55
:
1
–11.
14
Shirota Y, Stoehlmacher J, Brabender J, et al. ERCC1 and thymidylate synthase mRNA levels predict survival for colorectal cancer patients receiving combination oxaliplatin and fluorouracil chemotherapy.
J Clin Oncol
2001
;
19
:
4298
–304.
15
Martinez-Balibrea E, Manzano JL, Martinez-Cardus A, et al. Combined analysis of genetic polymorphisms in thymidylate synthase, uridine diphosphate glucoronosyltransferase and X-ray cross complementing factor 1 genes as a prognostic factor in advanced colorectal cancer patients treated with 5-fluorouracil plus oxaliplatin or irinotecan.
Oncol Rep
2007
;
17
:
637
–45.
16
Martinez-Balibrea E, Abad A, Aranda E, et al. Pharmacogenetic approach for capecitabine or 5-fluorouracil selection to be combined with oxaliplatin as first-line chemotherapy in advanced colorectal cancer.
Eur J Cancer
2008
;
44
:
1229
–37.
17
Safaei R. Role of copper transporters in the uptake and efflux of platinum containing drugs.
Cancer Lett
2006
;
234
:
34
–9.
18
Plasencia C, Martinez-Balibrea E, Martinez-Cardus A, Quinn DI, Abad A, Neamati N. Expression analysis of genes involved in oxaliplatin response and development of oxaliplatin-resistant HT29 colon cancer cells.
Int J Oncol
2006
;
29
:
225
–35.
19
van't Veer LJ, Dai H, van de Vijver MJ, et al. Gene expression profiling predicts clinical outcome of breast cancer.
Nature
2002
;
415
:
530
–6.
20
Ruzzo A, Graziano F, Loupakis F, et al. Pharmacogenetic profiling in patients with advanced colorectal cancer treated with first-line FOLFOX-4 chemotherapy.
J Clin Oncol
2007
;
25
:
1247
–54.
21
Bible KC, Kaufmann SH. Cytotoxic synergy between flavopiridol (NSC 649890, L86-8275) and various antineoplastic agents: the importance of sequence of administration.
Cancer Res
1997
;
57
:
3375
–80.
22
Churchill GA. Fundamentals of experimental design for cDNA microarrays.
Nat Genet
2002
;
32
Suppl:
490
–5.
23
Dobbin K, Shih JH, Simon R. Questions and answers on design of dual-label microarrays for identifying differentially expressed genes.
J Natl Cancer Inst
2003
;
95
:
1362
–9.
24
Giulietti A, Overbergh L, Valckx D, Decallonne B, Bouillon R, Mathieu C. An overview of real-time quantitative PCR: applications to quantify cytokine gene expression.
Methods
2001
;
25
:
386
–401.
25
Vandesompele J, De Preter K, Pattyn F, et al. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes.
Genome Biol
2002
;
3
:
RESEARCH0034
.
26
Boyer J, Allen WL, McLean EG, et al. Pharmacogenomic identification of novel determinants of response to chemotherapy in colon cancer.
Cancer Res
2006
;
66
:
2765
–77.
27
Faury D, Nantel A, Dunn SE, et al. Molecular profiling identifies prognostic subgroups of pediatric glioblastoma and shows increased YB-1 expression in tumors.
J Clin Oncol
2007
;
25
:
1196
–208.
28
Perona R, Sanchez-Perez I. Control of oncogenesis and cancer therapy resistance.
Br J Cancer
2004
;
90
:
573
–7.
29
Dihlmann S, Kloor M, Fallsehr C, von Knebel Doeberitz M. Regulation of AKT1 expression by β-catenin/Tcf/Lef signaling in colorectal cancer cells.
Carcinogenesis
2005
;
26
:
1503
–12.
30
Luo Y, Shoemaker AR, Liu X, et al. Potent and selective inhibitors of Akt kinases slow the progress of tumors in vivo.
Mol Cancer Ther
2005
;
4
:
977
–86.
31
Kim D, Dan HC, Park S, et al. AKT/PKB signaling mechanisms in cancer and chemoresistance.
Front Biosci
2005
;
10
:
975
–87.
32
Piccolo E, Vignati S, Maffucci T, et al. Inositol pentakisphosphate promotes apoptosis through the PI 3-K/Akt pathway.
Oncogene
2004
;
23
:
1754
–65.
33
Liu LZ, Zhou XD, Qian G, Shi X, Fang J, Jiang BH. AKT1 amplification regulates cisplatin resistance in human lung cancer cells through the mammalian target of rapamycin/p70S6K1 pathway.
Cancer Res
2007
;
67
:
6325
–32.
34
Kim E, Chen F, Wang CC, Harrison LE. CDK5 is a novel regulatory protein in PPARγ ligand-induced antiproliferation.
Int J Oncol
2006
;
28
:
191
–4.
35
Lin H, Juang JL, Wang PS. Involvement of Cdk5/p25 in digoxin-triggered prostate cancer cell apoptosis.
J Biol Chem
2004
;
279
:
29302
–7.
36
Sharma P, Veeranna, Sharma M, et al. Phosphorylation of MEK1 by cdk5/p35 down-regulates the mitogen-activated protein kinase pathway.
J Biol Chem
2002
;
277
:
528
–34.
37
Burgermeister E, Chuderland D, Hanoch T, Meyer M, Liscovitch M, Seger R. Interaction with MEK causes nuclear export and downregulation of peroxisome proliferator-activated receptor γ.
Mol Cell Biol
2007
;
27
:
803
–17.
38
Gupta RA, Dubois RN. Controversy: PPARγ as a target for treatment of colorectal cancer.
Am J Physiol Gastrointest Liver Physiol
2002
;
283
:
G266
–9.
39
Eichele K, Ramer R, Hinz B. Decisive role of cyclooxygenase-2 and lipocalin-type prostaglandin D synthase in chemotherapeutics-induced apoptosis of human cervical carcinoma cells.
Oncogene
2008
;
27
:
3032
–44.
40
Wang LH, Yang XY, Zhang X, Farrar WL. Inhibition of adhesive interaction between multiple myeloma and bone marrow stromal cells by PPARγ cross talk with NF-κB and C/EBP.
Blood
2007
;
110
:
4373
–84.
41
Biliran H, Jr., Banerjee S, Thakur A, et al. c-Myc-induced chemosensitization is mediated by suppression of cyclin D1 expression and nuclear factor-κB activity in pancreatic cancer cells.
Clin Cancer Res
2007
;
13
:
2811
–21.
42
Park ES, Choi S, Kim JM, et al. Early embryonic lethality caused by targeted disruption of the TRAF-interacting protein (TRIP) gene.
Biochem Biophys Res Commun
2007
;
363
:
971
–7.
43
Lee SY, Lee SY, Choi Y. TRAF-interacting protein (TRIP): a novel component of the tumor necrosis factor receptor (TNFR)- and CD30-TRAF signaling complexes that inhibits TRAF2-mediated NF-κB activation.
J Exp Med
1997
;
185
:
1275
–85.
44
Regamey A, Hohl D, Liu JW, et al. The tumor suppressor CYLD interacts with TRIP and regulates negatively nuclear factor κB activation by tumor necrosis factor.
J Exp Med
2003
;
198
:
1959
–64.
45
Wang CY, Mayo MW, Baldwin AS, Jr. TNF- and cancer therapy-induced apoptosis: potentiation by inhibition of NF-κB.
Science
1996
;
274
:
784
–7.
46
Arnold SM, Fessler LI, Fessler JH, Kaufman RJ. Two homologues encoding human UDP-glucose:glycoprotein glucosyltransferase differ in mRNA expression and enzymatic activity.
Biochemistry
2000
;
39
:
2149
–63.
47
Ollendorff V, Noguchi T, deLapeyriere O, Birnbaum D. The GARP gene encodes a new member of the family of leucine-rich repeat-containing proteins.
Cell Growth Differ
1994
;
5
:
213
–9.
48
Snow BE, Krumins AM, Brothers GM, et al. A G protein γ subunit-like domain shared between RGS11 and other RGS proteins specifies binding to Gβ5 subunits.
Proc Natl Acad Sci U S A
1998
;
95
:
13307
–12.
49
Szepetowski P, Ollendorff V, Grosgeorge J, et al. DNA amplification at 11q13.5-q14 in human breast cancer.
Oncogene
1992
;
7
:
2513
–7.
50
Cui Q, Ma Y, Jaramillo M, et al. A map of human cancer signaling.
Mol Syst Biol
2007
;
3
:
152
.