Characterization of RhoA-mediated Chemoresistance in Gastric Cancer Cells

1) Materials

The lovastatin was obtained from Merck & Co. (San Diego, CA). The other chemicals, protease inhibitors and chemotherapeutic drugs were purchased from Sigma Chemical Co. (St Louis, MO). Fetal bovine serum and tissue-culture media were purchased from Life Technologies, Inc. (Gaithersburg, MA). The enhanced chemiluminescence detection reagents were obtained from Amersham Corp. (Arlington Heights, IL) and the antibodies for RhoA and PARP from Santa Cruz Biotechnology (Santa Cruz, CA).

2) Cell-growth and cytotoxicity assays

The SNU638 human carcinoma cell line (Korean Type Culture Collection, Seoul, Korea) was maintained in RPMI 1,640 (Life Technologies, Gaithersburg, MD), containing 1 mM glutamate, 100 U/ml penicillin, 100 ng/ml streptomycin and 10% FCS. Cells were cultured and passaged at 37℃ in a humidified atmosphere of 5% CO₂. For the MTT assay, cells were seeded at 1~5×10³ cells/well in a flat-bottomed 96-well tissue-culture plate, treated with or without drugs, and incubated for various times. Before the end of the experiment, the MTT labeling mixture was added and the cells incubated for another 4 h. The viable cell mass was assessed as recommended in the manufacturer's protocol (Promega, Madison, WI). The absorbance of the product was measured at a wavelength of 540 nm, with a reference wavelength of 650 nm. The cell viability was expressed as the ratio of viable cell mass following a given treatment to that of the untreated cells ×100(%).

3) Transfection of SNU638 cells

SNU638 cells were transfected using FuGene (Roche Molecular Biochemicals, Indianapolis, IN), as recommended by the manufacturer, with either a RhoA plasmid in which the full-length RhoA cDNA was cloned into the EcoRI site of the expression vector pcDNA3.1 (Invitrogen, Carlsbad, CA) or a similar plasmid expressing caRhoA as the result of the V14 mutation. After transfection, the cells were incubated for 48 h; the medium was changed, and the transfected cells selected for 2 weeks in the presence of geneticin.

4) Western blot analysis

After experimental treatments, the cells were harvested and the cell pellets washed with ice-cold phosphate-buffered saline and lysed in lysis buffer, containing 10 mM Tris (pH 7.5), 150 mM NaCl, 10 mM ethylene diamine tetra acetic acid (EDTA), 1% sodium dodecyl sulfate (SDS), 1 mM sodium orthovanadate and a mixture of protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 µg/ml pepstatin A, 2 µg/ml aprotinin). Lysates were sonicated for 10 s, centrifuged for 20 min at 20,000×g, and stored at -70℃. Equal amounts of lysate proteins were run on 10% SDS-polyacrylamide gels and transferred electrophoretically to nitrocellulose. The nitrocellulose blots were blocked with 5% skimmed milk in Tris-buffered saline with 0.05% Tween-20 (TBST), then incubated at 4℃ with antibody in TBST, containing 5% bovine serum albumin, overnight. Immunoreactivity was detected using enhanced chemiluminescence.

5) cDNA microarray analysis

Total RNA from SNU638/pcDNA3 and SNU638/RhoA cells was purified using the RNeasy mini kit (Qiagen, GmbH, Germany). The relative abundance of a specific transcript was evaluated using the MICROMAX human cDNA microarray system (PE Applied Biosystems, Foster, CA), which contains probes for the detection of approximately 2,400 human gene sequences. A list of a set of genes on the filter is shown in the web site: (http://www.perkinelmer.com/lifesciences). Microarray probe and hybridization were performed as described in the instruction manual. The gene expression images were quantified by measuring the intensity of the signals using Imagene (BioDiscovery, Los Angels, CA). The signal intensity of the filters was compared using the Genespring program (Silicon Genetics, Redwoods, CA).

The fluorescence intensity measured for each sample's chip was normalized to the average fluorescence of the entire sample. To identify differentially expressed RNA species, mocktransfected SNU638/pcDNA3 was considered the baseline sample, with SNU638/RhoA as the experimental sample. Only genes significantly detected in two samples (considered significant based on a p value of <0.05) were considered in the identification of the differentially expressed transcripts. Out of these transcripts, a pair-wise comparison analysis was performed, and the changes denoted as an increase or decrease in the gene expression. The fold change in the signal intensity of a gene was calculated, and only those with more than a 2-fold increase in the RhoA transfectant over that of the base line control were considered as examples of increased gene expression.

6) Statistical analysis

Where appropriate, data were expressed as the mean±SD based on at least three experiments. The Student's t test was used for directly comparing data, with a p value <0.05 considered significant.

1) RhoA expression enhances lovastatin resistance in gastric cancer cells

To study the chemoresistance-associated role of RhoA, stable wild-type RhoA (SNU638/RhoA) and constitutively active RhoA (SNU638/caRhoA) expressing systems in SNU638 gastric cancer cells were generated. Ectopic expression of transfected RhoA or caRhoA was confirmed by Western blot analysis (Fig. 1A). Because RhoA is a target of lovastatin, the response of SNU638/RhoA and SNU638/caRhoA cells to lovastatin were compared with parental cells to verify the activity of the expressed recombinant RhoA. IC₅₀ curves were generated after treatment of cells with a range of different lovastatin concentrations. Forced expressions of RhoA and caRhoA increased the lovastatin resistance in SNU638 cells, with an overall increase in the IC₅₀ from 7.66±0.55 to 23.55±0.75 and 71.88±3.23 µg/ml, respectively (Fig. 1B).

Fig. 1

Ectopic expression of RhoA induces resistance to lovastatininduced cell death. (A) Western blot analysis demonstrating ectopic expressions of myc-tagged recombinant RhoA and caRhoA in SNU638 cells. (B) Dose response curves detailing the effect of stable expressions of RhoA and caRhoA on the cytotoxicity of lovastatin in SNU638 cells. Cells (5×10³/well) were plated in 96-well plates and treated with lovastatin for 72 h on the following day. A MTT assay was performed to detect live cells. IC₅₀ values were calculated from the sigmoidal dose response curves.

2) RhoA is a differential modulator of chemosensitivity

To examine the role played by RhoA in mediating resistance to chemotherapeutic drugs, dose response curves were generated for the SNU638/pcDNA3.1, SNU638/RhoA and SNU638/caRhoA cells after treatment with the DNA cross-linking agents cisplatin, anti-metabolite 5-fluorouracil, topoisomerase II poison etoposide, anti-microtubule agent taxol and vinca alkaloid vincristine. The cells were plated at equal densities and exposed to increasing doses of each drug for 96 hours, after which time the cell survival was estimated using the MTT assay. SNU638/caRhoA cells displayed a 1.3-fold increase in resistance to 5-FU, with the IC₅₀ value increasing from 0.33±0.03 µg/ml in the SNU638 cells to 0.44±0.03 µg/ml in the SNU638/caRhoA cells (p=0.0094, Table 1). Similar to that observed for 5-FU, the SNU638/caRhoA cells displayed a significant increase in resistance to taxol compared to the SNU638/RhoA cells. Specifically, the SNU638 cells displayed an IC₅₀ value of 4.73±0.26 ng/ml compared with respective IC₅₀ values of 39.12±3.45 and 12.82±0.43 ng/ml in the SNU638/caRhoA and SNU638/RhoA cells (p<0.005), suggestive of RhoA-induced resistance to taxol in these cells (Table 1). SNU638/RhoA and SNU638/caRhoA cells displayed significant increases in resistance to vincristine compared with SNU638/pcDNA cells, displaying IC₅₀ values of 0.47±0.02 and 1.01±0.13 ng/ml, respectively, compared with that of 0.39±0.02 ng/ml in the control SNU638/pcDNA cells (p<0.0005), suggesting RhoA induces resistance to vincristine in these cells (Table 1).

Table 1

Differential chemosensitivity of SNU638 clones expressing RhoA or caRhoA

However, treatment with cisplatin or etoposide resulted in respective IC₅₀ values of 0.43±0.04 and 0.46±0.01 µg/ml in the parental control cells, whereas the respective IC₅₀ values were 0.44±0.05 and 0.42±0.03 µg/ml in the SNU638/RhoA cells (p>0.05, Table 1). These results reflect a differential resistance pattern of RhoA, which was clearly dependent on the chemotherapeutic drugs (Table 1).

3) PARP cleavage is blocked by RhoA

To examine whether the observed RhoA-induced resistance was due to the inhibition of an apoptotic pathway component, Western blots were conducted and the PARP cleavage after the treatment with drugs analyzed, which induced about 50% apoptosis in the treated cells after 2 days at the given concentration. The PARP assays demonstrated cleavage of the PARP protein when SNU638 cells were treated with cisplatin, etoposide, 5-Fu, taxol and vincristine (Fig. 2). However, in agreement with the cell survival data, SNU638/RhoA and SNU638/caRhoA cells were resistant to the cleavage induced by taxol or vincristine; whereas, the parental or mock cells displayed the characteristic 85 kDa PARP cleavage product.

Fig. 2

RhoA is a selective modulator of chemotherapeutic drug-induced apoptosis. PARP cleavage assay demonstrates the absence of the cleaved PARP fragment in RhoA (or caRhoA)-transfected cells compared with the parental (SNU638) or mock-transfected cells (pcDNA3.1), after treatments with either 2 µg/ml cisplatin, 5 µg/ml etoposide, 1 µg/ml 5-FU, 2 ng/ml taxol or 10 ng/ml vincristine. Identical blots were reprobed with β-actin as a control for loading.

4) Microarray analysis reveals RhoA targets that may account for drug resistance

The ability of the transcription factor RhoA to increase chemoresistance in a gastric cancer cell line was examined via up-regulation of the target genes. To identify potential target genes, a microarray analysis was carried out, and the differential gene expressions compared between mock and SNU638/RhoA cells. A total of 19 up-regulated genes were identified with at least 2-fold increases in their expressions in the SNU638/RhoA cells (Table 2). Among these there were proteins involved in antioxidant defense and potential detoxification of chemotherapeutic drugs, such as superoxide dismutase. A particular gene, involved in protein ubiquitination, was also up-regulated in the RhoA-transfectants, including deubiquitinating enzyme UnpEL. Genes associated with glucose uptake, such as glucose transporter-like protein-III, were elevated in RhoA-transfected cells. Interestingly, IL-1 receptor-associated kinase (IRAK), which is known to be involved in the IKK phosphorylation and subsequent activation of NF-κB, is also induced by RhoA, implicating that cell death resistance of RhoA could be generated by elevation of these anti-apoptotic proteins.

Table 2

Results of microarray analysis

5) RhoA increases the expression of drug resistance-related proteins

To further examine whether drug resistance-related proteins could be modulated by RhoA, Western blot analyses were performed using SNU638, SNU638/pcDNA3, SNU638/RhoA and SNU638/caRhoA cells. Drug resistance-related proteins, such as vascular endothelial growth factor (VEGF), epidermal growth factor receptor (EGFR), lung resistance protein (LRP) and glutathione S-transferase (GST-π) (18~21), were modulated by either RhoA or caRhoA (Fig. 3), indicating that RhoA expression or signaling induces these proteins, which may subsequently result in cellular chemoresistance.

Fig. 3

Western blot analyses of chemoresistance-related proteins in total cell lysates from parental (SNU638), mock-transfected (pcDNA3.1), RhoA overexpressing (RhoA) and caRhoA expressing (caRhoA) cells. Identical blots were reprobed with β-actin as a control for loading.