All procedures performed in studies involving human participants were in accordance with the ethical standards of Ethics Committee of Soonchunhyang University Cheonan Hospital and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.

Human colorectal cancer cell lines HCT116, HT29, LoVo, SW480, SW620, colo201, colo205, and CaCO₂ were purchased from the Korean Cell Line Bank (KCLB). Cells were grown in RPMI-1640 medium (Cellgro, USA) supplemented with 10% fetal bovine serum (Equitech-bio, USA) and 1X penicillin/streptomycin (Cellgro) at 37°C in a humidified atmosphere containing 5% CO₂.

Short hairpin RNA (shRNA) constructs were designed and cloned into the H1-shRNA vector (Genolution Pharmaceuticals Inc.). The target sequence for RhoA is 5′-CAGAAAAGTGGACCCCAGAA-3′. The sequence of nonsense shRNA against luciferase was provided by Genolution Pharmaceuticals Inc. The plasmids were transfected into HCT116 cells using Lipofectamine 2000 (Invitrogen, USA) according to the manufacturer's instructions. Briefly, HCT116 cells were cultured and transfected using Lipofectamine 2000. Medium was changed 6 h after transfection, and transfected cells were selected using Zeocin (Invitrogen). To evaluate transfection efficiency, semi-quantitative reverse transcriptase-polymerase chain reaction (RT-PCR) was used.

RNA was isolated using TRIzol (Invitrogen); equal amounts of RNA were converted to cDNA using ReverTra Ace^® qPCR kit (Toyobo, Japan) according to the manufacturer's instructions. To determine RhoA expression, PCR was performed using the Maxime PCR PreMix kit (iNtRON, Korea). Primer sequences are as follows: RhoA-F, 5′-CATCCGGAAGAAACTGGT-3′; RhoA-R, 5′-TCCCACAAAGCCAACTC-3′; GAPDH (glyceraldehyde-3-phosphate dehydrogenase)-F, 5′-CTTAGCACCCCTGGCCAAG-3′; GAPDH-R, 5′-GATGTTCTGGAGAGCCCCG-3′; Zeocin-F, 5′-CGACGTGACCCTGTTCATCAG-3′; Zeocin-R, 5′-GTTCGTGGACACGACCTCCGA-3′. The expected amplicon sizes were 168 bp (RhoA), 121 bp (GAPDH), and 130 bp (Zeocin). The PCR cycles consisted of a pre-denaturation step at 95°C for 10 min, followed by 35 3-temperature cycles (95, 59.5 and 72°C) for 30 sec each, and a final extension at 72°C for 5 min. The PCR products were confirmed using the QIAxcel auto electrophoresis system (Qiagen, USA).

Cell lysates were harvested using PRO-PREP (iNtRON, Korea) for 30 min on ice and centrifuged at 13,000 rpm for 5 min at 4°C. Protein concentration of the supernatant was determined by BCA assay (Thermo, USA). An equal amount of each protein extract (50 μg) was resolved using 10% polyacrylamide gel and electro-transferred onto 0.2 μm polyvinylidene fluoride (PVDF) membrane (Millipore, USA) using Trans-blot turbo (Bio-Rad Laboratories, Inc., USA). Membranes were immunoblotted with either 1:1,000-diluted mouse anti-RhoA monoclonal antibody (Abnova, Taiwan) or 1:5,000-diluted mouse anti-β-actin monoclonal antibody (Sigma, USA) overnight at 4°C. Membranes were incubated with 1:10,000-diluted horseradish peroxidase-conjugated anti-mouse immunoglobulin (Sigma) for 1 h at room temperature. The protein signal was detected by enhanced chemiluminescence (Advansta, USA) using the Molecular Imager ChemiDoc XRS+ System (Bio-Rad Laboratories, Inc.).

MTT assay was used to evaluate cell proliferation after transfection. Cells (1.0×10⁵ cells/well) were seeded into a 96-well plate and incubated for an additional 24–72 h post-transfection. After time-dependent incubation, the medium was removed and the cells were washed with PBS. The cells were incubated in a 5-mg/ml MTT (Sigma) solution for 4 h. Then, the media was substituted with dimethyl sulfoxide (DMSO; Sigma) and placed on the plate shaker for 15 min. Absorbance was read at 570 nm using a plate reader.

Cell migration was analyzed in vitro using the Culture insert system 24 (ibidi, Germany). The culture insert was attached to the bottom of a 6-well plate, and 100 μl of media containing 1.0×10⁶ cells were seeded into each well of the insert. The culture inserts were removed from the plate after 48 h, and cells were further cultured with fresh RPMI-1640 medium contained 10% FBS. The cell gap was monitored every 12 h for 48 h. Cell images were taken every 12 h with a phase contrast microscope, AxionCam camera (Zeiss, Germany).

The transwell culture insert was pre-coated with 50 μl of Matrigel (BD, USA) according to the manufacturer's instructions. Cells (5.0×10⁶) were suspended in serum-free RPMI-1640 medium and seeded into the pre-coated insert. Eight hundred microliters of RPMI-1640 medium containing 10% FBS was added outside the transwell culture insert. Cells were incubated at 37°C for 24 h in a humidified atmosphere with 5% CO₂. After washing the transwell insert twice with PBS, cells were fixed with 10% formaldehyde for 2 min. The cells were permeabilized with methanol for 20 min and then stained with methylgreen for 15 min. The transwell insert was washed twice with PBS, wiped using cotton swab, and then observed using an inverted microscope.

Immunostaining was conducted as previously described, with slight modification, to determine morphology change after transfection (15). Briefly, 1×10⁵ cells were seeded on a cover slip placed in a 6-well plate and incubated at 37°C in a humidified atmosphere with 5% CO₂ for 24 h. The wells were washed twice with cold PBS and fixed in cold methanol/acetone 1:1 (vol/vol) for 30 min at −20°C. The methanol/acetone mixture was then removed and the wells were dried for 15 min at room temperature. After rehydration with PBS for 15 min at room temperature, cells were incubated with blocking buffer (2 mg/ml BSA in PBS) for 1 h at room temperature, and then rinsed with blocking buffer three times. Cells were stained with 1 μg/ml DAPI for 2 min, sealed with clear nail polish, and visualized using an Olympus FV10 confocal microscope.

A total of 150 colorectal carcinoma tissue specimens were obtained from Soonchunhyang University Cheonan Hospital, Korea, where samples were collected from patients who underwent surgery between 2002 and 2007. These tissues were formalin-fixed and paraffin-embedded (FFPE). Clinicopathological data including age, gender, TNM classification, and distal metastasis are shown at Table I. Tumor stage was identified according to the American Joint Committee on Cancer TNM classification system. Sample collection for this study was approved by the Ethics Committee of Soonchunhyang University, Cheonan Hospital.

Immunohistochemical staining was performed using tissue microarray (TMA) block sections to determine RhoA expression in patient samples. The FFPE tumor tissues were re-embedded from each FFPE block to the recipient block in duplicate. Each TMA block contained 60 cores from 30 samples. For immunohistochemistry, 4-μm sections were obtained using a microtome, deparaffinized in xylene, and rehydrated in 100–70% alcohol series. Antigen retrieval was achieved in citrate buffer (pH 6.0) using a microwave for 15 min. To eliminate endogenous peroxidase activity, the sections were incubated in peroxidase blocking solution (Dako, Denmark) for 30 min and then washed with phosphate-buffered saline containing 0.1% Tween-20 (PBST). The sections were incubated with anti-mouse RhoA antibody (Abnova, 1:500) for 2 h at room temperature, followed by incubation in enhancer for 30 min and treatment with polymer for 1 h at room temperature. After washing with PBST, sections were incubated with DAB, counterstained with hematoxylin, and observed under a microscope.

The RhoA-stained tissue cores were examined by 2 independent observers (Chang-Jin Kim and Dongjun Jeong), and a consensus score was determined for each specimen. A positive reaction was scored into 4 grades, according to the intensity of the staining: 0, 1+, 2+, and 3+. The percentages of RhoA-positive cells were also scored into 4 categories: 0 (0%), 1 (1–30%), 2 (31–70%) and 3 (71–100%). The final score, calculated as the product of the intensity score multiplied by the percentage score, was classified as follows: 0 for negative; 1–3 for weak; 4–6 for moderate; and 7–9 for strong. Samples with a final score ≤3 were grouped together as RhoA expression negative while those with a score ≥4 were grouped together as RhoA expression positive.

Statistical analysis was conducted using SPSS 19.0 (Chicago, IL, USA) program. The results of RT-PCR, western blotting, and functional characterization of cells were analyzed with one-way ANOVA test and Student's t-test. The relationship between the result of immunohistochemistry and clinicopathological data was analyzed with Chi-square test. Hazard ratio and 95% confidence interval were evaluated using Cox regression models. Kaplan-Meier method was used to analyze disease-free survival rate using the log-rank test. A p-value of <0.05 was considered statistically significant in all assessments.

As RhoA has been implicated in multiple cancers including breast cancer, liver cancer, ovarian carcinoma, and gastric carcinoma, we examined the expression of RhoA in both non-metastatic and metastatic colorectal cancer cell lines. A non-metastatic cancer cell line, SW480, expressed a very small amount of RhoA, whereas many cancer cell lines including colo205, LoVo, and HCT116 exhibited high levels of RhoA expression (Fig. 1). Notably, the metastatic cell line HCT116, had the highest RhoA protein expression among the cell lines tested. Therefore, we used the HCT116 cancer cell line for further studies to determine the function of RhoA in the carcinogenesis of colon cancer.

To determine the functional consequence of RhoA, we used shRNA to knock down RhoA in HCT116 cells. The knockdown efficiency of RhoA was confirmed using RT-PCR and immunoblotting (Fig. 2). Specifically, HCT116 cells were transfected with either nonsense control (HCT116-vec) or shRhoA (HCT116-RhoAi). RNA and protein lysates from each pair of transfected cells were extracted and determined to assess RhoA levels. To determine the knockdown efficiency at the RNA level, RT-PCR was performed using cDNA from the samples. Amplicon size is 168 bp for RhoA, 121 bp for GAPDH, and 130 bp for zeocin. The PCR products were analyzed with the auto electrophoresis system, QIAxcel (Qiagen, USA). Zeocin was used to compare the transfection efficiency between nonsense control and shRhoA, and indicated similar transfection efficiencies (Fig. 2A). The expression of RhoA was significantly downregulated in shRhoA-treated HCT116 cells (HCT116-RhoAi) compared to untransfected (HCT116-con) and nonsense control (HCT116-vec) cells (p=0.028 and p=0.035, respectively). Specifically, the RhoA RNA level decreased by ~70% after shRhoA transfection, and there was no difference between the untransfected group (HCT116-con) and the control transfected group (HCT116-vec) (Fig. 2B).

As changes at RNA level do not always correlate with changes at the protein level, it is necessary to determine the protein level after knockdown. To test this, we also determined RhoA protein level after shRNA-mediated depletion. Protein lysates from each pair of transfected cells were collected and immunobloted with anti-RhoA Ab (Fig. 2C and D). Protein expression also displayed the same pattern observed in RNA after knockdown of RhoA. The protein level of RhoA was decreased by 70% in HCT116-RhoAi compared to untransfected cells or vector-transfected cells (p<0.0001). These results led us to confirm that RhoA was successfully knocked down.

To functionally define the role of RhoA, we first determined proliferation ability in the HCT116 cell line after RhoA knockdown using an MTT assay. After transfection of either control vector or shRhoA, cells were incubated for different time periods, ranging from 24 to 72 h, to determine proliferation rate. As shown in Fig. 3A, the proliferation of shRhoA-transfected cells was lower than that of HCT116-con or HCT116-vec cells (24 h, p<0.0001; 48 h, p<0.0001; 72 h, p=0.002). The proliferation between HCT-con and HCT116-vec cells was not significantly different. These results suggest that RhoA plays a role in proliferation of colon cancer cells.

In order to assess the effect of RhoA on migration of HCT116 cells, migration ability was determined after knocking down RhoA using a wound healing assay. Cells were seeded into culture insert, which was attached to the bottom of a 6-well plate, and the culture inserts were removed two days later. The cells were further cultured and monitored every 12 h (Fig. 3B and C). Over time, cells in control groups migrated gradually. However, the shRhoA-transfected group displayed approximately a 2-fold delay in migration at 12 h and a 3-fold delay at 24 h compared to control groups. More significantly, while control groups showed almost no gaps at 36 h, shRhoA group still displayed ~400-μm gaps, suggesting that RhoA-deficient cells were significantly impaired in the ability to migrate (Fig. 3C).

As previous results showed that Rho is essential for migration in HCT116 cells, we aimed to determine whether RhoA is also required for invasion. To achieve this goal, HCT116 cells were transfected and seeded on a Matrigel-coated insert. Cells were then incubated for an additional 48 h after seeding, and cells were imaged to assess cell invasion. HCT116 cells transfected with shRhoA exhibited significantly less invasion than control cells (Fig. 4A). Thus, we examined the morphological characteristics of HCT116 cells following RhoA knockdown. To assess cell morphology, nuclei of RhoA-depleted cells were stained with DAPI and analyzed with a confocal microscope. Vector-transfected or untransfected cells exhibited elongated morphology with pseudopodia, whereas RhoA-depleted cells were more rounded with less pseudopodia, and multi-nucleated in ~20% cells tested (Fig. 4B). These results indicate that RhoA is essential for cell migration and invasion of colon cancer cells, two important factors contributing to mortality caused by colon cancer.

Based on the conclusion that RhoA is essential for the migration and invasion in the HCT116 cell line, we then analyzed RhoA expression in patient colorectal cancer samples to define whether RhoA is relevant in human disease. The samples, obtained from 150 patients who had undergone surgery, were stained with RhoA antibody to determine RhoA expression. RhoA protein stained primarily in the cytoplasm with a wide range of intensity; RhoA stain intensity was graded from negative to strong expression (Fig. 5A–C and Table II). RhoA expression was confirmed to be positive in 86 cases out of 150 samples (57.3%). We further correlated RhoA expression with several factors including age, gender, pN stage, metastasis, and invasion, among others. RhoA expression was not significantly correlated with age, gender, and pN stage (Table III). However, RhoA expression was significantly associated with pT stage, vascular invasion, lymphatic invasion, and clinical stage (Fig. 5C–E and Table III), which is consistent with in vitro data demonstrating defective migration and invasion following RhoA knockdown (Figs. 3 and 4). Specifically, RhoA(+) samples exhibited an approximately 3- to 4-fold higher invasion rate of vascular invasion and lymphatic invasion compared to RhoA(−) samples (vascular invasion, 25.6% of RhoA(+) versus 7.8% of RhoA(−); lymphatic invasion, 29.1% of RhoA(+) versus 10.9% of RhoA(−); Fig. 5C and D and Table III). In addition, RhoA expression was associated with clinical stages of colorectal cancer; RhoA(−) specimens associated with lower clinical stages, while RhoA(+) specimens had higher clinical stages [ clinical stage I, 20.3% RhoA(−) versus 5.8% RhoA(+); clinical stage III, 39.1% of RhoA(−) versus 52.3% of RhoA(+); Fig. 5E and Table III]. These results indicate that, as tumors express more RhoA, the disease becomes more invasive, progressing to a higher clinical stage. Finally, we correlated RhoA expression with 5-year survival rate. Using univariate analysis, we found that RhoA expression was significantly associated with patient survival rate. As shown in Fig. 6, 92% of RhoA(−) patients survived whereas only 56% of RhoA(+) patients survived five years after surgery. These results indicate that RhoA is associated with invasion and a poor prognosis in colorectal cancer, and could serves as a promising therapeutic target for cancer therapy.