The colorectal cell lines HT-29, SW620, RKO and LOVO were obtained from Dr Zeng (State Key Laboratory of Oncology in Southern China, Sun Yat-sen University Cancer Center, Guangzhou, China) and were grown in RPMI-1640 medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS). 293T and SW480 cell lines were purchased from the Shanghai Institute of Cell Biology (Shanghai, China). 293T and SW480 cells were separately cultured in Dulbecco's modified Eagle's medium (DMEM) and RPMI-1640 medium with 10% FBS at 37°C in a humidified atmosphere of 5% CO₂.

Paraffin-embedded, archived colorectal cancer samples obtained from 94 patients were histologically and clinically diagnosed at Fudan University Pudong Medical Center between 2007 and 2009. Of the 94 colorectal cancer tissues, 52 matched adjacent non-cancerous tissues were used as controls. Prior to the use of these clinical materials for investigation, informed consent from patients and approval from the Institute Research Ethics Committee were obtained. Primary cancers of the colorectum were classified according to the pathological TNM classification (15). Clinical information of the samples is described in detail in Table I. Patients included 47 males and 47 females, with ages ranging from 24 to 85 years (mean, 65.4 years). The data for metastasis pertain to its presence at any time during the follow-up. The median follow-up time for overall survival was 59.0 months for patients still alive at the time of analysis and ranged from 6 to 86 months. A total of 39 (41.5%) patients died during follow-up.

Three target sequences for VMP1 mRNA were chosen according to the RNAi Consortium (TRC) shRNA Library (Broad Institute) (16). The VMP1 shRNA single-strand oligonucleotides were: 1F, 5′-CCGGGTGCTTATAGCTACGTATTATCTCGAGATAATACGTAGCTATAAGCACTTTTTG-3′ and 1R, 5′-AATTCAAAAAGTGCTTATAGCTACGTATTATCTCGAGATAATACGTAGCTATAAGCAC-3′; 2F, 5′-CCGGCAACAGTATGTGCAACGTATACTCGAGTATACGTTGCACATACTGTTGTTTTTG-3′ and 2R, 5′-AATTCAAAAACAACAGTATGTGCAACGTATACTCGAGTATACGTTGCACATACTGTTG-3′; 3F, 5′-CCGGGCAATGAACAAGGAACATCATCTCGAGATGATGTTCCTTGTTCATTGCTTTTTG-3′ and 3R, 5′-AATTCAAAAAGCAATGAACAAGGAACATCATCTCGAGATGATGTTCCTTGTTCATTGC-3′.

Two complementary single-strand oligonucleotides containing the target sequences were synthesized (Shanghai, China) chemically and annealed. The double-stranded oligonucleotides were inserted between the AgeI and EcoRI restriction sites in the pLKO-green fluorescent protein (GFP) small interfering RNA (siRNA) vector that contains cytomegalovirus-driven enhanced green fluorescent protein (EGFP) reporter gene. The ligated plasmid was transformed into E. coli DH5α competent cells for plasmid amplification. The plasmids from positive colonies were identified by RT-PCR and DNA sequencing.

Recombinant lentiviruses were produced by co-transfecting 293T cells with three combinant lentiviral vectors, Δ8.91 and pVSV-G (10:10:1) using the cationic lipid complex method (X-tremeGENE HP DNA Transfection Reagent; Roche). The culture supernatants containing the produced viruses were harvested at 48 h after transfection, and concentrated by centrifugation at 4,000 × g at 4°C for 10–15 min. SW480 cells were subcultured at a density of 1×10⁶ cells/well. Cells were divided into 4 groups: CON (infected with negative control lentiviral vector) and VMP1 shRNA 1–3 (infected with the VMP1 lentiviral vectors). SW480 cells were plated into 6-well dishes at 5×10⁵ cells/well. The next day, the cells were infected with the same viral titer with 2 μg/ml Polybrene. After 24 h post-infection, the media were replaced with media containing 2 μg/ml puromycin. Cells were maintained and allowed to grow for 7–9 days; media were replaced without puromycin and then passaged for follow-up assays.

Equal numbers of cells were lysed in lysis buffer composed of 0.6 M Tris-HCl (pH 6.8), 10% SDS and protease inhibitor cocktail. Samples were incubated at 4°C for 10 min, and then centrifuged at 10,000 × g for 15 min at 4°C. The supernatants were transferred, mixed and boiled in sample buffer. The supernatants were separated by polyacrylamide gel electrophoresis and transferred to a PVDF membrane (Bio-Rad, Hercules, CA, USA). Then incubation of the membrane was carried out at room temperature in blocking buffer consisting of 5% fat-free milk dissolved in 1X TBST (10 mM Tris-base, pH 7.5, 100 mM NaCl and 1% Tween-20) for 1 h followed by incubation with the blocking buffer containing the primary antibody, such as anti-VMP1, anti-AKT, anti-p-AKT, anti-ZO-1 (Cell Signaling Technology, Danvers MA, USA), anti-E-cadherin or anti-β-actin (BD, Shanghai, China) at 4°C overnight. The membrane was washed with TBS-T, and incubated with the secondary antibody for 1 h. The blot was exposed to ECL (Amersham) after TBS washing.

Cells of the SW480 CON group (infected with the negative control lentiviral vector) and the SW480 VMP1 shRNA1 group (infected with the VMP1 lentiviral vector) were seeded into a 96-well microplate at a density of 5×10³ cells/well and incubated overnight in 10% FBS medium. Each group had 3 wells. After incubation for 12, 24, 48 and 72 h at 37°C, MTT reagent (5 mg/ml) in phosphate-buffered saline (PBS) was added to the cells (10 μl/well), and the cells were incubated for 4 h at 37°C. The supernatant was discarded and 200 μl of dimethylsulfoxide (DMSO) was added to each well to solubilize the MTT-formazan product. Finally, samples were measured on a multi-well spectrophotometer (Thermo, USA) at a test wavelength of 560 nm with a reference wavelength of 650 nm.

Cells were seeded at 5,000 cells/well into 96-well plates (200 μl/well) and allowed to grow for 24 h before treatment with the complex. Different concentrations of cisplatin, 5-FU and oxaliplatin were added into the cells; each group had 3 wells. After a further 24 h, MTT reagent (5 mg/ml) in PBS was added to the cells (10 μl/well) and incubated for 4 h at 37°C. Two hundred microliters of DMSO was added to each well to solubilize the MTT-formazan product after removal of the medium. Samples were measured on a multi-well spectrophotometer (Thermo) at a test wavelength of 595 nm and a reference of 650 nm.

Assays were performed in Transwell chambers (Corning Inc., USA) with an 8-μm pore size coated with Matrigel (BD, USA). SW480 cells infected with the scrambled shRNA and VMP1 shRNA1 were trypsinized and suspended in 1% FBS. After counting, the cells were plated into the upper chamber and 600 μl medium supplemented with 10% FBS was placed into the lower chamber. After incubation at 37°C for 6 h, cells on the upper surface of the filters were removed and cells adhering to the undersurface of the filter membrane were dyed with 0.5% crystal violet for 30 min. The crystal violet was washed with PBS for 3 times. Cells on the lower chamber were counted under a microscope in four fields randomly. The mean cell numbers were recorded and analyzed. The experiment was repeated 3 times.

IHC was carried out to study altered protein expression in 94 human colorectal cancer tissues and 52 matched adjacent non-cancerous tissues. A commercially available antibody against VMP1 (1:100 lot ab116006, rabbit polyclonal immunoglobulin G; Abcam Biotechnology, USA) was used as the primary antibody. An immunohistochemical kit (SP-9001 rabbit SP kit, lot 50581654) was obtained from Zhongshan Golden Bridge Biotechnology Co., Ltd. (Beijing, China). For each sample, one score was given according to the percent of positive cells as: no positive cells, 0; <5% positive cells, 1 point; 5–35% positive cells, 2 points; 36–70% positive cells, 3 points; >70% positive cells, 4 points. To achieve objectivity, the intensity of positive staining was also used in a four-tier scoring system: 0 (negative staining), 1 (weak staining exhibited as light yellow), 2 (moderate staining exhibited as yellow brown), and 3 (strong staining exhibited as brown). A final score was then calculated by multiplying the above two scores. If the final score was ≥4, the tumor was considered to have high expression; otherwise, the tumor was considered to have low expression (17).

Cells were washed twice with cold PBS and were collected by centrifugation at 1,000 × g for 5 min. Total RNA was extracted using TRIzol^® reagent (Invitrogen) following the manufacturer's protocol. cDNA was generated with oligo(dT) primers. Primers were designed online using the Primer3 designer program (website, http://frodo.wi.mit.edu/) and were as follows: VMP1 forward, 5′-GAC CAG AGA CGT GTA GCA ATG-3′ and reverse, 5′-ACA ATG CTT TGA CGA TGC CAT AA-3′; GAPDH (housekeeping gene) forward, 5′-GGA GAT TGT TGC CAT CAA CG-3′ and reverse, 5′-TTG GTG GTG CAG GAT GCA TT-3′. The reaction systems consisted of 0.2 μl cDNA, 2.5 μl 10X Taq enzyme buffer, 1 μl 10 mol/l dNTP, the forward and reverse primers were 0.2 μl and 0.2 μl Taq enzymes. Water was then added to 25 μl. PCR conditions were as follows: denaturation at 95°C for 2 min; each cycle consisting of 1 min at 94°C, 1 min at 56°C; 1 min and 15 sec at 72°C; and a final extension of 5 min at 75°C. After the reaction, the PCR products were separated on 2% agarose gel (BioWest) and visualized under UV light.

Mice were treated in accordance with the NIH Guide for the Care and Use of Laboratory Animals (18), and as approved by the Ethics Committee of Ningbo University. Mice were housed in a temperature-controlled room with a proper dark-light cycle, fed a regular diet, and maintained under the care of the Laboratory Animal Unit, Ningbo University, China. The mice were acclimated for 1 week before the experiment. VMP1^KD and VMP1^scramble cells growth in logarithmic phase were trypsinized, and the cell suspension was harvested. After staining with trypan blue, the cells were counted under a hemocytometer to test the viability. Cell concentration was adjusted with culture medium to 1×10⁷ cells/ml. The nude mice were randomly divided into 2 groups. In the VMP1 knockout group, 10 mice were intraperitoneal injected with VMP1^KD cells, while in the scramble shRNA group, 10 mice were injected with the VMP1^scramble cells. Each nude mouse was inoculated with 0.5 ml of the cell suspension. The needle was stopped internally for 5 sec, rotated and pulled out to avoid leakage of the cell suspension. The activity, diet and mental state of the nude mice were observed daily. On day 20 of the experiment, all of the mice were sacrificed by cervical vertebra disjointing, the abdominal cavity was opened and the nodules within the abdominal were stripped, counted and confirmed by fluorescence microscopy.

All statistical analyses were carried out using the SPSS 13.0 statistical software package. The χ² test was employed to evaluate differences in expression of VMP1 between the two categories of tissues. The Mann-Whitney U test was used to analyze the relationship between VMP1 expression and clinicopathological characteristics. Survival curves were plotted by the Kaplan-Meier method and compared by the log-rank test. The significance of various variables for survival was analyzed by the Cox proportional hazards model in the multivariate analysis. P<0.05 in all cases was considered to indicate a statistically significant result.

Immunohistochemical assay showed that the expression of VMP1 in the same patients was significantly decreased in primary cancer tissues when compared with that in the matched adjacent non-cancerous tissues (χ²=12.962, P=0.008; Fig. 1). Among the matched adjacent non-cancerous tissues, 61.5% (32 of 52) of the individual tissues had either high or strong VMP1 expression (Fig. 1Aa, b, e and g). However among the colorectal cancer cases only 30.9% (29 of 94) of the cancer tissues were defined as having high expression of VMP1 (Fig. 1Ac–f). The subcellular location of VMP1 was mainly cytoplasmic.

Table I shows the relationship between the expression of VMP1 protein and clinical characteristics. There was no significant correlation between the expression level of VMP1 protein and age, histological classification, histological differentiation, tumor diameter, pT classification or distant metastasis of the colorectal cancer patients. However, the expression of VMP1 was closely associated with stage of the colorectal cancer patients (P=0.008) and pN classification (P=0.002). The expression of VMP1 protein was negatively correlated with staging and pN classification (Table II). As shown in Fig. 1B, higher staging was correlated with lower VMP1 expression.

Kaplan-Meier analysis and the log-rank test were used to calculate the effect of classic clinicopathological characteristics (including stage, T classification, N classification) and VMP1 expression on survival. The expression level of VMP1 protein in the colorectal cancer cases was significantly correlated with the patient survival time (P=0.021), indicating that lower levels of VMP1 expression were correlated with shorter survival time. The high VMP1 expression group had better survival, whereas the low VMP1 expression group had shorter survival (Fig. 2). The median survival of patients with high VMP1 expression was much longer (62 months) than the median survival of patients with low VMP1 expression (55 months) (P=0.021, log-rank).

In addition, T classification, N classification and stage were also significantly correlated with survival in the Kaplan-Meier analysis and log-rank test (for T classification, P=0.026; for N classification, P<0.001; for stage, P<0.001). We did multivariate survival analysis, which included VMP1 expression level, stage, T classification and N classification, to determine whether VMP1 expression is an independent prognostic factor of patient outcome. In this analysis, stage was recognized as an independent prognostic factor, but VMP1 expression was not an independent prognostic factor of outcome (Table II).

The expression of VMP1 at the protein and mRNA levels in different colorectal cancer cell lines was detected. A negative correlation between VMP1 expression and metastasis was found. Namely, the cell lines expressing VMP1 had a significantly lower ability for metastasis and vice versa. These results suggest that VMP1 may act as a key factor for preventing the progression of colorectal cancer as well as in balancing the level between autophagy and apoptosis (Fig. 3).

In order to clarify the role of VMP1 in the progression of colorectal cancer, a shRNA lentiviral vector was constructed to silence the VMP1 gene for heritable changes. We designed 3 pairs of shRNAs. After sequencing was confirmed and amplification in the plasmid, we packaged the 293T cells with liposome and plasmid for a transfection of 48 h. After that, the lentivirus in the culture supernatant was harvested to infect the target cells. Results of fluorescence (Fig. 4A) showed that the cell infection rate reached ~98%. In contrast, the western blotting indicated that shRNA1 was the best sequence for silencing VMP1 at nearly 95% (Fig. 4B and C). Thus, the colorectal cancer cells were used for subsequent experiments.

First, we evaluated the cell proliferation after VMP1 silencing, which indicated a significantly higher proliferation rate at 48 h and a balanced level at 72 h compared with the scrambled shRNA group (Fig. 5A). As shown in Fig. 5C, determination of drug sensitivity indicated an increasing sensitivity to 5-FU after VMP1 silencing but an unconspicuous function on platinum, which resulted from the increased sensitivity to apoptosis signaling.

We used a Transwell chamber model to study the ability of cell migration. As expected, a significant increase in migration was found after VMP1 silencing (Fig. 5B). Western blotting showed an increase in AKT/p-AKT as well as a downregulation of ZO-1, which was related to the activation of the PI3K/AKT signaling pathway and the phosphorylation of ZO-1. In addition, expression of E-cadherin was also downregulated at the same time that acts as a key factor inducing the EMT of tumor cells and promoting migration (Fig. 6).

Finally, we compared the tumor formation in a nude mouse model. The results indicated that the tumor formation rate was 100%. There were no significant differences among the two groups of mice at day 7 after inoculation of the normal SW480 cell and the silenced ones. Furthermore, listlessness, poor appetite and slow action were found in the VMP1 shRNA group but no significant changes were noted in the scrambled shRNA group 7 days later. In addition, 20 days after inoculation, separation of nodules surrounded by intestinal tissue was carried out, to evaluate the biological characteristics by counting the nodules with fluorescent signaling. An increasing number of nodules was formed in the VMP1 silenced group compared with the GFP groups (Fig. 7).