Human colon cancer cell lines SW480, HCT116, SW620 and cervical cancer cell line HeLa were purchased from the American Type Culture Collection (ATCC), cultured in Dulbecco’s modified Eagle’s medium (DMEM; Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (FBS; Invitrogen, Carlsbad, CA, USA) and 1% penicillin-streptomycin. All cell lines were maintained at 37°C in an atmosphere of 5% CO₂.

Fresh frozen specimens of matched colon cancer tissues and adjacent non-tumor tissues were from 20 patients in the Department of General Surgery of the Fourth Hospital of Hebei Medical University, obtained between March 2011 and January 2013, and stored at −80°C for further use. All of the tissues were from untreated patients who were undergoing surgery. Written informed consent was obtained from each patient before the surgery, and the study protocol was approved by the Institutional Research Ethics Committee.

Total RNA was isolated from colon cancer cell lines using TRIzol reagent (Invitrogen). Total cDNA was synthesized using the M-MLV Reverse Transcriptase (Fermentas, Vilnius, Lithuania). The mRNA expression levels were measured by qRT-PCR using the 7500 Real-Time PCR detection system (Applied Biosystems, Foster City, CA, USA). Amplification was performed with the SYBR Premix Ex Taq™ II kit (code: DRR081A; Takara, Dalian, China) according to a 2-step cycle procedure consisting of 45 cycles of denaturation at 95°C for 10 sec and annealing/elongation at 60°C for 30 sec. We measured mRNA levels semi-quantitatively by the Δ/Δ threshold cycle (Ct) method. GAPDH was used as internal control. Fold-changes were calculated and normalized as relative to the parental cells. The primers were used as follows: p21^cip1 F, 5′-AAATCGTCCAGCGACCTTCC-3′ and p21^cip1 R, 5′-GCCCTGTCCATAGCCTCTACT-3′; GAPDH F, 5′-CAAG GGCATCCTGGGCTACA-3′, GAPDH R, 5′-AAGTGGTCG TTGAGGGCAAT-3′.

Cells and clinical tissues were lysed on ice in lysis buffer (50 mM Tris-HCl, pH 7.4; 150 mM NaCl; 2 mM EDTA; 1% NP-40; and 0.1% sodium dodecyl sulfate SDS) containing freshly added protease inhibitor cocktail (Complete Mini; Roche Diagnostics, Branchburg, NJ, USA). Aliquots of samples with the same amount of protein were determined using the Bradford assay (Bio-Rad Laboratories, Hercules, CA, USA), and mixed with loading buffer (final concentrations of 62.5 mM Tris-HCl, pH 6.8, 2.3% SDS, 100 mM dithiothreitol and 0.005% bromophenol blue). Then the protein extracts were boiled and separated by SDS-PAGE and blotted to activated polyvinylidene difluoride (PVDF) membranes (Millipore, Billerica, MA, USA). Then the membranes were blocked with 5% fat-free milk and probed with first antibodies overnight. The first antibodies included the following: anti-Msi1 (1:500, sc-98845; Santa Cruz Biotechnology, Santa Cruz, CA, USA), anti-p21 (1:500, sc-397; Santa Cruz Biotechnology), and anti-β-actin (1:500, sc-47778; Santa Cruz Biotechnology). The membranes were then washed four times in PBST and incubated with a secondary antibody coupled to horseradish peroxidase (Thermo Fisher Scientific, Inc., New York, NY, USA), followed by ECL detection (Millipore) and visualization on X-ray film. Relative quantitation was determined with the AlphaView system (Cell Biosciences, Santa Clara, CA, USA) using β-actin as the loading control.

To generate plasmids that express Msi1-specific small interfering RNA (siRNA) which targeting MSI1, the oligonucleotide inserts were designed using an online siRNA design tool (Ambion, Austin, TX, USA). The double oligonucleotides were annealed and cloned into the lentivirus GV248-GFP shRNA vector (Shanghai Genechem Co., Ltd., Shanghai, China) as per the manufacturer’s introductions. Lentivirus was produced after GV248-GFP shRNA vector and two package plasmids were co-transfected to 293T cells with Lipofectamine 2000 transfection reagent (Invitrogen). Then colon cancer cells were infected by recombined lentivirus, and cell clones were selected to the amplification culture. Stably transfected clones were extracted and identified by western blotting.

A luciferase reporter vector (pMIR-REPORT; Ambion) was used to generate reporter constructs. Fragment of the 3′ UTR of human p21^cip1 was amplified from HCT116 cDNA by PCR using primers p21^cip1 WT F/R and cloned into the pMIR-REPORT luciferase plasmid. The mutant p21^cip1 3′ UTR was generated by using the p21^cip1 WT F/MU R and p21^cip1 MU F/WT R primers. The following primers were used: shMsi1-1F, 5′-GATCCTGTTACATGGTGTTTCGAATTCAAGA GATTCGAAACACCATGTAACATCA-3′ and shMsi1-1R, 5′-GACAATGTACCACAAAGCTTAAGTTGAGAAAGCT TTGTGGTACATTGTAGTTCGA-3′; shMsi1-2F, 5′-GATCC TCCTGTATCATATGTAAATTTCAAGAGAATTTACATA TGATACTGGACGA-3′ and shMsi1-2R, 5′-AGGACATAGT ATACATTTAAAGTTCTCTTAAATGTATACTATGACCT GCTTTCGA-3′; p21^cip1 WT F, 5′-ATTGAGCTCTAATCCGC CCACAGGAAG-3′ and p21^cip1 WT R, 5′-CTCAAGCTTACA AGTAAAGTCACTAAG-3′; p21^cip1 MU F, 5′-TGGGAAGCA GTGTCTTTCCTGGCACTAACGTT-3′ and p21^cip1 MU R, 5′-AGACACTGCTTCCCAGCCCCATATGAGCCCAC-3′.

Cells (5×10⁴) were cultured in triplicate in 35-mm cell culture dishes. The cells were harvested longitudinally, and counted every day for one week using a hemocytometer under light microscopy.

Cell viability was assessed every other day using 3-(4,5-dimethylthiazol-yl)-2,5-diphenyl tetrazolium bromide (MTT; Sigma-Aldrich) dye according to the standard protocol. Approximately 1,000 cells/well were seeded in a 96-well plate. MTT solution (20 μl) was added to 200 μl of culture media and incubated for 4 h, and then cell growth was determined by measuring the absorbance at 490 nm.

Cells (1×10³) of each type were cultured in 10-cm culture dishes and exposed to fresh media every 3 days. Colonies with diameters >0.2 mm were counted at day 14.

One hundred Msi1-KD or -control cells were seeded in 6-well plates in 2 ml of serum-free stem cell medium as described above. Fresh medium were added to each well every 3 days. The tumorspheres were analyzed on day 14 for tumorsphere forming ability.

Exponentially growing cells collected from stable transfectants were bilaterally inoculated into subcutaneous sites of 4- to 6-week-old Balb/c nude mice. Tumor dimensions were measured with calipers once every week, and volumes (cm³) were calculated according to the standard formula, length × width²/2. At the end of the experiment, tumors were dissected out, and the net weight per mouse was measured. The experimental protocols used herein were evaluated and approved by the Animal Care and Use Committee of the Fourth Hospital of Hebei Medical University.

Cells (1×10⁶) were harvested and washed twice with PBS, followed by fixation with 75% cold ethanol at 4°C overnight. After the cells were washed twice in PBS, the cells were suspended in PBS with 50 μg/ml propidium iodide (PI; Sigma) and 10 μg/ml RNaseA (Sigma). They were then incubated at 4°C for 30 min in the dark. Cells were analyzed for DNA content by FACSCalibur (BD Biosciences) and the results of the cell cycle were analyzed by FlowJo software.

Cells were lysed in 100 μl of passive lysis buffer (Promega, Madison, WI, USA) at 48 h after transfection and assayed with a Dual-luciferase report assay kit according to the manufacturer’s instructions. For each assay, cell extract (20 μl) was used and the reaction was started by injection of 50 μl of luciferase substrate. Each reaction was measured for 10 sec in the Luminometer. Luciferase activities were expressed as the ratio of firefly to Renilla luciferase activity.

Data are presented as the mean ± standard deviation (SD). The Student’s t-test was performed using the SPSS 16.0 (SPSS, Inc., Chicago, IL, USA). P<0.05 was regarded as statistically significant.

To assess Msi1 expression in clinical patients, western blot analysis was conducted on 20 pairs of colon cancer tissues and matched adjacent non-tumor tissues (Fig. 1A). Msi1 expression in all 20 cancer tissues was markedly higher than that in the corresponding non-tumor tissues (Student’s t-test, P<0.01; Fig. 1B), suggesting that the activation of Msi1 may contribute to the development and progression of colon cancer.

We also evaluated the expression of Msi1 in colon cancer cell lines by western blotting. We detected a high level of Msi1 expression in SW620, HCT116 and SW480 cells (Fig. 1C).

To determine whether Msi1 regulated the proliferation of colon cancer cell lines, the endogenous Msi1 was knocked down by short hairpin RNA (shRNA) in HCT116 and SW480 cells (Fig. 2A). The Msi1 knockdown HCT116 and SW480 cells (HCT116-KD and SW480-KD) had much lower proliferation ability than the corresponding control cells (HCT116-control and SW480-control), as measured by both cell growth curve assay (P<0.01; Fig. 2B) and cell viability assay (P<0.01; Fig. 2C), indicating that the knockdown of Msi1 suppressed the growth of colon cancer cells in vitro.

Furthermore, colony formation assays showed that the efficiency of foci formation was dramatically decreased in Msi1 knockdown clones compared with control clones (P<0.01; Fig. 2D). These findings suggested that enforced inhibition of Msi1 led to retardation of colon cancer cell growth in vitro.

As Msi1 modulates the proliferation in vitro, we explored whether it alters tumorigenic potential of colon cancer cells in vivo. Xenograft assays were created with nude mice, and the development and growth of solid tumors were monitored every week. The palpable tumor formation of HCT116-KD and HCT116-control cells occurred at the same time after inoculation. However, tumor development with HCT116-KD cells was significantly delayed (P<0.01; Fig. 3A). The tumor weights generated from HCT116-KD were reduced significantly (P<0.01; Fig. 3B). Similar results were observed from SW480 cells (Fig. 3A and B), suggesting that the silencing of Msi1 induces the suppression of the tumor formation and development of colon cancer, and this inhibition may have a close relationship with the decreased cell proliferation. Together, these data indicated that knockdown of Msi1 expression inhibited the growth of colon cancers.

Several studies have recently indicated the existence of CSCs (cancer stem cells) in various cancer types (23–25), and CSCs are critical for the maintenance of tumor growth, progression and resistance to chemotherapy or radiotherapy, as well as recurrence and metastasis (26). It has been demonstrated that Msi1 regulates the proliferation and differentiation of CSCs. In the present study, we used a tumorsphere culture system to conduct the latent role of Msi1 on tumorsphere formation. Knockdown of Msi1 inhibited the tumorsphere formation and growth of HCT116-KD and SW480-KD cell lines, as demonstrated by the significantly reduced numbers of tumorspheres compared with the numbers in control cells (P<0.01; Fig. 3C). Our results indicate that Msi1 is a key regulator of the growth of tumorspheres.

To further investigate the potential mechanism by which silence of Msi1 regulates growth of colon cancer cells, we characterized cell cycles by FACS analysis. As shown in Fig. 4A, the proportion of HCT116-KD cells in G₀/G₁ phase increased markedly to 69.3%, while proportion in S phase decreased to 19.0%. The ratio of G₁ phase to S phase of HCT116-KD cells was much higher than that of HCT116-control cells. A similar effect was observed in SW480-KD cells (P<0.01; Fig. 4B), indicating that the knockdown of Msi1 expression blocked G₁/S phase transition of cell cycle.

It has been reported that Msi1 blocked the G₁/S phase transition of cell cycle by negative regulation of cyclin-dependent kinase (CDK) inhibitor p21^cip1 in bladder carcinoma (21) and breast cancer (19). To test whether it also occurs in colon cancer cells, p21^cip1 mRNA and protein levels were examined by qRT-PCR and western blot analysis. Significantly, increased levels of p21^cip1 protein were observed in the lysates of both HCT116-KD and SW480-KD cells, but there was little change in p21^cip1 mRNA levels, suggesting that Msi1 may reduce p21^cip1 protein levels by inhibiting translation.

To confirm the direct interaction between Msi1 and its putative target site, the human p21^cip1 3′ UTR containing the wild-type and mutant Msi1 binding sequence was cloned downstream from the luciferase report element (Fig. 5E). The relative luciferase activity of wild-type p21^cip1 3′ UTR reporter was increased by 48% in HCT116-KD cells relative to controls, while the augmentation was diminished in cells transfected with the mutant reporter (P<0.01; Fig. 5F). Similarly, knockdown of Msi1 enhanced the relative luciferase activity of the wild-type in SW480 cells (P<0.01; Fig. 5F). Mutation of the Msi1 binding site abrogated the elevate effect, demonstrating the specificity of the Msi1 target sequence.