HCT8 cells were maintained in Roswell Park Memorial Institute (RPMI)-1640 (Invitrogen, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS) at 37°C in a humidified 5% CO₂ atmosphere. DNA transfection was performed using Lipofectamine (Invitrogen) according to the manufacturer's protocol. For stable transfection, cells were selected using full growth medium containing G418 (600 μg/ml) for 14 days, and the positive cells were detected by western blot analysis.

Wild-type pcDNA3.1/UCHL1 expression plasmid against UCHL1 was a gift from Dr Julia Shackelford. The pcDNA3.1/UCHL1-C90S mutant was generated by inserting a specific mutation at the Cys 90 site in wild-type pcDNA3.1/UCHL1 plasmid by overlap PCR.

Cells were seeded in triplicate into 96-well plates with 3×10³ cells/well. Cells were maintained for 5 days, and then 20 μl of MTT (Sigma, USA) solution (5 mg/ml) was added and cells were further incubated at 37°C for 4 h. Absorbance at 492 nm was determined with a microplate reader. All MTT assays were repeated three times.

Cell invasion and migration assays were carried out in a chemotaxis chamber (Millipore). Cells (1×10⁵) suspended in 200 μl of serum-free medium were loaded into the upper chamber and allowed to pass through an 8-μm-pore polyethylene terephthalate filter. The lower chamber was filled with complete medium. The chamber was either pre-coated with Matrigel (1:8; BD Biosciences, Bedford, MA, USA) for invasion assays or left uncoated for migration assays. Cells that failed to pass through the filters were removed by scrubbing with cotton swabs after 20 h (migration assay) or 24 h (invasion assay). Cells on the undersurface were fixed in methanol and stained with 0.5% crystal violet, then photographed and quantified in 5 random fields per membrane. Each sample was assayed in triplicate.

For the tumor growth assay, cells (40 μl, 4×10⁶ cells) were collected and inoculated subcutaneously into the nail pad of 4-week-old male nude mice. Experimental and control groups had 5 mice each. Mice were sacrificed 38 days later. Tumor growth curves were calculated. After tumor excision, the tissues were fixed in 10% buffered formalin. All formalin-fixed and paraffin-embedded samples were carefully examined after staining with hematoxylin and eosin (H&E) and UCHL1 antibody.

For the tumor metastasis assay, cells (50 μl 2.5×10⁶ cells) were collected and injected into the tail vein of 4-week-old male nude mice. Each group included 6 mice. Mice were sacrificed 4 weeks later and examined for metastasis development. Lungs and visibly observable lymph nodes were excised. All lungs were fixed with Bouin's fluid. All lymph nodes were formalin-fixed and paraffin-embedded, then carefully examined after staining with H&E and UCHL1 antibody.

All animal experiments were conducted according to the institutional and national guidelines for animal experiments.

Immunohistochemical (IHC) staining was performed as previously described (16). Specimens were incubated with anti-UCHL1 (1:500, Abcam) overnight at 4°C, and then incubated with a biotinylated second antibody (goat anti-rabbit/IgG 1:100 dilution) for 30 min at 37°C.

Cells were treated with cyclohexamide (CHX, 100 μg/ml; Sigma) for 0, 3 or 6 h to inhibit further protein synthesis. Then cells were harvested and analyzed by western blot analysis.

For whole-cell protein extraction, cells were lysed with lysis buffer [50 mmol/l Tris-HCl, pH 7.4; 150 mmol/l NaCl; 1.5% NP-40; 0.1% SDS; 50 μg/ml phenylmethylsulfony fluoride (PMSF); with freshly added proteinase inhibitor cocktail] for 40 min on ice, followed by centrifugation at 14,000 × g for 15 min at 4°C. The supernatants were collected as total protein.

For preparation of cytoplasmic and nuclear protein, cells were collected and rinsed with cold phosphate-buffered saline (PBS) twice, then centrifuged at 3000 × g for 5 min at 4°C. The pellets were suspended in buffer A (10 mM HEPES, pH 7.9; 1.5 mM MgCl₂; 10 mM KCl; 0.5 mM DTT; 0.2% NP-40; 50 μg/ml PMSF; with freshly added proteinase inhibitor cocktail) and kept for 15 min at 4°C, then centrifuged at 5,000 × g for 5 min. The supernatants (cytoplasmic protein) were harvested. The insoluble pellets were then centrifuged at 14,000 × g for 1 min, again after being washed with cold buffer A. The pellets were lysed in lysis buffer and kept for 40 min at 4°C, then centrifuged at 14,000 × g for 10 min. The supernatants were harvested as nuclear protein. Western blot analysis was performed using conventional protocols as described previously (16).

All statistical analyses were performed using SPSS 13.0 software (SPSS Inc., Chicago, IL, USA). Data were expressed as means ± SD. The Student's t-test was used for statistical comparison. P-values <0.05 indicated statistical significance.

As we previously reported (16), HCT8 did not express endogenous UCHL1. Thus, we used HCT8 to generate the UCHL1 stable cell line. The HCT8/UCHL1 stable cells expressed high exogenous UCHL1, and the HCT8/pcDNA3.1 cells were used as empty vector control (Fig. 1A). HCT8/UCHL1 cells showed increased cell proliferation, migration, and invasion capabilities in vitro compared with control cells (P<0.05) (Fig. 1B–D). These results implied that UCHL1 promoted the proliferation, migration, and invasion capabilities of HCT8 cells in vitro.

To investigate UCHL1 functions in vivo, we injected HCT8/pcDNA3.1 and HCT8/UCHL1 cells into nude mice through the nail pad. Mice were humanely sacrificed and dissected for tumors 38 days later. Tumor nodules resected from the HCT8/UCHL1 group showed positive UCHL1 staining (Fig. 2D). In addition, the nodules from this group were larger and heavier than those from the control group (P<0.05) (Fig. 2A–C). However, we did not find lymph node metastasis or lung metastasis in either group in this model.

To investigate whether UCHL1 promotes tumor metastasis in vivo, we injected HCT8/pcDNA3.1 and HCT8/UCHL1 cells into nude mice via tail veins. Mice were sacrificed 4 weeks later. In the HCT8/UCHL1 group, 2 mice died before the end of the experiment. In the remaining 4 mice, all of them developed lymph node metastases, accompanied with positive UCHL1 staining (Fig. 3A). However, no lymph node metastasis was found in HCT8/pcDNA3.1 group. The percentage of metastatic lymph nodes with respect to the total visible lymph nodes per mouse was significantly higher in the HCT8/UCHL1 group (22%) than in the control group (0%; P<0.05) (Fig. 3B). Furthermore, UCHL1 re-expression dramatically promoted lung metastasis of HCT8 cells. The metastatic loci in the HCT8/UCHL1 group (58±30) were significantly more than those in the control group (7±6; P<0.05) (Fig. 3C–D). These results indicated that UCHL1 promoted tumorigenesis and metastasis of HCT8 cells in vivo.

To investigate if the deubiquitinating capability is essential for UCHL1 function, we constructed a C90S mutant (with cysteine 90 converted to serine), with impaired deubiquitinase activity (17). Then the HCT8/UCHL1-C90S stable cell line was generated (Fig. 4A). The proliferation, migration, and invasion capabilities of HCT8/UCHL1-C90S cells were similar to those of HCT8/pcDNA3.1 cells (Fig. 4B–D). These results indicated that the malignant transformation promoting function of UCHL1 was dependent on its deubiquitinating activity.

To see whether UCHL1 could affect β-catenin expression in HCT8 cells, we detected β-catenin in three stable cell lines. Our results confirmed that β-catenin expression was upregulated by UCHL1, but could not be upregulated by the UCHL1-C90S mutant (Fig. 5A).

To determine whether UCHL1 upregulates β-catenin expression by enhancing its stability, a half-life assay was performed. The data showed that UCHL1 re-expression significantly extended the half-life of β-catenin. In contrast, we failed to observe this function in the UCHL1-C90S mutant (Fig. 5B). These results indicated that UCHL1 upregulated β-catenin by enhancing its stability, depending on its deubiquitinating activity.

To further confirm the effect of UCHL1 on β-catenin/TCF signaling, the immunoblotting analyses were carried out using the protein isolated from the cell cytoplasm and nucleus in three stable cell lines. Our results demonstrated that UCHL1 re-expression increased the accumulation of β-catenin in cytoplasm, which resulted in increased β-catenin translocation into cell nucleus (Fig. 6A). However, β-catenin remained unchanged in HCT8/UCHL1-C90S cells compared with HCT8/pcDNA 3.1 cells.

Cyclin D1 and serine protease urokinase plasminogen activator (uPA) are transcriptionally induced by β-catenin in the nucleus (18,19). As shown in Fig. 6B, both molecules were upregulated in HCT8/UCHL1 cells, but not in HCT8/pcDNA 3.1 and HCT8/UCHL1-C90S cells. These data suggested that UCHL1 activated the β-catenin/TCF signaling, depending on its deubiquitinating activity.