A total of 240 cases of paraffin-embedded HCC samples and their matched non-HCC liver tissues which had been clinically and histologically diagnosed at the Shandong Provincial Hospital Affiliated to Shandong University (Shandong, China) from 2000 to 2004 were used. The CCR9 expression from 240 HCC samples was subjected to immunohistochemistry (IHC). All participants underwent hepatectomy with a median age of 48 years (range 29–78 years). In addition, fresh specimens with vascular invasion (n=39) and without vascular invasion (n=26) were also analyzed by quantitative PCR (qPCR) and western blotting. Prior written consent from patients and approval by the Institutional Ethics Committee were obtained.

Follow-up data were obtained by reviewing the medical records and from direct communication with patients. Follow-up visits were scheduled postoperatively at intervals of one month for six months, bimonthly for six months, quarterly for six months, and semiannually for life. Complete follow-up, ranging from 0 to 83 months, was available for all patients and the median survival was 34 months.

Immunohistochemical analysis was carried out to study CCR9 protein expression in 240 human HCC tissues. In brief, paraffin-embedded specimens were cut into 4-μm sections and baked at 65°C for 30 min. The sections were deparaffinized with xylenes and rehydrated. Sections were submerged into EDTA antigenic retrieval buffer and microwaved for antigenic retrieval. The sections were treated with 3% hydrogen peroxide in methanol to quench the endogenous peroxidase activity, followed by incubation with 1% bovine serum albumin to block nonspecific binding. Rabbit anti-CCR9 (1:50; LifeSpan Biosciences) was incubated with the sections overnight at 4°C. For negative controls, the rabbit anti-CCR9 antibody was replaced with normal goat serum, or the rabbit anti-CCR9 antibody was blocked with a recombinant CCR9 polypeptide by co-incubation at 4°C overnight preceding the immunohistochemical staining procedure. After washing, the tissue sections were treated with biotinylated anti-rabbit secondary antibody, followed by further incubation with streptavidin-horseradish peroxidase complex (both from LifeSpan Biosciences). The tissue sections were immersed in 3-amino-9-ethylcarbazole and counterstained with 10% Mayer’s hematoxylin, dehydrated and mounted in crystal mount.

The degree of immunostaining of formalin-fixed, paraffin-embedded sections was reviewed and scored independently by two observers, based on both the proportion of positively stained tumor cells and the intensity of staining (12,13). The proportion of tumor cells was scored as follows: 0 (no positive tumor cells), 1 (<10% positive tumor cells), 2 (10–50% positive tumor cells) and 3 (>50% positive tumor cells). The intensity of staining was graded according to the following criteria: 0 (no staining), 1 (weak staining = light yellow), 2 (moderate staining = yellow brown) and 3 (strong staining = brown). The staining index was calculated as staining intensity score x proportion of positive tumor cells. Using this method of assessment, we evaluated the expression of CCR9 by determining the staining index, which scores as 0–4, 6 and 9. Cut-off values for CCR9 were chosen on the basis of a measure of heterogeneity with the log-rank test statistical analysis with respect to overall survival (OS). An optimal cut-off value was identified: the staining index score of ≥4 was used to define tumors as high CCR9 expression and ≤3 as low expression of CCR9.

Total RNA from cells and primary tumor materials was extracted using the TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. The extracted RNA (5 μg) from each sample was used for cDNA synthesis primed with random hexamers. Real-time PCR primers were designed using the Primer Express v 2.0 software (Applied Biosystems). The primers were: CCR9 forward, 5′-GTGCCTCCCT GAGATCATGT-3′ and reverse, 5′-TGTGCTTTTGGCATCT TTTG-3′; β-actin forward, 5′-GCTGTATTCCCCTCCATC GT-3′ and reverse, 5′-GCCATGTTCAATGGGGTACT-5′. For PCR amplification of CCR9 cDNA, an initial amplification using CCR9-specific primers was performed with a denaturation step at 95°C for 10 min, followed by 35 cycles of denaturation at 95°C for 60 sec, primer annealing at 58°C for 30 sec and primer extension at 72°C for 30 sec. Upon completion of the cycling steps, a final extension at 72°C for 5 min was performed before the reaction was stored at 4°C. Expression data were normalized to the geometric mean of housekeeping gene β-actin to control the variability in expression levels. The results were analyzed using the ΔΔCt method.

Total protein was extracted and determined by the Bradford assay using a commercial kit purchased from the Bio-Rad Laboratories. Equal quantities of protein were separated electrophoretically on 10% SDS/polyacrylamide gels and transferred onto polyvinylidene difluoride membranes (Roche). The membrane was probed with a 1:500-diluted anti-CCR9 antibody (Abcam). Expression of CCR9 was determined with horseradish peroxidase-conjugated anti-rabbit IgG (1:2,000) and enhanced chemiluminescence (Pierce) according to the manufacturer’s suggested protocols. The membranes were stripped and reprobed with an anti-β-actin antibody (1:2,000; Sigma) as a loading control.

The cell lines HepG2, Huh7, HEP3B and HCCLM3 were purchased from the American Type Culture Collection (ATCC) and were grown in DMEM (Invitrogen) supplemented with 10% fetal bovine serum (FBS) (HyClone). Overexpression of CCR9 human full-length CCR9 cDNA was amplified by PCR and cloned into a pMSCV-puro retroviral vector by OriGene Technologies (Rockville, MD, USA). For CCR9 knockdown, shRNA sequences were cloned into the pSUPER-retro-puro plasmid to generate pSUPER-retro CCR9 shRNA obtained from OriGene Technologies.

The MTT assay was used to assess the cell proliferation. Briefly, cells were seeded in 96-well plates at a density of 2×10³ cells/well. One plate was taken out at the same time every day after the cells adhered. Twenty microliters of MTT (5 mg/ml) were added to each well, and the cells were incubated for another 4 h. The medium was removed and the formazan precipitate was solubilized in 150 ml dimethyl sulfoxide. The absorbance at 490 nm was measured using a microplate reader. All experiments were performed in triplicate.

Regarding the colony formation, cells were seeded in 6-well plates (2×10³ cells/plate) and cultured for 10 days. Subsequently, cells were fixed with ice-cold methanol for 10 min, followed by staining with 1% crystal violet for 1 min.

HCC cells were seeded on coverslips at a density of 5×10⁴ cells and 24 h post-seeding, the cells were incubated with BrdU for 1 h, followed by staining with anti-BrdU antibody (Sigma) according to the manufacturer’s instructions (Roche Diagnostics). Images were captured using a laser scanning microscope (Olympus).

The anchorage-independent growth ability was determined in soft agar, as previously described (14). Briefly, 3 ml of 0.5% agar in basal modified Eagle’s medium supplemented with 10% FBS was layered onto each well of 6-well tissue culture plates. Cells (3×10⁴ cells) suspended in 1 ml normal medium were mixed with 2 ml of 0.5% agar-basal modified Eagle’s medium supplemented with 10% FBS, and 1 ml of mixture was added into each well over the top of the 0.5% agar layer. Plates were incubated at 37°C in 5% CO₂ for 2 weeks, and the colonies with >32 cells of each were counted.

All statistical analyses were carried out using the SPSS 17.0 statistical software package. The Mann-Whitney U test was used to analyze the correlation between CCR9 expression and the clinicopathological characteristics. Survival curves were plotted using the Kaplan-Meier method and compared with the log-rank test. The significance of various variables for survival was analyzed by the Cox proportional hazards model in the univariate and multivariate analyses. P<0.05 was considered to indicate a statistically significant difference.

CCR9 expression was detected mainly in the cytoplasm of tumor cells (Fig. 1). According to the classification described in the Materials and methods, immunohistochemical analysis clearly showed that high CCR9 expression was present in 55.8% (134 of 240) of HCC tissues. In the non-HCC tissues, we observed high CCR9 protein expression only in 9.6% (23 of 240) (P<0.01). These results indicated that CCR9 expression was significantly elevated in HCC samples.

Immunohistochemical analysis showed that the high CCR9 levels correlated with tumor node number, high Edmondson-Steiner grade and vascular invasion (Table I). As indicated in the Materials and methods, 240 HCC tissues were divided into the group with high CCR9 expression (n=134) and the group with low CCR9 expression (n=106). As a result, HCC patients with high CCR9 expression had markedly reduced OS (P=0.0041; Fig. 1E) compared with patients with low CCR9 expression. Multivariate analysis showed that CCR9 was an independent prognostic factor for the OS of HCC patients (Table II, hazard ratio=2.35, 95% confidence interval=1.09–5.02, P=0.014). These findings support that CCR9 plays an important role in HCC.

We also investigated the relationship between CCR9 expression and vascular invasion in fresh HCC tissues using qPCR and western blotting. The results showed that CCR9 mRNA level in HCC tissues with vascular invasion was significantly higher than in tissues without vascular invasion (P<0.01, Fig. 2A). Consistent with the mRNA expression, the CCR9 protein expression level was also significantly elevated in HCC tissues with vascular invasion (P<0.01; Fig. 2B).

Furthermore, we detected the expression levels of CCR9 in HCC cell lines. Among the four HCC cell lines, HCCLM3, with the highly metastatic ability (15), had the highest CCR9 expression at mRNA and protein levels, followed by HEP3B, Huh7 and HepG2 cells (Fig. 2C and D). These data suggest that there may be a correlation between CCR9 and the metastasis potential of HCC.

To further investigate the potential role of CCR9 in HCC, we assessed the effects of CCR9 silencing on the cell proliferation of HCC cell lines. Based on the CCR9 expression levels, we chose HEP3B and HCCLM3 as the study cells.

Western blot analysis showed that CCR9 expression in the two cell lines were knocked down (Fig. 3A). MTT assays showed significant reduction in cell proliferation of HEP3B-shCCR9 and HCCLM3-shCCR9 cells compared with control cells at 4 and 5 days (Fig. 3B). Silencing of CCR9 also reduced the colony formation abilities (Fig. 3C) in both cell lines. Herein, we showed that CCR9 silencing inhibited the proliferation of HCCLM3 and HEP3B cells compared with control cells.

Furthermore, we investigated the potential role of CCR9 overexpression in HCC cell lines. Western blot analysis showed that stable CCR9-overexpressed cell lines were established (Fig. 4A). MTT assays showed a significant increase in cell proliferation of HEP3B^CCR9+ cells and HCCLM3^CCR9+ cells compared with control cells at 4 and 5 days (Fig. 4B). Overexpression of CCR9 also increased the colony formation abilities (Fig. 4C) in the two cell lines. Therefore, we showed that CCR9 overexpression promoted the proliferation of HEP3B and HCCLM3 cells compared with control cells.

Since CCR9 promoted cell proliferation, we further explored the potential mechanisms. The percentage of cells at S phase was assessed by immunodetection of BrdU. Ectopic expression of CCR9 markedly increased the percentage of BrdU-positive cells in HCC cell lines; 29.4 vs. 43.5% for HEP3B and 27.2 vs. 41.8% for HCCLM3 (Fig. 5A). Conversely, CCR9 silencing significantly decreased the S phase fraction of BrdU-positive cells; 32.27 vs. 17.03% for HEP3B cells and 31.23 vs. 17.98% for HCCLM3 cells (Fig. 5B). Flow cytometric analysis of cell cycle also confirmed that CCR9 overexpression markedly increased the percentage of cells at S phase and decreased the percentage of cells in the G1/G0 phase (Fig. 5C). In contrast, CCR9 knockdown increased the percentage of cells in the G1/G0 phase and decreased the percentage of cells at S phase (Fig. 5D). These results showed that CCR9 promoted cell proliferation by increasing the fraction of HCC cells at S phase.

The p21, p27 and cyclin D1 play key roles in the control of cell cycle associated with tumorigenicity (16,17). qPCR showed that CCR9 overexpression significantly reduced p21 and p27, and increased cyclin D1 (Fig. 6A). In contrast, CCR9 silencing markedly enhanced p21 and p27 expression, and inhibited cyclin D1 expression in both cell lines (Fig. 6B). These results indicated that CCR9 regulates p21, p27 and cyclin D1 to promote cell proliferation and tumorigenicity.