A total of 89 CRC patients undergoing surgical resection between January 1, 2004 and October 1, 2010, in Guangzhou First People’s Hospital, Guangzhou Medical College, were included. One hundred and six cancerous samples, 35 adjacent normal, 39 liver metastatic samples from 89 CRC patients were collected. The median age of the patients was 54 years at operation, range 28–81 years. Follow-up and clinicopathological characteristics including tumor location, stage, and differentiation were recorded. OS was calculated from the date of surgery until the date of last contact. This study was approved by the Ethics Committee of the Guangzhou First People’s Hospital, Guangzhou Medical College.

Surgical samples containing primary tumors, adjacent normal tissues and liver metastatic tumors were collected in 10% buffered formalin. The details of deparaffinization and IHC were previously described (20). Briefly, following deparaffinization, the endogenous peroxidase activity was blocked with 3% H₂O₂. The array slides were incubated with normal goat serum for 20 min, and then applied with primary antibody for 20 min at room temperature. After 7 min of H₂O₂ treatment, the array slides were incubated with horseradish peroxidase-labeled polymer conjugated with corresponding antibodies for 30 min. Each slide was counterstained with hematoxylin (Dako, Carpinteria, CA, USA). PBS was used as a negative control.

The rabbit anti-human GATA6 monoclonal antibody (Sigma-Aldrich, USA) was used for IHC staining (1:100 dilution). To reduce the image reader bias, an automated imaging system was employed to obtain digital images of the stained sections for subsequent quantitative analyses. In addition, each sample was evaluated by two independent investigators in a double-blind manner. Investigators reviewed and assessed the subcellular localization (cytoplasm vs. nucleus), staining intensity (integrated optical density), and/or percentage of stained cells (total area or percentage of cells positive) for each image. Discrepancies in samples were resolved after joint review by the readers.

The two CRC cell lines (Colo-205 and SW-480 cells) were obtained from the American Type Culture Collection and maintained in RPMI-1640 with 10% calf serum (HyClone) at 37°C with 5% CO₂.

Cells were transfected with lentiviral vectors encoding short hairpin RNA targeting human GATA6 for GATA6 knockdown (shGATA6) or a scrambled shRNA as control (shControl) (Sigma-Aldrich). Multiplicity of infection was 10. Cells were cultured for 72 h after transfection. Cells were grown to 80% confluency in 60-mm dishes. GATA6 (10 lg) plasmid (ExGATA6) or empty plasmid (ExControl) was transfected using Lipofectamine Reagent (Invitrogen). Cells were cultured for 48 h after transfection.

Cells (2×10⁵) were suspended in 400 μl serum-free RPMI-1640 medium and seeded in the top chamber that had been coated with a layer of extracellular matrix (BD Biosciences, USA). The complete medium with serum (500 μl) was added to the bottom chamber. After 48 h of incubation, the cells which had invaded through the extracellular matrix layer to the lower surface of the filters were stained. Images of three randomly selected fields of the fixed cells were captured, and cells were counted. Experiments were repeated independently three times.

Cells were seeded in a 6-well plate, grown until confluence and then starved for 24 h. A linear wound was made by scraping a pipette tip. The cell motility in terms of wound closure was measured by photographing at three random fields 72 h after wounding. Experiments were repeated independently three times.

Total RNA was extracted from CRC cells using RNA Extraction kit (Qiagen, China). The cDNA synthesis was performed according to the manufacturer’s instructions (SYBR-Green PCR kit; Qiagen). Quantitative PCR was performed by SYBR-Green PCR kit (Qiagen) using a LightCycler system. PCR reaction conditions for all assays were 94°C for 30 sec, followed by 40 cycles of amplification (94°C for 5 sec, 58°C for 30 sec and 72°C for 30 sec). β-actin mRNA was used to normalize RNA. Primer sequences were: GATA6 forward, CCAACTTCCACCTCTTCTAAC and reverse, TTGACCCGAATACTTGAGC; β-actin forward, CCATGTACGTTGCTATCCAGG and reverse, TCTCCTT AATGTCACGCACGA.

Nuclear protein or total proteins of CRC cells were isolated using RIPA (Cell Signaling Technology). For immunoblotting, equal amounts of proteins were separated on 5–8% SDS-PAGE and were electrophoretically transferred onto nitrocellulose membranes (Millipore), which were blocked in TBST containing 5% milk for 2 h at room temperature and blotted with antibody overnight at 4°C: anti-GATA6 (1:1,000; Abcam) and anti-histone H3 (1:500; Cell Signaling Technology). After washing with TBST and incubating with either anti-rabbit or anti-mouse horseradish peroxidase-conjugated secondary antibody (Cell Signaling Technology) for 2 h at room temperature, immunocomplexes were visualized using the chemiluminescence (GE Healthcare Life Sciences, USA) following the manufacturer’s protocol.

Data were analyzed using SPSS 17.0. Continuous data were measured by the t-test. For categorical data, Chi-square analysis or Fisher’s exact test was used. Multivariate logistic regression was used to adjust for covariate effects on the odds ratio (OR). Kaplan-Meier was applied for OS. P<0.05 was considered to indicate a statistically significant difference.

We examined GATA6 protein expression in metastatic CRC by IHC. In general, GATA6 was predominantly nuclear staining in IHC. Some perinuclear granulation in cytoplasm was also observed. GATA6 expression was quantified using a visual grading system based on the extent of staining. Only immunoreactivity in the nucleus was evaluated. GATA6 was defined as: negative nuclear staining (−), <5%; positive (+), ≥5% and <50% or strong positive (++), >50% positive nuclear staining from CRC cells. The GATA6 IHC standards of negative (−), weak positive (+) and strong positive (++) are displayed in Fig. 1A–C, respectively. GATA6 expression was predominantly observed in the primary cancer cells, but not in the adjacent normal colorectal epithelium (Fig. 1D–F). Meanwhile, strong positive GATA6 was also observed in metastatic CRC to liver (Fig. 1G–I).

Based on the IHC staining of GATA6, 32 of 89 CRC samples were defined as GATA6 nuclear positive staining [including weak positive (+) and strong positive (++)]. The TNM stage of CRC was based on clinical diagnosis. According to the univariate analysis results, the GATA6 staining was positively and significantly associated with hepatic metastasis (P<0.01) and tumor invasion (P<0.05) of CRC, but not with age, gender, tumor location and lymph node involvement (Table I).

To validate this finding, non-conditional logistic analysis was employed for univariate and multivariate analyses. The GATA6 positive and negative were stratified as unfavorable and favorable subsets, respectively. Tumor invasion, lymph node involvement, and hepatic metastasis were considered as the endpoint in logistic analysis. The expression of GATA6 significantly impacted the risk of hepatic metastasis but not tumor invasion and lymph node involvement according to either univariate or multivariate analysis. After adjusting for age and gender, the OR of GATA6-positive for hepatic metastasis was 3.53 (95% CI, 1.37–9.70) (Table II). Therefore, our analyses revealed that GATA6 significantly affected hepatic metastasis of CRC, suggesting that GATA6 may be used to prognosticate CRC.

The prognostic significance of GATA6 was determined by GATA6 staining and the corresponding clinical follow-up records. Kaplan-Meier survival analysis revealed a correlation between higher GATA6 expression levels and shortened OS times (Fig. 2). The Log-rank test indicated that the positive GATA6 expression was significantly related to OS (P<0.05). The strong positive GATA6 (++) exhibited a more significantly reduced survivability (P<0.001). Taken together, these observations indicate that overexpression of GATA6 is significantly associated with CRC liver metastasis and poor prognosis in patients with CRC.

As clinical data showed positive correlation between GATA6 and metastatic behavior in CRC, the roles of GATA6 in invasion and metastasis were investigated in two human CRC cell lines, Colo-205 and SW-480 cells. Both mRNA and nuclear protein levels of GATA6 were significantly higher in SW-480 cells than in Colo-205 cells (Fig. 3A and B), and therefore, knockdown of GATA6 was performed in SW-480 cells by GATA6-shRNA lentivirus transfection and overexpression of GATA6 was performed in Colo-205 cells by GATA6 plasmid transfection, respectively. The GATA6 nuclear protein levels after transfection were also validated (Fig. 3C). By transwell assay, the number of invading cells was markedly reduced by knockdown of GATA6 in SW-480 cells and was markedly increased by overexpression of GATA6 in Colo-205 cells (Fig. 4A). Consistently, cell migration was significantly reduced by GATA6 knockdown and enhanced by GATA6 overexpression by wound-healing assay (Fig. 4B). Collectively, the data suggested a role of GATA6 in promoting CRC cell invasion and metastasis.