KLF4-Mediated Negative Regulation of IFITM3 Expression Plays a Critical Role in Colon Cancer Pathogenesis

Colon cancer is one of the most common cancers worldwide (1). For metastatic colon cancer, the 5-year overall survival rate is about 10% (2). A mechanistic understanding of colon cancer initiation, recurrence, and metastasis is therefore an important goal. Genome-wide gene-expression analysis of colon tumor and corresponding normal colon tissue cells is crucial to understanding how oncogenes and tumor suppressor genes alter the complex cellular molecular context of colon cancer and thus drive colon tumor progression (3).

In previous experimental and bioinformatic genome-wide gene-expression analyses, we found that expression of the IFITM3 gene was aberrantly higher in colon tumors than in matched normal colon mucosa (4). Researchers first isolated IFITM3 from a genetic screen aimed at identifying the genes involved in the acquisition of germ-cell competence (5). Investigators proposed that epiblast cells having the highest levels of expression of IFITM3 initiated germ-cell specification and that homotypic association discriminated germ cells from their somatic neighbors. Studies then showed that IFITM3 belongs to a family of 5 murine genes (along with IFITM1, IFITM2, IFITM5, and IFITM6) clustered within a 68-kb genomic region on chromosome 7 (6). The IFITMs are short, 2-transmembrane-domain proteins (5–18 kDa) with high core sequence similarity but divergent N- and C-termini. The human homologues (IFITM1, IFITM2, and IFITM3) are clustered on chromosome 11 within an 18-kb genomic sequence (7–9). These genes respond to treatment with type I and II IFNs and encode for IFN-induced transmembrane proteins involved in the homotypic cell adhesion functions of IFNs (8, 10).

Expression of the IFITM3 gene is implicated in diverse cellular processes, including immune-cell regulation, somitogenesis, germ-cell homing and maturation, and bone mineralization (10–13). Recent studies identified possible roles for IFITM genes involved in carcinogenesis. For example, researchers first isolated the IFITM3 gene from tumor tissue and severely inflamed mucosa in the colons of patients with ulcerative colitis, describing it as a preferential marker for ulcerative colitis–associated colon cancer (14, 15). Another study showed that IFITM1 and IFITM3 were expressed at higher levels in astrocytoma cells than in normal astrocytes in mice (16). Furthermore, authors reported that upregulation of IFITM1 expression played a critical role in both the precancerous stage and carcinogenesis in patients with gastric mucosa infected with Helicobacter pylori and cervical cancer (17–19). Also, overexpression of IFITM1 promotes and knockdown of IFITM1 expression significantly suppresses the invasiveness of head and neck tumor cells (20). However, the precise function and underlying mechanisms of IFITM3 in colon cancer pathogenesis remain unclear.

Furthermore, the expression and regulation of IFITM1, IFITM2, and IFITM3 are not clear. One recent study identified IFITMs as potential targets of Wnt/β-catenin signaling in carcinogenesis in the intestinal epithelium (21). Importantly, KLF4, a critical tumor repressor in the initiation and progression of gastrointestinal cancers (22, 23), directly interacts with Wnt/β-catenin signaling. However, whether KLF4 critically regulates IFITM expression is unknown. In the present study, we sought to determine the role of IFITM3 and molecular mechanisms underlying dysregulated expression of it in colon cancer progression.

For tissue microarray (TMA) construction and immunohistochemical analysis, we used human colon tumor specimens obtained from 203 patients with colon cancer that were preserved in the Colon Cancer Tissue Bank at Shanghai Jiaotong University Affiliated First People's Hospital (Shanghai, PR China). The primary colon cancer in these patients was diagnosed (and later confirmed by at least 2 pathologists) and the patients were accepted for colectomy at Affiliated First People's Hospital from 2001 to 2003. The 203 formalin-fixed, paraffin-embedded specimens were selected to represent all of the stages and histologic types of colon cancer. Tumor staging for the specimens was carried out according to the American Joint Committee on Cancer staging criteria (24). The patients' disease-free survival (DFS) and overall survival (OS) durations were defined as the interval from initial surgery to clinically or radiologically proven recurrence or metastasis and from initial surgery to death, respectively. The follow-up period for this analysis concluded on June 29, 2008. The use of human specimens was approved by proper institutional review boards.

After screening hematoxylin and eosin (H&E)-stained slides for optimal tumor tissue and tumor adjacent tissue up to 2 cm from the tumor, TMA slides were constructed (in collaboration with Shanghai Biochip). Two cores were taken from each formalin-fixed, paraffin-embedded tumor and matched normal colon tissue specimen as well as at least 1 lymph node metastasis core by using punch cores that measured 2 mm in greatest dimension from the non-necrotic areas of tumor, lymph node metastasis, and matched normal colon tissue specimens. Among the 203 study patients, there were 66 from whom primary colon tumor and matched lymph node metastasis specimens were obtained. These specimens were included in the TMA. Sections (4 μm thick) of formalin-fixed, paraffin-embedded colon tumor specimens were prepared and processed for immunostaining by using a rabbit polyclonal antibody against human KLF4 [H-180 (1:200 dilution); Santa Cruz Biotechnology], a monoclonal antibody against IFITM3 [H00010410-M01 (1:1,000 dilution); Abnova], and a rabbit polyclonal antibody against human proliferating cell nuclear antigen [Ki67 (1:200 dilution); Santa Cruz Biotechnology].

The human colon cancer cell lines HCT116 and SW480 were purchased from the American Type Culture Collection. All cells were maintained at 37°C in 5% CO₂ in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% FBS.

Total RNA was isolated from cell cultures or tissues using TRIzol reagent (Invitrogen). First-strand cDNA was synthesized from 1 μg of total RNA by using Moloney murine leukemia virus reverse transcriptase (Invitrogen). Real-time transcriptase-PCR analysis of expression of the IFITM3 gene was carried out using 2 μL of cDNA and the SYBR Green Master Mix (Bio-Rad) as recommended by the manufacturer for the IFITM3 primers 5′-caaggaggagcacgagg-3′ (forward) and 5′-ttgaacagggaccagacg-3′ (reverse). Glyceraldehyde-3-phosphate dehydrogenase was used as an internal control. Each PCR product was run in triplicate for the target and internal control genes.

Whole-cell lysates were prepared from the colon cells as described previously (25). Standard Western blot analysis of the lysates was carried out with an antibody against KLF4 (Santa Cruz Biotechnology) or against IFITM3 (Abnova) and a second anti-IgG antibody (Amersham Life Sciences). The membranes were then stripped and blotted with an anti-β-actin antibody (Sigma Chemical Co.) and used as loading controls. The probe proteins were detected using an enhanced chemiluminescence system (Amersham Life Sciences) according to the manufacturer's instructions.

To induce overexpression of KLF4 in HCT116 and SW480 cells, the cells were transfected with adenovirally infected KLF4 (Ad-KLF4) or enhanced green fluorescent protein [EGFP (Ad-EGFP)] as described previously (26). To inhibit IFITM3 expression in these cells, they were transfected with a pool of IFITM3 siRNA oligonucleotides (Santa Cruz Biotechnology; 50 nmol/L) or control siRNA oligonucleotides (Santa Cruz Biotechnology; 50 nmol/L). Also, the cells treated with Oligofectamine reagent alone were included as mock controls.

For an MTT assay, transfected cells (3 × 10³) were seeded in a 96-well culture plate and subsequently incubated with MTT reagent (0.5 mg/mL) at 37°C for 2 hours and then with dimethyl sulfoxide at 37°C for 1 hour. The MTT assay was carried out at 24, 48, and 72 hours after transfection. The results were plotted as the mean ± SD from 3 separate experiments with 8 replicates per experiment for each experimental condition.

Colon cancer cells were suspended in 0.3% agar medium (DMEM containing 10% FBS) and then plated on a 0.6% agar base layer at a concentration of 1.5 × 10³ cells per 6-well plate. The cells were incubated in a humidified atmosphere (5% CO₂) at 37°C. The colonies that were 50 μm in diameter or larger were counted 10 days after cell plating.

Cell migration and invasion assays were conducted using a modified 24-well Boyden chamber with a membrane that was uncoated or coated with Matrigel (BD Biosciences), respectively. Briefly, 24 hours after transfection of both HCT116 and SW480 cells either with a control (mock or control siRNA-treated) or IFITM3 siRNA, the cells were harvested and suspended in DMEM at a concentration of 8 × 10⁴/mL. Cells prepared in 500 μL of DMEM were loaded in the upper wells, and a medium containing 20% FBS was placed in the lower wells as a chemoattractant stimulus. Migrated cells on the bottom surface of the filter were fixed, stained with H&E, and counted under a microscope in 3 randomly selected fields at a magnification of 200×.

HCT116 and SW480 cells transfected with Ad-KLF4 or Ad-EGFP were cultured on Falcon chamber slides (BD) at up to 50% to 60% confluence before being fixed with 4% paraformaldehyde and permeabilized with 0.3% Triton X-100. The cells were then immersed 3 times in PBS, incubated with indicated primary antibodies overnight at 4°C and corresponding Alexa Fluor–conjugated secondary antibodies (Invitrogen) for 1 hour at room temperature, and mounted using a mounting medium containing 4′,6-diamidino-2-phenylindole. Microscopic images of cells were obtained using an Axio Observer A microscope (Zeiss).

A 0.5-kb IFITM3 promoter was cloned into pGL3 basic luciferase reporter vectors. Site-specific mutagenesis of the IFITM3 promoter was carried out using a QuikChange Site-Directed Mutagenesis kit (Stratagene) according to the manufacturer's instructions. The primers used to generate the mutant vector were as follows: site 1, 5′-gggcctggagtgtggaatcctcagcgcaggcctg-3′ (sense) and 5′-caggcctgcgctgaggattccaca ctccaggccc-3′ (antisense); site 2, 5′-cagctgctggcctggattctggagacccccaaca-3′ (sense) and 5′-tgttgggggtctccagaatccaggccagcagctg-3′ (antisense). The mutation was confirmed via DNA sequencing. The IFITM3 promoter activity was normalized by cotransfection with a β-actin/Renilla luciferase reporter containing a full-length Renilla luciferase gene (25). We quantified both firefly and Renilla luciferase activity using a dual-luciferase assay system (Promega) 24 hours after transfection.

HCT116 and SW480 cells were seeded to about 80% confluence in 15-cm culture dishes and transfected with Ad-KLF4 or Ad-EGFP. Twenty-four hours later, a chromatin immunoprecipitation (ChIP) assay was carried out using a ChIP assay kit (Cell Signaling Technology) according to the manufacturer's instructions. Briefly, DNA cross-binding proteins were cross-linked with DNA and lysed in SDS lysis buffer. The lysate was sonicated to shear DNA to around 500 bp. Anti-KLF4 antibodies were then added, and the samples were incubated overnight at 4°C. Immunoprecipitation using a normal rabbit IgG as a negative control was carried out as well. Resulting immunocomplexes were precipitated for 2 hours with Protein A-Sepharose beads, and DNA was recovered using phenol–chloroform extraction. Next, the DNA was subjected to PCR analysis to amplify a 512-bp region (−438 to +74 bp) of the IFITM3 promoter using the primers 5′-tgagggttatgggagacg-3′ and 5′-gggttactgggatggttc-3′. The PCR products were resolved electrophoretically on a 2% agarose gel and visualized using ethidium bromide staining.

For orthotopic model of colon cancer, female athymic BALB/c nude mice were purchased from The Jackson Laboratory. The mice were housed in laminar flow cabinets under specific pathogen-free conditions and used when they were 8 weeks old. To produce cecal tumors, 5 × 10⁵ cells in 50 μL of Hank's balanced salt solution were injected into the cecum wall of nude mice. Mice were sacrificed 35 days after tumor implantation. Cecal tumors were weighted and hepatic metastases were determined as described previously (27). For colon tissue–specific KLF4 knockout mice, the derivation and use of Klf4-LoxP and Villin-Cre mice were described previously (28). These Klf4-LoxP mice had LoxP in the introns flanking exons 2 and 3. The mice had a C57BL6 genetic background. These animals were mated with Villin-Cre mice, which were obtained from The Jackson Laboratory [004586; B6.SJL-Tg(Vil-Cre)997Gum/J], which have expression of Cre under the control of a 12.4-kb regulatory region of the murine Villin promoter. Tg(Vil-Cre) mice also had a C57BL6 genetic background. Pairs of mice heterozygous for the floxed Klf4 allele and positive for Villin-Cre were intercrossed to produce offspring homozygous for the floxed allele and containing the Villin-Cre transgene and Klf4^fl/fl (control) offspring. The animals were housed and maintained in facilities approved by the Association for Assessment and Accreditation of Laboratory Animal Care International in accordance with the current regulations and standards of the U.S. Department of Agriculture, Department of Health and Human Services, and NIH.

The 2-tailed χ² test was used to determine the significance of the difference among the covariates. Survival durations were calculated using the Kaplan–Meier method. The log-rank test was used to compare the cumulative survival rates in the patient groups. A Cox proportional hazards model was used to calculate univariate and multivariate HRs for the study variables. The IFITM3 expression level, patient age, disease stage (American Joint Committee on Cancer staging system), and tumor differentiation and distant metastasis were included in the model. The significance of the in vitro data was determined using the Student t test (2-tailed). In all of the tests, the values of P < 0.05 were considered statistically significant. The SPSS software program (version 12.0; SPSS Inc.) was used for statistical analyses.

We first investigated the expression of IFITM3 protein in the 203 primary colon tumor and paired adjacent normal colon mucosa specimens as well as the 66 lymph node metastasis specimens in a TMA. We observed IFITM3-positive staining in the cytoplasm of the cancer cells with IFITM3-negative or weak IFITM3-positive staining in adjacent normal colon cells (Fig. 1A, Supplementary Table S1). We confirmed the presence of aberrant upregulation of IFITM3 expression at both the mRNA and protein levels in colon tumor and normal colon tissue specimens by using real-time PCR and Western blotting (Fig. 1B).

We further analyzed the relationship between clinicopathologic features and IFITM3 expression levels in colon cancer cases. IFITM3 expression (Supplementary Table S2) was positively correlated with disease stage (P < 0.001), pT classification (P < 0.001), regional lymph node metastasis (P = 0.004), and distant metastasis (P = 0.009). In contrast, IFITM3 expression was inversely correlated with OS and DFS rates (Fig. 1C, Supplementary Tables S3 and S4). Moreover, in the subgroups of stage III and IV patients who underwent standard postoperative chemotherapy (Supplementary Table S5), there were significant differences in OS and PFS between IFITM3-negative and IFITM3-positive subgroups (Supplementary Tables S3 and S4). These findings strongly indicated that IFITM3 expression plays a critical role in colon cancer development and progression and is a valuable biomarker for this disease.

Prior studies have shown that loss of KLF4 expression contributed to colon carcinogenesis, whereas the underlying mechanisms remain to explore (22, 23). It is unknown whether loss of KLF4 in fact is, in part, responsible for IFITM3 overexpression. To identify the molecular mechanisms underlying IFITM3 overexpression and its impact on colon cancer pathogenesis, we carried out immunostaining for KLF4 and Ki67 in normal colon tissue, colon tumor, and nodal metastasis specimens (Fig. 2A, Supplementary Tables S6 and S7). The levels of IFITM3 expression in the primary tumors were higher than those in the normal colon tissue, whereas the levels of IFITM3 expression in the nodal metastases were higher than those in the primary tumor and normal tissue specimens (Fig. 2B, left). We observed a significant inverse correlation between the levels of IFITM3 and KLF4 expression (Fig. 2B, middle), whereas the level of IFITM3 expression correlated directly with that of Ki67 expression (Fig. 2B, right; Fig. 3A). These results indicated that IFITM3 expression was significantly associated with loss of KLF4 expression and increased colon cancer cell proliferation.

To determine the role of IFITM3 in colon cancer cell proliferation, we treated HCT116 and SW480 cells with IFITM3 siRNA. We confirmed the efficacy of knockdown of IFITM3 expression by using Western blotting (Fig. 3B). We determined cell viability by using an MTT assay and colony-formation ability by using a soft agar assay. We found that knockdown of IFITM3 expression significantly inhibited the growth of cancer cells (Fig. 3C). This result was consistent with the finding showing a direct correlation between increased IFITM3 expression and a high proliferation index in colon tumor specimens. These data suggested that overexpression of IFITM3 contributes to hyperproliferation of colon cancer cells.

Because IFITM3 expression was correlated with both distant and nodal metastases of colon cancer (Supplementary Tables S8 and S9), we investigated the role of IFITM3 in the migration and invasion of colon tumors. We found that the levels of both migration and invasion of IFITM3 siRNA–transfected HCT116 and SW480 cells were significantly lower than those of control cells (P < 0.001; Fig. 4). Consistent with those in vitro findings, knockdown of IFITM3 expression suppressed the growth and metastasis of HCT116 cells in nude mouse model whereas overexpression of IFITM3 did the opposite (Fig. 5A and B).

To further investigate the relationship between IFITM3 and KLF4 expression, we generated colon-specific KLF4 knockout mice, confirming the gene deletion by histologic analysis (Fig. 5C), which showed an absence of goblet cells in colon mucosa, and Western blotting analysis (Fig. 5D). We then measured the expression of both KLF4 and IFITM3 in KLF4 knockout (KLF4^−/−), KLF4 heterogeneous (KLF4^+/−), and KLF4 wild-type (KLF4^+/+) mice using both Western blot and immunohistochemical analyses. The KLF4^+/+ mouse had the highest KLF4 but the lowest IFITM3 expression levels (Fig. 5D and E). Therefore, data from our studies of both human colon tumor specimens and this mouse model showed that decreased KLF4 expression was significantly correlated with increased IFITM3 expression.

To identify the molecular mechanisms of negative regulation of IFITM3 expression by KLF4, we first examined the effects of KLF4 overexpression on IFITM3 expression in HCT116 and SW480 cells. Restored KLF4 expression significantly repressed IFITM3 expression at both the mRNA (Fig. 6A1) and protein levels (Fig. 6A2 and A3). Furthermore, to determine the role of KLF4 in regulation of IFITM3 transcription, we generated an IFITM3 promoter and mutations of IFITM3 in 2 KLF4-binding sites in HCT116 and SW480 cells using site-specific mutagenesis (Fig. 6B1). Transfection of KLF4 significantly inhibited the IFITM3 promoter activity (Fig. 6B2), whereas mutations of the IFITM3 promoter reporter in either (Mut1 or Mut2) or both (Mut3) KLF4-binding sites significantly increased this activity (Fig. 6B3). These results suggested that the KLF4-binding sites were negative regulatory elements in the IFITM3 promoter. Finally, we carried out a ChIP assay to determine how KLF4 directly interacts with the IFITM3 promoter. We transduced both HCT116 and SW480 cells with Ad-EGFP and Ad-KLF4. We then immunoprecipitated the chromatins using an anti-KLF4 antibody and amplified the regions from −438 to +74 bp flanking the putative KLF4-binding side by using specific PCR primers. As shown in Figure 6C, anti-KLF4 antibodies, but not control IgG, amplified the predicted size DNA fragments from the precipitates of the cells. These results suggested that endogenous KLF4 bound to the IFITM3 promoter in both HCT116 and SW480 cells and that enforced KLF4 expression increased the binding of KLF4 to the IFITM3 promoter.

In this study, we discovered 4 lines of evidence supporting a critical role for IFITM3 in colon cancer pathogenesis. First, we observed a direct correlation of elevated IFITM3 expression with colon tumor progression. IFITM3-positive staining of colon cancer cells could be used to identify a greatly increased risk of recurrence and metastasis in patients after colectomy. Second, overexpression of IFITM3 increased the proliferation and metastasis of human colon cancer cells, indicating that targeting of IFITM3 is a potential therapeutic strategy for colon cancer. Third, genetically enforced KLF4 overexpression led to decreased IFITM3 expression in colon cancer cells, and we observed a negative correlation between KLF4 expression and IFITM3 expression in both human colon tumor samples and a KLF4 knockout mouse model. Fourth, KLF4 directly regulated the expression of the IFITM3 gene at the transcriptional level via binding to the IFITM3 promoter. Therefore, loss of KLF4 expression during the initiation and development of colon tumors contributed to abnormal IFITM3 expression and activation; the latter may be a novel molecular marker for poor prognosis and contribute to colon tumor pathogenesis and aggressive colon cancer biology.

Numerous studies have indicated the importance of IFITM3 expression in the oncogenesis of several malignancies, including colorectal cancer (16, 21). In addition, authors have reported upregulated IFITM3 expression in early and late intestinal neoplasms (4, 14, 15). However, the predictive significance of IFITM3 in colon cancer remains unclear. In the present study, we specifically investigated the expression of IFITM3 in TMA containing tumor specimens, matched normal colon tissue specimens, and lymph node metastasis specimens obtained from 203 patients with colon cancer. We observed the strongest IFITM3-positive staining in the nodal metastasis specimens. We also found that IFITM3 expression was significantly correlated with an advanced cancer biology, which was indicated by invasion depth, lymph node metastasis, and distant metastasis. Also, IFITM3-positive staining was associated with poor OS and DFS rates. Therefore, we provide the first evidence that IFITM3 can be used as a novel biomarker for outcome in patients with colon cancer after colectomy. This means that at the time of initial diagnosis of colon cancer, IFITM3 expression can be used not only to design optimal, individualized treatment but also to distinguish patients who would benefit from close monitoring after surgery from those who would not.

Researchers have suggested that IFITM genes function in a variety of contexts, including immune cell regulation, somitogenesis, and germ-cell development (8, 10–13). However, the precise roles of IFITM3 in carcinogenesis remain to be elucidated. Herein we provide evidence indicating that IFITM3 expression may affect the proliferation, migration, and invasiveness of colon cancer cells. Consistent with our findings, investigators have established that expression of IFITM1 promotes head and neck tumor invasion in the early stages of disease progression by mediating the expression of molecules downstream, including matrix metalloproteinases 12 and 13 (20). Recently, Lange and colleagues reported that targeted IFITM3 mutations have no detectable effects on development of the germ line or generation of live young mice (29), which contrasts with a previous study showing that IFITM3 mediates the migration of early primordial germ cells (9, 11). The different roles of IFITM3 in the development of the germ line and tumorigenesis may depend primarily on the involvement of different signaling pathways in these processes. Our current results suggested that IFITM3 could regulate both colon cancer cell growth and invasion. Whether IFITM3 is more important in regulating cell growth than cell migration and the underlying mechanism remain to be determined further.

Until now, the mechanisms responsible for IFITM3 overexpression in malignancies have been unknown. However, a study showed that in conditional Apc mutant mice, deletion of Apc alleles, which leads to the formation of colon adenomas, rapidly induced IFITM3 expression (21). That study also showed that induction of wild-type adenomatous polyposis coli (APC) resulted in a strong reduction of β-catenin signaling and IFITM3 expression in HT29 colon cancer cells (21). Thus, IFITM3 is a potential therapeutic target of activated β-catenin signaling in carcinogenesis in the intestinal epithelium. Nevertheless, the molecular mechanism of IFITM3 overexpression underlying the progression of colon cancer remains undefined. In a previous study, we found that loss of KLF4 expression plays a critical role in the pathogenesis and development of digestive system tumors, including gastric and pancreatic cancer (26, 30). In the present study, we investigated whether expression of the IFITM3 gene is a downstream molecule of KLF4 in colon cancer cases. Our results indicated that KLF4 could bind directly to the IFITM3 promoter and downregulate transcriptional activation of IFITM3. Also, we observed an inverse correlation between KLF4 expression and IFITM3 expression in human colon tumor specimens and a murine model of colon cancer. KLF4 has been known to be a highly expressed transcription factor in the adult intestine and critical for intestinal differentiation. Researchers have observed loss of KLF4 expression in many malignancies and that restoration of KLF4 expression can antagonize the development and progression of intestinal tumors (31). Consistently, a recent study has clearly shown that expression of KLF4 is a prognostic predictor for colon cancer (32) and a haploinsufficiency of Klf4 promotes APC-dependent intestinal tumorigenesis (31). In addition, KLF4 directly interacts with the C-terminal transactivation domain of β-catenin and inhibits Wnt/β-catenin signaling in intestinal tumors (33). Thus, KLF4 can mediate IFITM3 gene expression via both direct transcriptional inhibition and attenuation of the Wnt/β-catenin signaling pathway. It warrants further investigations on how the interactions between Wnt/β-catenin and KLF4 regulate the expression and function of IFITM3 and contribute to colon carcinogenesis. Therefore, improved understanding of the mechanism of KLF4-mediated inhibition will lead to novel therapies for colorectal cancer.

In summary, this study provided critical insight into the role of the IFITM3 gene in the progression of colon cancer. We showed that IFITM3 expression plays important roles in colon tumor progression by downregulating the proliferation, migration, invasion, and metastatic potential of tumor cells. The frequent upregulation of IFITM3 expression in human colon cancer cells highlights its potential as a novel therapeutic target for this cancer, which is closely associated with deregulated Wnt signaling and KLF4 signaling.

No potential conflicts of interest were disclosed.

We thank Don Norwood for editorial comments.

The work is supported in part by grants from the National High Technology Research and Development Program of China (“863” Program 2007AA022003), Key Basic Research Project of the Science and Technology Commission of Shanghai Municipality (05JC1402), and Outstanding Medical Academic Leader of Shanghai Municipality (LJ06024; to Z. Peng); and grants R01-CA129956 from the National Cancer Institute, NIH (to K. Xie).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.