|Home | About | Journals | Submit | Contact Us | Français|
C-terminal tensin-like (cten) is a focal adhesion molecule belonging to the tensin family. Previous studies have suggested that cten may function as a prostate specific tumor suppressor. Here, we show that although cten is expressed at a very low level in normal colon, its expression is significantly up-regulated in colon cancer. Furthermore, a high population of cten is found in the nucleus, where it interacts with β-catenin, a critical player in the canonical Wnt pathway. This interaction may contribute to the role of cten in enhancing the colony formation, anchorage-independent growth, and invasiveness of colon cancer cells. Our studies have identified cten as a novel nuclear partner of β-catenin, demonstrated an oncogenic activity of cten in colon cancers, and revealed cten as a potential biomarker and target for colon cancers.
Colon cancer is the third most common cancer worldwide and, like other cancers, it results from the cumulative effects on sequential genetic alterations in proto-oncogenes, tumor suppressors and DNA repair genes. It is estimated that greater than 90% of all colon cancers will have an activating mutation of canonical Wnt signaling pathway, ultimately leading to the accumulation of β-catenin in the nucleus of a cell (1). In the absence of Wnt, β-catenin is in a cytoplasmic Apc/Axin/GSK3/CK1 protein complex where it is phosphorylated by GSK3. Phosphorylation targets β-catenin for ubiquitination and degradation. Wnt binds to their receptors and activates Disheveled that inhibits the GSK3-directed phosphorylation of β-catenin, preventing its degradation. Consequently, levels of β-catenin increase and the protein accumulates in the nucleus where it binds to the transcription factors Tcf/Lef and other co-activators, leading to transcriptional activation of multiple target genes, such as c-myc and cyclin-D1. The activated transcriptional program directs cell proliferation and survival and alters cell fates (2, 3).
Focal adhesions are the transmembrane junctions between the extracellular matrix and the cytoskeletal network, and regulate biological events such as cell adhesion, migration, proliferation, differentiation, and gene expression (4, 5). When the regulation process goes awry, it may lead to cell transformation and cancer development. Previously we have identified a tensin family member, cten (C-terminal tensin-like), whose expression is relatively restricted at normal prostate gland and placenta. Our studies have demonstrated that cten is down-regulated in prostate cancers (6), is cleaved by caspase 3 during apoptosis (7), recruits the Deleted in Liver Cancer 1 (DLC-1) to focal adhesions and this interaction is critical for DLC-1′ tumor suppression activity (8, 9). Furthermore, the cten gene is located on chromosome 17q21, a region often deleted in prostate cancer. These data strongly support that cten is a tumor suppressor in the prostate. Nonetheless, we and others have found that although cten was not normally expressed in some tissues, its expression was up-regulated when tissues become cancerous (10-12), suggesting that cten might possess an oncogenic activity in certain tissues.
In this study, we report the expression and nuclear localization of cten in colon cancer, the identification of β-catenin as a binding partner of nuclear cten, and the role of cten in the tumorigenicity of colon cancer cells. Our findings have linked a focal adhesion molecule cten to canonical Wnt signaling pathway, demonstrated an oncogenic activity of cten in colon cancers, and revealed cten as a potential biomarker and target for colon cancers.
TissueScan Colon Cancer Panel and Colon qPCR Array - Matched Pairs (OriGene Technologies, Rockville, MD) were used to examine cten mRNA expression profile. The PCR was performed by initial activation for 2 min at 50°C and denaturation for 5 min at 95°C, followed by 42 cycles of 95°C for 15 s, 60°C for 30 s and 72°C for 30 s using the SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA). Primers used for cten mRNA detection were 5′- GGACCCCAGAGGACCTTGACTCCTA -3′ (forward) and 5′- GGGAAGCAGCTGGAAGGTGGGGT -3′ (reverse). Relative quantification of cten expression was measured by normalization against β-actin using the ΔCT method. Fold changes between the average of normal tissues (control) and the samples in Fig. 1A were quantified as 2-(ΔCT sample-ΔCT control).
Colon cancer tissue arrays from Imgenex (San Diego, CA), BioChain Institute (Hayward, CA), and UCD Cancer Center Specimen Repository were deparaffinized and rehydrated. After Antigen retrieval, endogenous peroxidase activity was blocked with 0.3% hydrogen peroxide for 20 min followed by normal serum blocking. Slides were then incubated with cten antibodies for 60 min at room temperature. Detection and visualization were performed by avidin-biotin complex system and DAB was used as the substrate. Slides were counterstained with hematoxylin.
Human colon tissues were obtained from the UCD Cancer Center Specimen Repository. These tissues were collected and frozen at -80°C immediately after surgical removal. The frozen sample was pulverized by mortar and pestle in liquid nitrogen, and reconstituted in ice-cold lysis buffer (1% Triton X-100, 50 mM Tris-HCl/pH8.0, 150 mM NaCl, 5 mM EDTA) with proteinase inhibitors (10 μg/ml aprotinin, 10 μg/ml leupeptin, 1 μg/ml pepestatin and 1 μM PMSF). Tissue lysates were clarified by centrifugation.
For subcellular fractionation, cells or pulverized tissues were suspended in buffer A (10 mM Hepes/pH 7.9, 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 0.15% NP-40, and proteinase inhibitors), swollen on ice for 10 min, and then centrifuged at 12,000 xg for 30 s at 4°C. Supernatant was collected as cytoplasmic extract. The pellet was washed, resuspended in buffer B (20 mM Hepes/pH 7.9, 400 mM NaCl, 1 mM EDTA, 1 mM EGTA, 0.5% NP-40, and proteinase inhibitors), and rocked for 15 min at 4°C. The supernatant after centrifugation was used as nuclear extract.
Colony formation assay was performed as previously described (9). The siRNA targeting APC (Dharmacon, Chicago, IL), β-catenin (Santa Cruz Biotechnology, Santa Cruz, CA), cten (Invitrogen) or a nonspecific siRNA (Santa Cruz Biotechnology) were transfected together with pEGFP (for drug selection) into SW480 using Lipofectamine 2000.
5000 cells of SW480 were resuspended in medium containing 0.3% low-melting agarose 24 h after transfection of indicated siRNA and then layered on top of a solidified 0.5% base agarose. After 3 weeks incubation, the colonies were stained with crystal violet.
SW480 cells transfected with the siRNA targeting β-catenin, cten or a nonspecific siRNA were seeded on 8.0 μm pore Transwells (Corning) previously covered with Matrigel. DMEM medium with 10% serum was used as a chemoattractant in the lower chamber. After 40 h incubation, the invading cells were fixed in formaldehyde and stained with crystal violet.
To examine cten expression in colon cancers, the mRNA and protein levels of cten were analyzed. Cten transcripts were up-regulated (> 4 folds) in 91% (39/43) (or >10 folds in 74%) of colon cancers by qPCR analysis (figure 1A). Additionally, over 4-fold increases of cten transcripts were detected in 83% (20/24) of the cancerous samples comparing with those in the adjacent normal counterparts in the same individual (figure 1B). At the protein levels, cten immunohistochemical (IHC) staining in normal colon tissues was very weak and was significantly enhanced in 76% (90/118) of cancer samples (figure 1C). Interestingly, we often observed nuclear, in addition to cytoplasmic, staining of cten in cancer samples. We further examined cten protein levels in matched cancer and adjacent normal colon tissue pairs by immunoblotting (figure 1D). Cten protein was only detected in the cancer samples and the levels matched very well with the percentage of tumor lesion in each sample. To confirm cten nuclear staining, we have isolated cytoplasmic and nuclear fractions from colon cancer tissue for immunoblotting analysis and detected cten proteins in both compartments (figure 1E). In human cell lines, cten was expressed at higher levels in SW480, SW620, HT29 colon cancer cell lines and MLC-SV40 non-malignant prostate epithelial cells, while lower levels were found in HCT116 and LoVo cells (figure 2A). In contrast, cten was not detected in the FHC human normal colon epithelial cells. Cten subcellular localization was further analyzed in SW480 and HCT116 cells. Again, cten was present in nuclear and cytoplasmic fractions in both cell lines cells (figure 2B). These studies have demonstrated that (a) cten is expressed at a very low level in normal colon tissues and cells; (b) cten is overexpressed at mRNA and protein levels in colon cancers; and (c) a population of cten is present in the nuclei of colon cancer cells.
Because Wnt signaling is highly involved in colon cancers, we speculated that cten might participate in β-catenin canonical pathway and tested whether cten interacted with β-catenin. Indeed, β-catenin was co-immunoprecipitated with cten (figure 3A). Since both cten and β-catenin travel between the cytoplasm and nucleus, we asked where the interaction took place by separating nuclear and cytoplasmic fractions for co-immunoprecipitation assay. Interestingly, even more β-catenin and cten proteins were found in the cytoplasm, β-catenin/cten interaction was only detected in the nucleus (figure 3B).
To establish the function of cten in tumorigenicity of colon cancer cells, we have carried out colony formation, soft agar and invasion assays. For colony formation, we first manipulated cten expression levels in two cell lines by suppression in cten high expressing SW480 cells or enhancement in cten low expressing HCT116 cells. As shown in figure 4A and B, suppression of cten expression reduced the colony formation activity of SW480, while over-expression of cten in HCT116 further enhanced its colony formation activity. We next asked whether the tumor promoting effect of cten is associated with the canonical Wnt pathway. Since APC is a scaffolding protein in part of a complex modulated by Wnt signaling and APC negatively regulates the activation of β-catenin, colony formation assays were carried out by suppressing endogenous truncated APC (1-1337) and wild-type β-catenin alone or together with cten in SW480 cells. Although knockdown of truncated APC showed a mild effect, it is not statistically significant (figure 4A). Nonetheless, cten knockdown significantly reduced the number of colonies to the similar level with that of β-catenin knockdown. Similarly, treatment of cten siRNA remarkably suppressed the anchorage-independent growth (Fig. 4C) and profoundly inhibited cell invasion (Fig. 4D) to a level comparable to those in β-catenin-depleted SW480 cells. However, no additive or synergistic effect was detected when both cten and β-catenin expressions were suppressed simultaneously. Taken together, theses results suggested that cten regulates the tumorigenicity of colon cancer cells through the β-catenin-dependent pathway.
In this study, we have detected a consistent up-regulation of cten in all stages of colon cancer (figure 1) and its effect on colony formation, anchorage-independent growth, and cancer cell invasion (figure 4), indicating that cten over-expression is associated with colon tumors at very early stage and is likely to be involved in tumor formation as well as progression. Recently, Sabates-Bellver et al (13) reported the transcriptome profile of human colorectal adenomas by analysis of microarray data from 32 adenoma/normal mucosa tissue pairs. In agreement with our finding, they had detected a 4.8-fold increase of cten mRNA in adenomas, which were the precancerous lesions in the colon. Together with its very low basal expression level in normal colon tissues makes cten a very attractive biomarker for colon cancer.
A main role of cten at focal adhesions is to regulate the integrin-actin cytoskeleton organization and focal adhesion complex for cell adhesion/migration events. This role is accomplished, at least in part, by the tensin-cten expression switch (12) and by recruiting DLC-1 to focal adhesion sites. Focal adhesion localization of DLC-1 is essential for DLC-1’s tumor suppression activity (8, 14). This also suggests a role of cten in preventing cell transformation. However, this function of cten probably is not relevant in the colon, since cten is not expressed in normal colon cells and DLC-1 is often absence or down-regulated in colon cancers (15).
Another two novel findings in this report are that cten is a nuclear protein and interacts with nuclear β-catenin, linking a focal adhesion molecule to the canonical Wnt pathway. It is known that a group of focal adhesion molecules, including Abl family, zyxin family, paxillin family, and CRP family, is able to travel to the nucleus (16). Some of these molecules bind to transcription factors and function as co-regulators and others may serves as a chaperone that directs the targeting of specific mRNAs to nascent focal adhesions for localized protein translation. For example, LPP interacts with PEA3 transcription factor and transactivates PEA3-dependent gene expression (17). TRIP6 is recruited to AP-1 and NF-kB promoters, where it serves as a co-activator (18). Paxillin binds to PABP1 (PolyA-binding protein) and promotes the export of PABP1 bound mRNAs into cytoplasm and targets the ribonucleoprotein complex to the leading edge of a migrating cell (19). Cten has joined this list of nucleus/focal adhesion molecules as it interacts with nuclear β-catenin. Although many functions of cten rely on the present of β-catenin (figure 4), the role of cten/β-catenin interaction remains to be investigated. It is possible that cten may be involved in the nuclear retention of β-catenin or regulation of β-catenin/Tcf transcriptional activity.
This work is supported by grants from the National Institutes of Health (CA102537) to SHL.
No potential conflicts of interest