THBS2 promotes cell migration and invasion in colorectal cancer via modulating Wnt/β‐catenin signaling pathway

Abstract

Colorectal cancer (CRC) is the most common human digestive malignancy with a poor prognosis; the pathophysiology of colon cancer involves multiple linkages of regulatory networks. Recently, thrombospondin 2 (THBS2) has been extensively studied for its role in cancer progression. In this study, we evaluated the expression of THBS2 in CRC tissues and studied the possible mechanism by which THBS2 regulates CRC progression. Our results showed that the upregulation of THBS2 in CRC tissues and CRC cell lines and high expression of THBS2 was correlated with poor overall survival. The in vitro experimental data showed that THBS2 overexpression promoted CRC cell growth, invasion, and migration, while THBS2 inhibition exerted tumor‐suppressive actions on CRC cells. THBS2 knockdown suppressed the activity of Wnt/β‐catenin signaling. Collectively, the results implied that THBS2 exerted promotional effects on CRC cell proliferation, invasion, and migration, partly by modulating the Wnt/β‐catenin signaling pathway.

1. INTRODUCTION

Colorectal cancer (CRC) is one of the most common human malignancies and is a leading cause of cancer‐related deaths worldwide. ¹ CRC cells exert distant invasive potential and metastatic ability; this is considered responsible for approximately 90% of colon cancer‐associated mortality. ² CRC metastasis is a complex and multistep process, and many molecules are involved in this process. The canonical Wnt/β‐catenin signaling pathway is considered a vital pathway for enhancing the metastatic properties of cancer cells. ³ Nuclear accumulation of β‐catenin has been found in up to 80% of patients with colon cancer. ⁴

Thrombospondins are multifunctional glycoproteins released from various types of cells, including stromal fibroblasts, endothelial cells, and immune cells. ⁵ Thrombospondin‐2 (THBS2) exerts diverse biological effects, such as angiogenesis, cell motility, apoptosis, and cytoskeletal organization by binding to extracellular matrix (ECM) proteins and cell surface receptors. ⁶ Additionally, THBS2 plays the role of an oncogene and an anti‐oncogene in epithelial tumors; its oncogenic effect has been given increasing attention. ⁷ , ⁸

THBS2 and the Wnt/β‐catenin signaling pathway are known to be closely related in various cancers. ⁹ THBS2, a microRNA‐744‐5p target, modulates MMP‐9 expression through CUX1 in pancreatic neuroendocrine tumors. ¹⁰ Silencing of THBS2 inhibits gastric cancer cell proliferation, migration, and invasion, while promoting apoptosis through the PI3k‐Akt signaling pathway. ¹¹ THBS2 is highly expressed in colon cancer and correlates with patient survival. ¹² Moreover, one preliminary study revealed that THBS2 silencing could inhibit CRC cell proliferation, migration, and invasion. ⁵ To date, the molecular mechanisms by which THBS2 regulates CRC cell growth and metastasis‐related traits are largely unknown.

In the present study, we determined that THBS2 expression was significantly upregulated in CRC tissues compared to that in nontumorous tissues. We showed that THBS2 is a critical activator of CRC cell migration and invasion. Moreover, THBS2 activated the Wnt/β‐catenin signaling pathway to promote CRC cell aggressiveness.

2. MATERIALS AND METHODS

2.1. Tissue specimens

In total, 31 pairs of CRC specimens and surrounding nontumorous tissues were obtained from patients who were diagnosed histopathologically at the Inner Mongolia Forestry General Hospital. All resected tissues were immediately snap‐frozen in liquid nitrogen and stored at −80°C until use. Patient consent and approval from the Institutional Research Ethics Committee were obtained for the use of clinical materials for research. The clinical characteristics of the patients are shown in Table 1.

TABLE 1.

The clinicpathological factors of 31 CRC patients

Variables	Number of cases
Age (years)
<60	13
≥60	18
Gender
Male	22
Female	9
Tumor size
<5	20
≥5	11
Histology
Adenocarcinoma	31
Others	0
Tumor stage
I–II	19
III–IV	12
T stage
T1‐T2	6
T3‐T4	25
N stage
N0	20
N1 + N2	11
M stage
M0	23
M1	8

Abbreviation: CRC, colorectal cancer.

2.2. Cell culture

The normal colonic epithelial cell line, FHC, was purchased from Nanjing Cobioer Biotechnology Co., Ltd (Nanjing, Jiangsu, China), and colon cancer cell lines LOVO and HCT‐116, were purchased from the Cell Bank of Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). HCT‐116, LOVO, and FHC cells were authenticated by short tandem repeat (STR) genotyping (Procell Life Science & Technology Co., Ltd., China). The STR genotype was consistent with the published genotype. All cell lines were cultured in Dulbecco's Modified Eagle Medium (Thermo Fisher Scientific, Waltham, MA, USA) with 10% fetal bovine serum in a humidified incubator with 5% CO₂ at 37°C. The Wnt/β‐catenin inhibitor, XAV939, was obtained from Selleck (Houston, TX, USA).

2.3. Plasmid, shRNA, and transfection

The plasmids of the human THBS2 gene, shRNA targeting human THBS2 (sh‐THBS2), and their controls were synthesized by Genechem (Shanghai, China). The cells were transfected with shRNA, plasmids, or their controls using Lipofectamine 3000 (Thermo Fisher Scientific). Western blotting was used to detect transfection efficiency 24 h after transfection. Lentiviruses were generated in 293 T cells by co‐transfecting plasmids, including the lentiviral vector (pLKO‐sh‐Ctrl or pLKO‐sh‐THBS2), using Lipofectamine 3000 reagent. Virus‐containing supernatants were collected for infection 48 h post‐transfection. For viral transduction, lentiviruses were incubated with LOVO cells overnight at 37°C in a humidified cell culture incubator. Cells were then selected in the presence of 1.5 μg/ml puromycin 24 h post‐infection.

2.4. Quantitative real‐time PCR

Total RNA from colon cancer tissues and cells was extracted using TRIzol reagent (Thermo Fisher Scientific). Total RNA (0.5 μg) was reverse transcribed to cDNA using PrimeScript™ RT reagent Kit with gDNA Eraser (Takara, Dalian, China). SYBR Green Polymerase Chain Reaction Master mix (Takara) was used to determine the mRNA levels of THBS2 on an ABI‐7900 Fast Real‐time Detection System (Applied Biosystems, Carlsbad, CA, USA). The relative expression of THBS2 was normalized to that of the internal reference GAPDH. The 2^ΔΔCt method was used to analyze the data. The primers used in the study were as follows: GAPDH forward primer: 5′‐TGTGGGCATCAATGGATTTGG‐3′, GAPDH reverse primer: 5′‐ACACCATGTATTCCGGGTCAAT‐3′; β‐catenin forward primer: 5′‐TGCCATTCCACGACTAGTTCAG‐3′, β‐catenin reverse primer: 5′‐CGTACGGCGCTGGGTATC‐3′; THBS2 forward primer: 5′‐GACACGCTGGATCTCACCTAC‐3′, and THBS2 reverse primer: 5′‐GAAGCTGTCTATGAGGTCGCA‐3′.

2.5. Cell proliferation and colony formation assay

Cell viability was assessed by Cell Counting Kit‐8 (CCK‐8) assay (Dojindo Laboratories, Kumamoto, Japan). Cells (3 × 10³ cells per well) were seeded in 96‐well plates. With supernatant removed, 10 μl CCK‐8 reagent in 100 μl medium was added to each well at 0, 24, 48, 72, and 96 h. The plates were incubated in the dark at 37°C for 2 h, and absorbance at 450 nm was detected with a microplate reader (BioTek, Winooski, VT, USA). Colony forming ability was assessed using a colony formation assay. Cells were seeded into a six‐well plate. Fourteen days later, the number of colonies was counted.

2.6. Invasion assay

For the invasion assay, the cells were plated on Matrigel‐coated upper chambers (24‐well inserts; pore size, 8 mm; BD Biosciences, San Jose, CA, USA). The 200 μl cell suspension was plated on uncoated upper chambers. In the lower wells, medium was replaced with fresh medium with 5% FBS. After 48 h, the noninvading cells from the upper surface of the filter were wiped with a cotton swab. The cells that invaded to the lower side of the filter were counted and photographed with an optical inverted microscope. Five random fields in each assay were counted and averaged.

2.7. Wound healing assay

To confirm the migration ability, a wound healing assay was carried out. Cells were seeded into a six well plate and allowed to grow from 90% to 100% confluence in Dulbecco's modified eagle medium , and cell monolayers were then wounded by a 200 ml pipette tip. The distance covered by migrated cells was quantified at 0 and 24 h.

2.8. Western blotting

Total proteins were extracted using ice‐cold radioimmunoprecipitation assay buffer (Cell Signaling Technology, Danvers, MA, USA). NE‐PER™ Nuclear and Cytoplasmic Extraction Reagent (Thermo Fisher Scientific) was used to extract nuclear and cytoplasmic proteins. Equivalent proteins were separated using 10% SDS‐PAGE and transferred to polyvinylidene fluoride membranes (Millipore, Braunschweig, Germany). The membranes were then probed with antibodies against THBS2, GAPDH, β‐catenin, MMP‐9, cyclin D1, or laminB (Abcam, Cambridge, UK) overnight at 4°C and incubated with horseradish peroxidase‐conjugated secondary antibody (Cell Signaling Technology) for 2 h at room temperature. The bands were visualized using an enhanced chemiluminescence kit (Millipore).

2.9. TOP‐Flash/FOP‐Flash reporter assay

LOVO or HCT116 cells (5 × 10⁴) were seeded into a 24‐well plate and were transfected with β‐catenin‐LEF/TCF‐sensitive (TOP) or β‐catenin‐LEF/TCF insensitive (FOP) reporter vector (Addgene) together with a pRL‐TK Renilla luciferase construct (Promega) for normalization of transfection efficiency using Lipofectamine 2000 (Thermo Fisher Scientific). After transfection for 48 h, a Dual‐Luciferase Assay Kit (Promega) was used according to the manufacturer's instructions to detect luciferase activity. The ratio of TOP/FOP indicated the activity of Wnt/β‐catenin signaling pathway.

2.10. Luciferase reporter assay

Fragments of the β‐catenin promoter (predicted from position −2000 bp to +44 bp) was cloned into the pGL3‐Basic Vector (Promega) to generate β‐catenin promoter reporter plasmids. Cells were seeded in 24‐well plates and transfected with plasmids containing firefly luciferase reporters and a pRL‐TK Renilla luciferase construct (Promega). After transfection for 48 h, the luciferase activity was detected. The transfection efficiency was normalized with Renilla luciferase activity.

2.11. Immunohistochemistry

CRC specimens and surrounding nontumorous tissues were resected to at a thickness of 4 μm on SuperFrost slides (Braunschweig, Germany). The sections were then dewaxed, rehydrated and antigen‐retrieved by using the microwave. Endogenous peroxidase activity was eliminated via 0.5% H₂O₂. After blocking for 1 h at 25°C, the tissues were incubated with THBS2 antibody (Abcam) and then developed using 3,3′‐diaminobenzidine tetrahydrochloride developer. After counterstaining with hematoxylin, dehydration with gradient ethanol, and mounting with neutral resins, tissue sections were observed under a microscope (Leica Microsystems, Wetzlar, Germany). The staining intensity was scored as 0 (no staining), 1 (weak), 2 (moderate), 3 (strong). The staining area was scored as 0 (<10% positive staining), 1 (10%–25% positive staining), 2 (25%–50% positive staining), 3 (50%–75% positive staining), and 4 (>75% positive staining). The final staining score was determined using formula: overall score = intensity score × percentage score.

2.12. Co‐immunoprecipitation

Co‐immunoprecipitation of β‐catenin and THBS2 was performed in HCT‐116 and LOVO cells. In total, 5 × 10⁶ cells were lysed using ice‐cold radioimmunoprecipitation assay buffer containing protease and phosphatase inhibitors. After a 30‐min incubation at 4°C, the total extracts were clarified by centrifugation at 12,000 rpm for 30 min. In IP, the cell extracts were precleared with 50 μl protein A + G agarose (Yeasen, Shanghai, China) for 1 h at 4°C and were immunoprecipitated with 2 μg of anti‐rabbit IgG or anti‐THBS2, respectively. The immune complexes were washed three times with wash buffer for 3 min to avoid nonspecific binding of the associated proteins. Finally, immunoprecipitates were resuspended in an SDS‐PAGE sample buffer containing loading dye and used for western blotting.

2.13. In vivo tumorigenesis assay

Nude athymic BALB/c 6‐weeks‐old mice were examined according to institutional guidelines, and all procedures were approved by the Inner Mongolia Forestry General Hospital. LOVO cells stably transfected with sh‐Ctrl or sh‐THBS2 were injected subcutaneously into the flanks of nude mice, followed by measurement of the tumor volume at 2‐day intervals using the formula: volume = (short diameter)² × (long diameter)/2. The mice were sacrificed 2 weeks after injection, and the tumors were harvested to determine tumor mass and immunohistochemical analysis of the expression levels of MMP‐9 and cyclin D1.

2.14. Statistical analysis

Statistical analyses were performed using GraphPad Prism version 7.0 (GraphPad Software, La Jolla, CA, USA). The results are expressed as mean ± standard deviation (SD). We used Spearman's correlation analysis to describe the correlation between quantitative variables in mRNA‐seq data from the Cancer Genome Atlas (TCGA) tumors. Student's t‐test and one‐way analysis of variance were conducted to analyze the differences between groups. Statistical significance was set at p <0.05.

3. RESULTS

3.1. THBS2 is overexpressed in human CRC specimens

To determine the role of THBS2 in CRC, qRT‐PCR analysis was used to determine the mRNA levels of THBS2 in a set of 31 pairs of human CRC tumor specimens (T) and surrounding non‐tumorous tissues (N). The results showed that the mRNA levels of THBS2 were higher in human CRC specimens than in the surrounding nontumorous tissues (Figure 1A). Consistently, the results of immunohistochemical staining (Figure 1B,C) and western blotting (Figure 1D) showed that THBS2 was upregulated in CRC. Future analysis of the TCGA database using the online Gene Expression Profiling Interactive Analysis (GEPIA) tool (http://gepia.cancer‐pku.cn/) indicated that the expression of THBS2 was significantly higher in colorectal tumor tissues (tumor) than that in non‐neoplastic tissues (normal) (Fig. 1E). Notably, Kaplan–Meier survival analysis using the GEPIA tool revealed that patients with high THBS2 expression levels had poorer disease‐free survival and overall survival (Figure 1F,G). Finally, we examined the mRNA and protein levels of THBS2 in the normal colonic epithelial cell line, FHC, and CRC cell lines (LOVO and HCT‐116). The mRNA sand protein levels of THBS2 in CRC cell lines were significantly higher than those in FHC (Figure 1H,I). These results suggest that THBS2 may play a promoting role in CRC progression.

FIGURE 1.

THBS2 is elevated in human colorectal cancer (CRC) tissues. (A) Thrombospondin 2 (THBS2) mRNA level was upregulated in CRC tissues than surrounding nontumorous tissues, which was analyzed by quantitative real‐time PCR (qRT‐PCR). **p <0.01. (B) Immunohistochemical staining analysis of THBS2 protein expression in CRC tissues and surrounding nontumorous tissues (magnification, × 200). (C). The immunohistochemistry (IHC) scores of THBS2 staining in CRC and nontumorous tissues were performed. (D) Representative western blots of THBS2 protein expression were detected in CRC and nontumorous tissues. (E). The expression of THBS2 in CRC was analyzed according to Gene Expression Profiling Interactive Analysis (GEPIA) database. *p <0.05. (F) Overall survival analysis of THBS2 in patients with CRC was performed according to GEPIA database. (G) Disease free survival analysis of THBS2 in patients with CRC was performed according to GEPIA database. (H,I) mRNA and protein expression THBS2 in normal colonic epithelial cell line, FHC and CRC cell lines (LOVO and HCT‐116) were measured using qRT‐PCR and western blotting. GAPDH was used as the control. **p <0.01 compared with FHC

3.2. Downregulation of THBS2 inhibits CRC cell growth, migration, and invasion

To investigate the biological function of THBS2 in CRC cells, LOVO and HCT116 cells were transfected with a THBS2‐knockdown shRNA construct (sh‐THBS2) or shRNA negative control (sh‐Ctrl) (Figure 2A). THBS2 knockdown led to a reduction in cell proliferation rate (Figure 2B). Similarly, knockdown of THBS2 decreased the colony formation ability of LOVO and HCT116 cells (Figure 2C). Subsequently, a xenograft model was constructed by inoculating sh‐THBS2 stable transfected LOVO cells into nude mice. THBS2 depletion significantly reduced the tumor growth of LOVO cells in vivo (Figure 2D,E). We then measured the effects of THBS2 knockdown on the migratory and invasive abilities of CRC cells using wound healing and transwell invasion assays, respectively. The results showed that knockdown of THBS2 reduced the migratory and invasive abilities of LOVO and HCT116 cells (Figure 2F,G). These results indicated that THBS2 knockdown inhibited CRC cell growth, migration, and invasion.

FIGURE 2.

Reduced thrombospondin 2 (THBS2) inhibits the aggressive phenotypes of colorectal cancer (CRC) cells. (A) LOVO and HCT116 cells were transfected with sh‐THBS2 or sh‐Ctrl. Western blot analysis was performed to detect the expression of THBS2. (B) Cell growth was measured by CCK‐8 assay after THBS2 knockdown in LOVO and HCT116 cells. (C) THBS2 knockdown inhibited the growth of LOVO and HCT116 cells which was detected by the colon formation assay. (D) The xenograft pictures of sh‐Ctrl group (n = 6) and sh‐THBS2 group (n = 6). (E) Tumor volume in the nude mice injected with sh‐THBS2 stable transfected cells was measured. (F) Wound‐healing assay was used for detecting the migration ability in THBS2 knocked down LOVO and HCT116 cells. (G) The invading cells of the transwell assay were counted under a microscope in five randomly selected fields. **p <0.01 compared with sh‐Ctrl

3.3. Exogenous THBS2 promotes CRC cell proliferation, migration, and invasion

We assessed the changes in cellular behavior induced by exogenous THBS2 overexpression. The transfection efficiency of THBS2 in HCT‐116 cells was confirmed by western blotting (Figure 3A). Compared with the corresponding control cells, the proliferative capacity and colony formation of THBS2‐overexpressing HCT‐116 cells were significantly enhanced (Figure 3B,C). In addition, exogenous expression of THBS2 remarkably increased HCT‐116 cell migration and invasion (Figure 3D,E). Overall, overexpression of THBS2 plays an oncogenic role in CRC cells.

FIGURE 3.

Upregulation of thrombospondin 2 (THBS2) promotes the aggressive phenotypes of CRC cells. (A) HCT116 cells were transfected with vector or THBS2 overexpression plasmid. Western blot analysis was performed to detect the expression of THBS2. (B) Cell growth was measured by CCK‐8 assay after THBS2 overexpression in HCT116 cells. (C) THBS2 raised the growth of LOVO cells which was detected by the colon formation assay. (D) Wound‐healing assay was used for detecting the migration ability in THBS2 overexpressing HCT116 cells. (E) The invading cells of the transwell assay were counted under a microscope in five randomly selected fields; **p <0.01 compared with vector

3.4. THBS2 regulates the Wnt/β‐catenin pathway in CRC cells

The canonical Wnt pathway, the Wnt/β‐catenin pathway, is one of the pathways involved in the oncogenesis and progression of CRC. We postulated that THBS2 regulates the Wnt/β‐catenin signaling pathway to mediate tumor growth, migration, and invasion in CRC cells. Through mRNA expression correlation analysis using TCGA, we found that the expression of THBS2 was positively correlated with multiple molecules of the Wnt/β‐catenin signaling pathway, such as β‐catenin (CTNNB1), MMP‐9, and cyclin D1 (Figure 4A–C). The nuclear accumulation of β‐catenin is the central event in the activation of the Wnt/β‐catenin signaling pathway; therefore, we investigated whether THBS2 can regulate β‐catenin nuclear accumulation in CRC cells. As shown in Figure 4D, overexpression of THBS2 enhanced nuclear β‐catenin levels in CRC cells. We investigated whether the function of THBS2 in CRC cells was associated with the Wnt/β‐catenin signaling pathway. As shown in Figure 4E, the TOP/FOP‐Flash reporter activity in THBS2 overexpressing CRC cells was significantly higher than that in the vector group. To test whether THBS2 influences the expression levels of molecules downstream of the Wnt/β‐catenin signaling pathway, we next evaluated MMP‐9 and cyclin D1 expression and found that silencing THBS2 inhibited their expression and that exogenous THBS2 had the opposite effects (Figure 4F,G). Finally, a co‐immunoprecipitation assay indicated that THBS2 directly interacted with β‐catenin in CRC cells (Figure 4H). Moreover, silencing THBS2 inhibited the expression of MMP‐9 and cyclin D1 in subcutaneous tumor (Figure 4I,J). To elucidate the molecular mechanism of THBS2 activating Wnt/β‐catenin signaling, we analyzed whether THBS2 could regulate β‐catenin at transcriptional level. As shown in Figure 4K, the mRNA levels of β‐catenin were raised in CRC cells upon THBS2 overexpression. These observations indicated that THBS2 regulates β‐catenin at transcriptional level. We then cloned the β‐catenin promoter into pGL3‐basic luciferase reporter vector, and transfected it into LOVO and HCT116 cells to test the changes of the luciferase activities. The luciferase reporter results revealed that THBS2 significantly elevated the activity of β‐catenin promoter (Fig. 4L). Next, six luciferase reporters were constructed containing β‐catenin promoter fragments with different deletions between −2000 and + 44 upstream of transcriptional start site. As shown in Figure 4M, the luciferase activities of P1 (−2000 to +44) and P2 (−1756 to +44) promoters were significantly increased by THBS2 overexpression. In the other promoter regions, including P3 (−1472 to +44), P4 (−1188 to +44), P5 (−888 to +44) and P6 (−484 to +44), the luciferase activities failed to show significant differences by THBS2 overexpression. These results showed that the sequence between the nucleotides −1756 and − 1473 in the β‐catenin promoter may contain the THBS2‐binding site. Taken together, these data suggest that THBS2 regulates the nuclear accumulation of β‐catenin in CRC cells.

FIGURE 4.

Thrombospondin 2 (THBS2) promotes β‐catenin nuclear translocation. (A–C) The correlating THBS2 level with Wnt/β‐catenin pathway genes (β‐catenin, MMP‐9, and Cyclin D1) were detected using TCGA. (D) Western blot of β‐catenin in the nuclear fractions of the THBS2 overexpressing CRC cells. (E) THBS2 overexpressing LOVO and HCT116 cells were transfected with the TOP/FOP‐Flash reporter plasmid, and the reporter activities were detected 48 h after transfection by a luciferase assay. (F) Western blot of MMP‐9 and cyclin D1 expression in THBS2 knocked down LOVO cells. (G) Western blot of the MMP‐9 and cyclin D1 in THBS2 overexpressing HCT116 cells. (H) Co‐IP assays revealed the association between THBS2 and β‐catenin in HCT116 and LOVO cells. Inputs were used as control. Co‐IP was performed using antibodies as indicated. (I) The representative photographs of MMP‐9 and cyclin D1 staining of subcutaneous tumor are shown (magnification, × 200). (J) The percentage of MMP‐9 positive staining and cyclin D1 positive staining in subcutaneous tumor was counted under the microscope. **p <0.01 compared with sh‐Ctrl. (K) The mRNA levels of β‐catenin in THBS2‐overexpressing LOVO and HCT116 cells were determined using qRT‐PCR assay. (L) The luciferase reporter assays were performed using the luciferase reporter plasmid linked with full‐length native promoter of β‐catenin in LOVO and HCT116 cells. (M) The β‐catenin promoter structure was constructed and luciferase activity relative to Renilla control was measured in LOVO and HCT116 cells. **p <0.01 compared with vector

3.5. Effect of the β‐catenin inhibitor XAV‐939 is rescued by THBS2

Finally, we performed an assay to determine whether the inhibitory effect of the β‐catenin inhibitor XAV‐939 could be reversed by THBS2 overexpression. We cultured CRC (THBS2‐overexpression) cells in the presence of XAV‐939 and evaluated the expression of Wnt/β‐catenin signaling by western blotting. Treatment with 5 μM or 10 μM XAV‐939 decreased MMP‐9 and cyclin D1 expression in LOVO cell. In HCT116 cell, 5 μM XAV‐939 also reduced the expressions of MMP‐9 and cyclin D1 at both 24 h and 48 h. However, the expression of MMP‐9 and cyclin D1 was rescued by transfection with THBS2 (Figure 5A,B). Moreover, using colony formation and transwell assays, we found that when cells were treated with XAV‐939, the growth and invasion ability of HCT‐116 and LOVO cell lines were decreased. However, the growth and invasion ability of HCT‐116 and LOVO cells were rescued by transfection with THBS2 (Figure 5C,D).

FIGURE 5.

Thrombospondin 2 (THBS2) promotes colorectal cancer progression via regulating Wnt/β‐catenin signaling. (A‐B) Western blot analysis of the expression of MMP‐9 and cyclin D1 in THBS2 overexpressing HCT116 and LOVO cells with the XAV939 treated. (C) THBS2 overexpressing HCT116 and LOVO cells were treated with XAV939. Growth ability was measured by colony formation assay. (D) Invasion ability was measured by transwell assay. **P <0.01 compared with vector group, ^## p <0.01 compared with vector plus 5 μM or 10 μM XAV939 group

4. DISCUSSION

Metastasis involves the detachment of tumor cells from their primary location or tumor, their invasion into circulation or the lymphatic system, and their proliferation distant from the primary tumor. ¹³ Thus, we focused on the role of THBS2 in the migration and invasion of CRC cells and its underlying molecular mechanism. Our results demonstrated that THBS2 silencing markedly weakened cell growth, migration, and invasion of CRC cell lines in vitro and impaired their tumorigenic ability in vivo. Conversely, THBS2 overexpression significantly increased the migration and invasion of CRC cells, potentially explaining the association between THBS2 overexpression and worse overall survival in patients with CRC in this study.

The canonical Wnt/β‐catenin pathway plays a critical role in the proliferation and metastasis of cancer cells. ¹⁴ In particular, increased levels of β‐catenin have been associated with metastasis and poor prognosis in patients with breast cancer. During Wnt/β‐catenin signaling, β‐catenin translocates from the cytoplasm into the nucleus where it associates with T‐cell factor/lymphoid enhancer‐binding factor transcription factors and regulates downstream target genes that affect cell proliferation, invasion, and tumor progression, such as cyclin D1, c‐myc, and MMP‐9. ¹⁵ In the present study, western blotting revealed that THBS2 overexpression resulted in substantial accumulation of β‐catenin in the nuclei of CRC cells. We speculate that the tumor‐promoting effects of THBS2 are associated with the activation of Wnt/β‐catenin signaling. Accordingly, THBS2 overexpression increased the expression of the downstream target genes of Wnt/β‐catenin signaling, including cyclin D1 and MMP‐9. In addition, we demonstrated a positive correlation between THBS2 and Wnt/β‐catenin signaling pathway. Mechanistically, β‐catenin expression is regulated by THBS2 at the transcriptional level. To further explore whether THBS2 could directly activate the transcription of β‐catenin, the dual‐luciferase assays were conducted and revealed that THBS2 directly promoted the promoter luciferase activity of β‐catenin in CRC cells. ¹⁶ We further truncated the β‐catenin promoter and identified that the sequence between the nucleotides −1756 and − 1473 in the β‐catenin promoter may contain the THBS2‐binding site. Regrettably, the directly binding between THBS2 and β‐catenin promoter was not experimental validated using the chromatin immunoprecipitation (ChIP) assay, which needs to be followed up by future research. Although we revealed that THBS2 binds to β‐catenin using Co‐IP assay. However, whether other proteins (“scaffold”) are required for THBS2 binding to β‐catenin is unclear. The binding condition in vivo might not be modeled well with Co‐IP assay. Pull‐down combination with Mass spectrometry will be applied to address this issue in follow‐up study since the IP experiment cannot determine whether proteins bind directly or indirectly.

Finally, our research demonstrated that THBS2 activates the Wnt/β‐catenin pathway to regulate growth, migration, and invasion of CRC cells. To further verify the role of THBS2 in inducing CRC cell progression by β‐catenin nuclear translocation, we used the β‐catenin inhibitor, XAV‐939. Our data showed that the suppressive effect of XAV‐939 could be reversed by exogenous overexpression of THBS2. However, this study has certain limitations. The clinical sample sizes were insufficient, and therefore a large sample size will be needed to further verify this conclusion. Furthermore, a corresponding pulmonary metastasis mouse model should be constructed to verify the role of THBS2 in vivo.

In summary, our results highlight the crucial role of THBS2 in the regulation of migration and invasion in CRC cells by activating the Wnt/β‐catenin signaling pathway. Our study is the first study that elucidated the specific region of β‐catenin promoter, to which THBS2 directly bind to transcriptionally activate the β‐catenin, resulting in activation of the Wnt/β‐catenin signaling pathway. THBS2 could be a prognostic marker and an effective therapeutic target for CRC.

CONFLICT OF INTEREST

All authors declare no conflict of interest.

Qu H‐L, Hasen G‐W, Hou Y‐Y, Zhang C‐X. THBS2 promotes cell migration and invasion in colorectal cancer via modulating Wnt/β‐catenin signaling pathway. Kaohsiung J Med Sci. 2022;38:469–478. 10.1002/kjm2.12528