High Wnt2 Expression Confers Poor Prognosis in Colorectal Cancer, and Represents a Novel Therapeutic Target in BRAF-Mutated Colorectal Cancer

Background and Objectives: We aimed to investigate the role of Wnt2 expression in colorectal cancer (CRC) prognosis and evaluate its potential as a therapeutic target in BRAF-mutated CRC. Materials and Methods: Exactly 136 samples of formalin-fixed paraffin-embedded CRC tissue specimens were obtained from patients who underwent surgical resection for CRC. The gene mutation status of the samples was detected using fluorescence PCR. Wnt2 expression was detected using immunohistochemistry. Survival curves with high Wnt2 expression and BRAF mutations were compared using the Kaplan–Meier method. A nomogram was constructed to determine the estimated overall survival probability. We also predicted the 3-year and 5-year survival rates for patients with high Wnt2 expression and BRAF mutations. In total, 50 samples of BRAF-mutated CRC were collected and detected Wnt2 expression by immunohistochemistry. The Chi-squared test was used to analyze the association between Wnt2 expression and BRAF-mutated CRC. Results: High Wnt2 expression and BRAF mutations are associated with poor prognosis of CRC. Multivariate survival analyses indicated that high Wnt2 expression and BRAF mutations are significant independent predictors of CRC prognosis. Furthermore, high Wnt2 expression was significantly associated with BRAF-mutated CRC, and Wnt2 may be a potential therapeutic target for BRAF-mutated CRC. Conclusions: High Wnt2 expression confers poor prognosis in colorectal cancer and represents a novel therapeutic target in BRAF-mutated CRC.


Introduction
Colorectal cancer (CRC) is one of the most common malignant gastrointestinal tumors. A White Paper on Epidemiology, Prevention, and Screening of Colorectal Cancer in China shows that CRC ranks third in the incidence of malignant tumors. According to the 2019 China Health Statistical Yearbook, CRC ranks fifth among the top 10 malignant tumor mortality rates in China, with a mortality rate of 7.25 (1/100,000), including 8. 19 (1/100,000) for males and 6.26 (1/100,000) for females. By 2030, there will be more than 2.2 million newly diagnosed cases and 1.1 million deaths, and the global CRC burden is estimated to increase by approximately 60% [1]. As the early detection of CRC is still challenging, most patients are diagnosed at an advanced stage [2]. Although significant progress has been made in treating advanced CRC in recent years, the 3-year and 5-year survival rates remain very low [3]. Therefore, potentially effective molecular biomarkers and treatment indicators for CRC diagnosis and prognosis are needed. These can help with the early detection, monitoring, and treatment monitoring of patients with CRC, improve their prognosis and survival, and promote the development of personalized treatment.
Wnt2, a member of the Wnt family located on human chromosome 7q31, encodes a secretory protein that regulates the necessary developmental process and plays a critical role in tumorigenesis [4]. Increased Wnt2 expression has been found in human fetal lungs and the placenta, but rarely in the normal gastrointestinal tract [5]. An increasing number of studies has shown that the expression of Wnt2 is abnormally high in various cancers, including fibroadenoma [6,7], breast cancer [8][9][10], and gastric cancer [11]. Compared with normal stroma [12], Wnt2 is significantly overexpressed in CRC and promotes cancer cell invasion and metastasis by activating the Wnt signaling pathway. It is essential in maintaining the activated CAF (cancer-associated fibroblast) phenotype, which promotes angiogenesis [13]. It has also been found that CAFs-secreted Wnt2 suppresses the DCmediated antitumour T-cell response via the SOCS3/p-JAK2/p-STAT3 signalling cascades, thereby suppressing antitumor immunity [14].
Tumor cells accumulate genetic and epigenetic changes during growth [15]. They are composed of thousands of non-synonymous mutations that are positively associated with the occurrence and development of cancer [16]. Only a subset of these mutations can often be used as driving mutations, whereas the rest are considered random passenger mutations. These mutations accumulate and change the course of the tumors [17]. As the primary driving mutations, BRAF mutations are involved in the occurrence and development of various cancers. Nearly 60% of melanomas show BRAF mutations [18], and mutations in this gene are also found in non-Hodgkin's lymphoma [19], CRC [20,21], thyroid papillary carcinoma [22,23], non-small-cell lung cancer [24][25][26], glioblastoma [27], and inflammatory diseases [28,29]. BRAF mutations, are widely believed to be associated with poor prognosis in CRC [30]. Elaine Tan et al. found that patients with BRAF mutations had worse OS (Overall Survival) compared with the wild type with a median survival of 18.9 months versus 33.2 months [31].
This study aimed to investigate the clinical implications of Wnt2 expression and BRAF mutations in the prognosis of CRC. Moreover, we investigated the relationship between Wnt2 and several key genes in CRC.

Clinical CRC Specimen and Clinical Information
This study was approved by the Ethics Committee of the Affiliated Hospital of Jiangnan University. In total, 136 patients diagnosed with CRC who underwent surgical resection at the Affiliated Hospital of Jiangnan University between June 2014 and November 2016 were selected retrospectively. Overall, 50 samples of BRAF-mutated CRC were retrospectively screened from January 2010 to December 2019 at the Affiliated Hospital of Jiangnan University. All patients underwent mutation testing. The exclusion criteria were as follows: (1) receiving neoadjuvant antitumor therapy such as radiotherapy, chemotherapy, or immunotherapy before surgery; (2) the type of multiple tumors or metastases is unclear; (3) the patients' personal information is incomplete or lost to follow-up; (4) a history of other malignant tumors.
The clinicopathological features of all patients were analyzed and recorded in detail by two clinicians, and details regarding prognosis were obtained by telephone follow-up.

Immunohistochemical Analysis
The CRC tissue samples were fixed in 4% formalin, embedded in paraffin, and sectioned. The tissue sections were subjected to immunohistochemical analysis, deparaffinized in xylene, and hydrated with graded ethanol. Then, they were heated in sodium citrate buffer for 30 min at 100 • C for antigen repair and incubated with 3% hydrogen peroxide to block endogenous peroxidase activity. After blocking for 30 min, the cells were incubated with Wnt2 antibody (1:100, Abcam, ab150608, Hong Kong, China) overnight at 4 • C. Subsequently, the sections were washed with phosphate buffer salt solution and incubated with an amplifying agent, polymerase (reagent A, GTVisionTM Kit, Shanghai, China), and 3, 3 -diaminobenzidine chrome developer (DAB, reagent B, C; GTVisionTM kit, Shanghai, China) for 1-3 min. Then, they were stained with hematoxylin for 60s and examined under a microscope after dehydration and sealing.
The intensity and degree of Wnt2 immunostaining were semi-quantitatively analyzed, and the percentage of tumor cells with positive staining and intensity of staining were scored. According to the German semi-quantitative scoring system standards, each section was individually read by two associate chief physicians from the Department of Pathology, and the results were averaged.

Mutation Detection
The samples were tested for mutations using five mutant gene detection kits (AmoyDx, Xiamen, China) and the human PIK3CA gene Mutation Detection Kit (AmoyDx, Xiamen, China), according to the manufacturer's instructions. The samples were 136 FFPE specimens from postoperative CRC patients. DNA extraction of the samples was performed using a paraffin tissue DNA extraction kit (TianGen, Beijing, China), the procedure is as follows:30 mg of paraffin tissue was collected, deparaffinized by shaking with xylene, and centrifuged at 12,000× g for 2 min at room temperature. The supernatant was discarded, and absolute ethanol was added and mixed by shaking. The mixture was centrifuged at 12,000× g for 2 min at room temperature, and the supernatant was discarded. It was allowed to stand for 5 min to fully volatilize the ethanol. An additional 200 µL GA buffer and 20 µL Proteinase K were added, mixed, and incubated at 56 • C for 1 h until the sample was completely lysed. After further incubation at 90 • C for 1 h, 220 µL GB buffer was added and mixed, then 250 µL absolute ethanol was added and mixed. The liquid was put into the adsorption column at 8000 rmp and centrifuged at room temperature for 2 min. After the waste liquid was discarded, 500 µL GD buffer was added to the adsorption column CR2 and centrifuged at 8000 rpm for 60 s. Discard the waste solution and repeat twice. The adsorption column was opened for 5 min, and then 50 µL of TE eluate prewarmed at 65 • C was added. Finally, DNA was collected.
Tissue samples were tested for mutations by Q-PCR with the five mutant gene detection kits, the procedure is as follows: the DNA of the extracted tissue samples was mixed with LMG mixed enzyme in the kit, and then the machine was detected. The reaction program is as follows: first stage, 42 • C for 5 min, 95 • C for 5 min, a cycle; second stage, 95 • C 25 s, 64 • C 20 s, 72 • C 20 s, 10 cycles; third stage: 93 • C 25 s, 60 • C 35 s, 72 • C 20 s, 36 cycles. The detection sites are as follows (Table 1):

Statistical Analysis
Statistical analysis was performed using the R version 3.5.3 software (version 3.5.3, http://www.r-project.org, accessed on 11 March 2019). Either the Chi-squared or Fisher's exact test was used to assess the association between Wnt2 expression and clinicopathological features. Moreover, Kaplan-Meier survival analysis was used to draw the overall survival (OS) curve, and the log-rank test was used to compare the differences. The Cox proportional hazards regression model was used for univariate and multivariate analyses. The Chi-squared test was used to analyze the association between Wnt2 expression and BRAF-mutated CRC. p < 0.05 was considered statistically significant.

Exon
Base Changes

Clinicopathological Features of the Patients with CRC
This study included 63 men and 73 women. Among the enrolled patients, 34 were under 60 years old, and 102 were over 60. There were 35 patients in the T1-2 stages, 101 in the T3-4 stages, 87 in the N0 stage, 49 in the N1 and N2 stages, 113 in the M0 stage, and 23 in the M1 stage. ERBB2 mutations were found in 132 patients, KRAS mutations in 83, BRAF mutations in 14, and PIK3CA mutations in 112. The clinicopathological features of the patients with CRC are shown in Table 2.

High Wnt2 Expression and BRAF Mutations Are Associated with Poor Prognosis in Patients with CRC
To determine the expression of Wnt2 in CRC, we evaluated its expression levels in the tissues of the 136 patients with CRC using immunohistochemical analysis. Immunohistochemistry showed that Wnt2-positive staining was mainly confined to the cytoplasm of the CRC cells ( Figure 1A,B). Depending on the staining intensity, samples were divided into groups with high Wnt2 expression and low Wnt2 expression. The number of patients with high Wnt2 expression was 37 (27.21%), and the number of those with low Wnt2 expression was 99 (72.79%).
In addition, we evaluated the association between Wnt2 expression and clinicopathological features, including age, sex, T stage, N stage, TNM stage, and mutation status, in different groups of patients with CRC. The difference in Wnt2 protein expression (high vs. low) was significantly correlated with BRAF mutation status in patients with CRC (p = 0.0001), as shown in Table 2.
Next, we investigated whether high Wnt2 expression and BRAF mutations are associated with CRC prognosis. The Kaplan-Meier analysis showed that CRCs with high Wnt2 expression had significantly worse OS than those with low Wnt2 expression (p = 0.00033, Figure 2A). In addition, CRCs with BRAF mutations also had significantly lower OS than those with the wild type (p < 0.0001, Figure 2B).
Univariate analysis showed that Wnt2 (p < 0.0001), T stage (p = 0.008), M stage (p < 0.0001), N stage (p < 0.0001), TNM stage (p < 0.0001), and BRAF mutation status (p < 0.0001) were significantly associated with poor prognosis in CRC (Table 3). Subsequently, all important variables such as Wnt2, T stage, M stage, N stage, TNM stage, and BRAF mutation status were entered into the multivariate Cox proportional hazards model, and the results showed that Wnt2 expression (p = 0.035) and BRAF mutations (p < 0.0001) were prognostic factors for poor OS in CRC (Table 3).  A nomogram was constructed to predict the 3-and 5-year survival rates. A score was matched on the score table based on the value of each factor, and the score of each factor was added to obtain the total predicted score to predict OS at 3 and 5 years ( Figure 3A). BRAF mutations and M stage were the most important prognostic factors, followed by TNM stage, T stage, Wnt2 expression, and ERBB2 mutations. The nomogram validation included calibration of the primary and validation cohorts. Moreover, the Kaplan-Meier plot survival probabilities were used for the calibration. There was an agreement between the nomogram prediction and the actual OS at 3 and 5 years ( Figure 3B).

High Wnt2 Expression and BRAF Mutations Are Associated with Poor Prognosis in Patients with CRC
To determine the expression of Wnt2 in CRC, we evaluated its expression levels in the tissues of the 136 patients with CRC using immunohistochemical analysis. Immunohistochemistry showed that Wnt2-positive staining was mainly confined to the cytoplasm of the CRC cells ( Figure 1A,B). Depending on the staining intensity, samples were divided into groups with high Wnt2 expression and low Wnt2 expression. The number of patients with high Wnt2 expression was 37 (27.21%), and the number of those with low Wnt2 expression was 99 (72.79%). In addition, we evaluated the association between Wnt2 expression and clinicopathological features, including age, sex, T stage, N stage, TNM stage, and mutation status, in different groups of patients with CRC. The difference in Wnt2 protein expression (high vs. low) was significantly correlated with BRAF mutation status in patients with CRC (p = 0.0001), as shown in Table 2.
Next, we investigated whether high Wnt2 expression and BRAF mutations are associated with CRC prognosis. The Kaplan-Meier analysis showed that CRCs with high Wnt2 expression had significantly worse OS than those with low Wnt2 expression (p = 0.00033, Figure 2A). In addition, CRCs with BRAF mutations also had significantly lower OS than those with the wild type (p < 0.0001, Figure 2B).
Univariate analysis showed that Wnt2 (p < 0.0001), T stage (p = 0.008), M stage (p < 0.0001), N stage (p < 0.0001), TNM stage (p < 0.0001), and BRAF mutation status (p < 0.0001) were significantly associated with poor prognosis in CRC (Table 3). Subsequently, all important variables such as Wnt2, T stage, M stage, N stage, TNM stage, and BRAF mutation status were entered into the multivariate Cox proportional hazards model, and the results showed that Wnt2 expression (p = 0.035) and BRAF mutations (p < 0.0001) were prognostic factors for poor OS in CRC (Table 3).   factor was added to obtain the total predicted score to predict OS at 3 and 5 years (Figure 3A). BRAF mutations and M stage were the most important prognostic factors, followed by TNM stage, T stage, Wnt2 expression, and ERBB2 mutations. The nomogram validation included calibration of the primary and validation cohorts. Moreover, the Kaplan-Meier plot survival probabilities were used for the calibration. There was an agreement between the nomogram prediction and the actual OS at 3 and 5 years ( Figure  3B).

Wnt2 Expression Was Associated with BRAF-Mutated CRC
From the correlation analysis in Table 2, Wnt2 expression was positively correlated with BRAF mutation status. Considering the insufficient number of BRAF-mutated CRC, we collected 50 paraffin specimens from patients with BRAF mutations at the Affiliated Hospital of Jiangnan University from 2010 to 2019. Wnt2 expression was analyzed using immunohistochemistry ( Figure 4A,B). In 50 samples of BRAF-mutated CRC (contains the 14 samples of BRAF-mutated CRC from the previous 136 samples), Wnt2 expression was high in 41 (82%) samples. Compared with 118 samples with wild-type BRAF mutations in the 136 samples, the number of high Wnt2 expression was 26 (22%). Table 4 presents the correlation between BRAF mutations and Wnt2 expression in CRC (p < 0.001). In conclusion, there was a positive correlation between the Wnt2 expression and BRAFmutated CRC, and Wnt2 may be a potential therapeutic target for BRAF-mutated CRC.

Wnt2 Expression Was Associated with BRAF-Mutated CRC
From the correlation analysis in Table 2, Wnt2 expression was positively correlated with BRAF mutation status. Considering the insufficient number of BRAF-mutated CRC, we collected 50 paraffin specimens from patients with BRAF mutations at the Affiliated Hospital of Jiangnan University from 2010 to 2019. Wnt2 expression was analyzed using immunohistochemistry ( Figure 4A,B). In 50 samples of BRAF-mutated CRC (contains the 14 samples of BRAF-mutated CRC from the previous 136 samples), Wnt2 expression was high in 41 (82%) samples. Compared with 118 samples with wild-type BRAF mutations in the 136 samples, the number of high Wnt2 expression was 26 (22%). Table 4 presents the correlation between BRAF mutations and Wnt2 expression in CRC (p < 0.001). In conclusion, there was a positive correlation between the Wnt2 expression and BRAF-mutated CRC, and Wnt2 may be a potential therapeutic target for BRAF-mutated CRC.

Discussion
It is well established that multiple mutations of oncogenes are involved in the process of tissue transformation from normal epithelial cells to carcinomas in CRC [32]. BRAF, which regulates the Ras-Raf-MEK-ERK pathway and is associated with poor prognosis in CRC, was selected as a key driver, even though its mutation rate was slightly lower than that of other genes. BRAF mutations are also significantly associated with distant metastasis in Asian populations [33]. The prognosis of patients with BRAFmutated CRC remains poor due to recurrence and metastasis [34]. Moreover, discovering the BRAF status is critical for understanding patient survival and prognosis. Therefore, there is an urgent need to define the molecular mechanisms of carcinogenesis and identify novel therapeutic targets for BRAF-mutated CRC. We analyzed the relationship between Wnt2 expression and KRAS-2, BRAF-15, PIK3CA-9, PIK3CA-20, and ERBB2-20.

Discussion
It is well established that multiple mutations of oncogenes are involved in the process of tissue transformation from normal epithelial cells to carcinomas in CRC [32]. BRAF, which regulates the Ras-Raf-MEK-ERK pathway and is associated with poor prognosis in CRC, was selected as a key driver, even though its mutation rate was slightly lower than that of other genes. BRAF mutations are also significantly associated with distant metastasis in Asian populations [33]. The prognosis of patients with BRAF-mutated CRC remains poor due to recurrence and metastasis [34]. Moreover, discovering the BRAF status is critical for understanding patient survival and prognosis. Therefore, there is an urgent need to define the molecular mechanisms of carcinogenesis and identify novel therapeutic targets for BRAF-mutated CRC. We analyzed the relationship between Wnt2 expression and KRAS-2, BRAF-15, PIK3CA-9, PIK3CA-20, and ERBB2-20. Statistical analysis revealed that high Wnt2 expression (p = 0.035) and BRAF mutations (p < 0.0001) were closely related to poor survival in patients with CRC. Furthermore, high Wnt2 expression was significantly correlated with BRAF mutations in CRC. This finding provides a new possible therapeutic target for the treatment of BRAF-mutated CRC.
Wnt2 can increase the metastasis and invasion of fibroblasts and promote angiogenesis in CRC [35]. Meanwhile, upstream regulatory molecules of the Wnt signaling pathway are involved in multidrug resistance. In our study, Wnt2 was found to be positively correlated with BRAF mutations; it is important to explore whether high Wnt2 expression promotes invasion and metastasis and regulates drug resistance in patients with BRAF mutations.
In a recent study, anti-Wnt2 mAb was found to significantly restore intratumoral anti-tumor T cell responses and enhance the efficacy of anti-PD-1 by increasing active DCS in both mouse OSCC and CRC syngeneic tumour models. Direct interference with CAFderived Wnt2 restored DC differentiation and DC-mediated antitumor T cell responses [14]. Combined with our experimental results, we consider whether targeting Wnt2 can enhance the efficacy of ICI in BRAF-mutanted CRC. The specific results need to be further explored. The specific mechanism of the significantly high Wnt2 expression in BRAF-mutated CRC is still unclear. One study exploring the treatment of BRAF-mutated CRC using a xenograft model, found that both pyrvinium and axitinib were able to significantly increase the ability of vemurafenib to attenuate tumor growth in xenografts of BRAF-mutated colorectal cancer cells. Thus, this also demonstrated that Wnt treatment also has non-immune-specific effects on BRAF-mutated CRCs [36].
This study is the first to report that Wnt2 is significantly higher in BRAF-mutated CRC tissues than in wild-type tissues, which provides a new approach for exploring the development and treatment of patients with BRAF mutations.
Our study also has some limitations. Firstly, this study is limited by small sample size and single-center data; we did not separate the TNM stages for I, II, III and IV. In future studies, a multi-center large-scale study will be conducted. Secondly, the molecular mechanisms linking Wnt2 and BRAF mutations require further exploration.

Conclusions
The results of this study suggest that high Wnt2 expression and BRAF mutations could be used as new predictors of CRC prognosis. In addition, high Wnt2 expression and BRAF mutations may be potential new therapeutic targets and have important implications in the future treatment of CRC. Wnt2 expression is significantly higher in BRAF-mutated CRC than the wild-type, and Wnt2 can be used as a new therapeutic target for BRAF-mutated CRC patients.

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Affiliated Hospital of Jiangnan University Medical Foundation for studies involving humans. All methods were performed in accordance with approved guidelines; written informed consent was waived by the Institutional Affiliated Hospital of Jiangnan University Medical Foundation due to the retrospective design of this study.
Informed Consent Statement: Patient consent was waived due to the retrospective design of this study.

Data Availability Statement:
The data presented in this study are available in this article.

Acknowledgments:
We would like to express our gratitude to the pathologists Yankui Liu and Zhidang Yin, who involved in the WNT2 expression scoring, for their help in our study.

Conflicts of Interest:
The authors declare no conflict of interest.

Abbreviations
CRC colorectal cancer OS overall survival