The Mechanism Of JWA Inhibits Tumor Angiogenesis In Gastric Cancer

Gastric cancer is the fifth most common cancer and the third largest cause of cancer death worldwide. Significant treatment achievements have been made for the early stage gastric cancer patients, however, due to the peritoneum, lymph nodes and distant organ metastasis, the5-year survival rate of gastric cancer patients at advanced stage is still approximately15%despite aggressive treatments. Metastasis reduces the prognosis of cancer patients and is responsible for approximately90%of all cancer-related deaths. Tumor angiogenesis which provides a shortcut for tumor cell metastasis is considered to play an important role in metastasis. Tumor-associated angiogenesis has traditionally been defined as the sprouting of new vessels from preexisting vessels, including angiogenic growth factors activate receptors on endothelial cells present in pre-existing blood vessels; proteases degrade the basement membrane to allow endothelial cells to escape from the original (parent) vessel walls; the endothelial cells then proliferate into the surrounding matrix and form solid sprouts connecting neighboring vessels, endothelial cells migrate in tandem, these sprouts then form loops to become a full-fledged vessel lumen. Little is known about the molecular events critical to the gastric cancer angiogenesis, although a lot of studies have aimed directly to other tumor angiogenesis.JWA, a multifunctional microtubule-binding protein, was demonstrated to be indispensable for the rearrangement of actin cytoskeleton and activation of MAPK cascades to inhibit cell migration. Meanwhile, we also find that JWA plays an important role in regulating melanoma metastasis via inhibition of MMP-2and integrin avβ3. Furthermore, our recent retrospective data has shown that JWA expression is significantly down-regulated in gastric cancer compared with normal gastric tissues and negatively associated with lymph node metastasis, distant metastasis, TNM stage and overall survival of patients; JWA low expression was an unfavourable factor for the prognosis of gastric cancer patients.Purpose:In this study, we will go on studying the role and mechanism of JWA in gastric cancer angiogenesis which is the pivotal step of gastric cancer metastasis.Methods:Western blot was used to test the expressions of MMP-2, JWA and CD31in fresh gastric cancer and the corresponding normal gastric mucosa tissues. Simultaneously, two independent retrospective gastric cancer sets were studied, and immunohistochemisty was used to detect the expression of JWA, MMP-2and CD31+intratumoral microvessel density (MVD) in the gastric cancer tissue microarrays (TMAs). The association between MMP-2expression and clinicopathological parameters was evaluated by Fisher’s exact test. The significance of MMP-2staining in primary tumors compared with their corresponding non-tumors was assessed by the paired Wilcoxon test (raw scores). Analysis of variance (ANOVA) was used to test the significance of the number of MVD among different groups. Probability of differences in OS as a function of time was ascertained by use of the Kaplan-Meier method, with a log-rank test probe for significance. Univariate and multivariate COX proportional hazards regression analysis were performed to estimate the crude hazard ratios (HRs), adjusted HRs and95%CI of HRs. JWA in the gastric cancer cell lines SGC7901and HGC-27was over-expressed or knocked down, then the conditioned medium surposing to contain secret cytokines was collected from these overexpressed or knockdown gastric cancer cells and used for HUVECs growth and tube formation assays. Moreover, in vivo matrigel plug and Chick embryo chorioallantoic membrane assay (CAM) assays were used to further investigate whether JWA expression in gastric cancer cells could inhibit the neovessel formation in nude mice and chick embryo. Western blot, real time PCR and EMSA assays were used to identify how JWA regulate the expression of Spl and MMP-2.Results:(1) Ten pairs of human gastric cancer fresh samples, including primary gastric cancerous tissues and matched normal gastric mucosa were analyzed by Western blot. The results showed that reduced JWA and increased MMP-2and CD31expression were found in9out of10gastric tumors compared with the paired normal gastric mucosa. Using our established gastric cancer TMA, we further confirmed expression of these three molecules by immunohistochemical staining in the training cohort. We found that CD31represented number of intratumoral MVD was dramatically increased compared with the ones in normal tissue and the expression of MMP-2was significantly increased in44of80(55%) tumors compared with the matched normal gastric tissues (P<0.001). The ANOVA analysis indicated that MVD (CD31+) was negatively associated with JWA expression but positively with MMP-2expression in GC tissues. Moreover, when combined these two molecules as a variable, we found that the MVD in the group of low JWA and high MMP-2expression were significantly increased compared with the group of either high JWA or low MMP-2expression.(2) Fisher exact analysis revealed that protein expression of MMP-2in cancerous tissues of the two cohorts was positively associated with lymph node metastasis (P=0.014) and TNM stages (P=0.003), but negatively correlated with JWA expression (P<0.001). Our data revealed that the patients with high MMP-2expression had significantly poorer OS in these two gastric cancer cohorts. When these two markers were combined as a new variable, we found that the cases with low JWA and high MMP-2expression had the worse outcome of survival in both cohorts (P<0.001). Next, we performed univariate and multivariate COX regression analysis to study the effects of both factors on patient survival in both GC sets. In multivariate Cox regression analysis, our results also indicated that MMP-2expression was an independent prognostic factor for OS (HR=4.08, P<0.001in the training cohort; HR=1.94, P<0.001in the validation cohort). When JWA and MMP-2expression were combined, low JWA and high MMP-2was a more unfavorable prognostic factor than JWA or MMP-2alone for patients with surgery alone (HR=7.75, P<0.001in the training cohort; HR=2.31, P<0.001in the validation cohort).(3) To test the effect of JWA expression in gastric cancer angiogenesis, we knocked down or overexpressed JWA in SGC7901and HGC-27cells, respectively. The conditioned medium surposing to contain secret cytokines was collected from these overexpressed or knockdown gastric cancer cells and used for HUVECs tube formation assays. The data showed that the average number of complete tubular structures formed by HUVECs in conditioned medium from JWA knockdown in these two GC cells led to significant increase (P<0.001); whereas decreased dramatically in conditioned medium from SGC7901and HGC-27cells with overexpression of JWA (P<0.001). Then, we performed the in vivo matrigel plug and CAM assays to further validate that JWA expression in gastric cancer cells could inhibit the neovessel formation in nude mice and chick embryo when compared with the corresponding controls.(4) We observed that MMP-2expression was negatively regulated by JWA in both SGC7901and HGC-27cells, respectively. We next examined whether and how JWA and MMP-2was functionally related in angiogenesis. The expression of JWA and MMP-2in SGC7901was knocked down separately or combined, and their conditioned mediums were collected for testing their impact on tube formation of HUVECs. The tube formation assay showed that knocking down MMP-2abolished JWA knockdown-induced tube formation of HUVECs.(5) The data revealed that knockdown or overexpressed JWA in SGC7901and HGC-27cells elevated or downregulated Spl expression in both total and nuclear protein compared with the vector control. At the same time, EMSA was performed to further determine whether JWA affected Spl-binding affinity to a defined Spl-responsive DNA-binding element. Our result demonstrated that JWA knockdown increased the DNA binding activity of Sp1, whereas JWA overexpression decreased this binding, when compared to the respective controls. To confirm the role of Spl in JWA-regulated GC angiogenesis, we used siRNA to knock down either JWA or Spl, or both of them in SGC7901, and then collected the conditioned medium from these cells to test their impact on angiogenesis. The data showed that knockdown JWA enhanced MMP-2expression in SGC7901cells could be effectively blocked by inhibiting Sp1. The tube formation of HUVECs assay also revealed that knockdown JWA induced tubular formation could be abolished by knocking down Spl.(6) To investigate how JWA negatively regulates Sp1expression, we examined Sp1mRNA level which was not changed by JWA knockdown or overexpression. However unexpectedly, we found that the protein levels of Spl were increased in cells of JWA knockdown but decreased in cells of JWA overexpression compared with corresponding control cells. Therefore, we postulated that JWA may affect Spl protein stability. To determine this, SGC7901cells were treated with CHX, an inhibitor of protein synthesis, we found that knockdown JWA suppressed, whereas overexpression of JWA promoted the degradation of Spl. Simultaneously, we found that the ubiquitination of Spl was inhibited in the cells of JWA knockdown treated with MG132compared with non MG132treated control cells.Conclusion:In this study, we connected the gastric cancer cell lines, animals and the tissues of the gastric cancer patients combining with the clinical data to validate that the combination of JWA with MMP-2was a better biomaker for predicting gastric cancer prognosis and JWA was a promising biomarker in regulating gastric cancer angiogenesis. JWA was found to inhibit Spl activity via an ubiquitin-proteasome-dependent mechanism and downregulate the proangiogenic MMP-2expression. Our findings imply that JWA and MMP-2may serve as promising prognostic markers in resectable gastric cancer and JWA as biomarker of angiogenesis in gastric cancer and a therapeutic target by MMP-2modulation.