The Potential Role Of Hypoxia-induced DDR2in The Progression Of Breast Cancer

Discoidin domain receptor2(DDR2) is a receptor tyrosine kinase (RTK) thatbinds collagen, especially the native fiber collagen, as its ligands. DDR2highlyexpressed in the tissues of proliferativity and ductility, such as embryo, brain,liver, skin, connective tissue and so on. Previous work suggested that DDR2wasupregulated in pulmonary carcinoma and ovarian cancer. As the main function ofDDR2is to stimulate production of matrix metalloproteinase (MMP) which candigest the ECM proteins and promote cell migration, we speculated that DDR2may be induced by the changes of the tumor microenvironment, and the disorderof tumor structure may render the possibility that DDR2binds to its ligand,collagens, which is the main component of intercellular substance in tumor tissue.The upregulated and activated DDR2may play a role in tumor progression.Therefore, our present study focused on the relationship of DDR2and cancer.Firstly, we discovered that the mRNA and protein level of DDR2were bothsignificantly higher in breast cancer tissues samples than that in normal samples. Then, we detected the expression of DDR2in the specimens of breast cancerpatients by immunohistochemistry, and found that DDR2is highly expressed inductel carcinoma than in fibrocystic disease of breast. However, the DDR2levelin the breast cancer cell lines was not exactly high. This contradiction promotedthat the induction of DDR2in breast cancer may relate to some alteration ofmicroenvironment.It is reported that hypoxia could induce DDR2in vascular smooth musclecells. Hypoxia is not only the principal character of tumor microenvironment, butalso the promoter of malignant transformation. We suppose that DDR2could beinduced by hypoxia in cancer cells. To validate the presume, we treated severalcancer cell lines with hypoxia and its mimietics(CoCl2). By using quantitive PCRand Western Blot, we found that DDR2could be obviously up-regulated both atmRNA and protein level with the time going by hypoxia. And this was auniversal phenomenon in almost all the cancer cell lines which we used. And thedata further showed that there existed a certain relationship between hypoxiastress and DDR2, but not DDR1. Since the function of DDR2is relied on itsphosphorylation, we detected if hypoxia could also influence the activity ofDDR2. The results showed that hypoxia could not change the activity of DDR2.Nevertheless, the activation of DDR2could be indirectly up-regulated by thebinding of collagens with more DDR2induced by hypoxia.The previous study had convinced that DDR2involved in the tumor hypoxiastress. Then we focused on the function of DDR2under hypoxia condition incancer cells. We constructed RNAi lentivirus targeting DDR2. By MTT, we cansee the decreased proliferation rate of the cells in which DDR2had been silencedon the hypoxia. By Transwell and Millicell experiment, we can see the ability ofcell migration and invasion were both increased on the hypoxia condition, but decreased when DDR2were being silenced. And Zymography assays furtherindicated that those changes of cell were associated with a production of MMP-2which digested the ECM. To determine the in vivo effects of DDR2on tumorprogression, we generated two tagged cells which were orthotopically injectedinto the mammary fat pads of nude mice and the distribution of the labeled cellswas examined. Knock down DDR2had a minimal effect on the development ofmetastases, further confirming that the decreased metastatic potential wasspecifically due to the inhibition of DDR2expression and supporting thehypothesis that DDR2acts in the early steps of metastasis. Additionally, theshDDR2signifcantly enhanced the survival of the implanted mice. AlthoughDDR2promoted cell proliferation in vitro, we did not find any correlationbetween tumor size and DDR2expression levels.EMT is thought to promote cancer cell migration and invasion, and hypoxiais an independent factor to trigger EMT. As DDR2is responsible for the increasein cell migration and invasion under hypoxia, next we sought to determinewhether EMT is induced in the cells in which DDR2expression is changed inhypoxia. MDA-MB-468cells lost cell contacts and grew more independently inhypoxia. The level of mesenchymal markers like vimentin and N-cadherin wereincreased, while the level of epithelial marker like E-cadherin was decreased asdetermined by qPCR and Western Blot. These hypoxia-induced changes of EMTcan be significantly inhibited by DDR2-specific RNAi, especially the expressionof E-cadherin.To further test the role of DDR2in hypoxia-induced cell EMT, we examinedthe transcriptional control of DDR2on E-cadherin. The results demonstrated thatDDR2could repress the promoter activity of E-cadherin60%approximately.Howeverf, when the binding site of the transcriptional factor Snail was mutated, the inducing repression abolished. These indicated that the core promoter regionof E-cadherin, to which Snail binds, is responsible for the transcriptional controlof E-cadherin by DDR2. Furthermore, to investigate the protein kinases involvedin signaling downstream in hypoxia-activated DDR2, we examined thephosphorylation level of several protein kinases, including ERK, p38, c-JunN-terminal kinase (JNK), and phosphoinositide3-kinase (PI3K). And we foundthat the ERK MAPK cascade participates in the signaling pathway ofhypoxia-regulated DDR2activation in breast cancer cells.In summary, our results show that DDR2expression may be induced by hypoxiain epithelial cancer. And DDR2is involved in hypoxia-induced cell proliferation,migration, Matrigel invasion, and ECM digestion by MMP-2. Moreover, DDR2is required for hypoxia-induced EMT, which is promoted by transcriptionalrepression of E-cadherin via transcriptional factor Snail.