CAFs, as one of the most abundant components of tumor microenvironment, were involved in tumor response to chemoradiotherapy. Our study found CAFs conferred ESCC cells EMT-like phenotype. EMT is generally observed during embryonic development or in tissue repair processes in adult organs [26]. When this process was activated in human cancers, it was often associated with decreased sensitivity to antitumor therapies and increased cancer cell migration and invasion [27]. Recently, several reports demonstrated that EMT is not a binary process [28]. Different cancer cell populations presented different degrees of EMT in response to external stimuli. The EMT state of cancer cells was influenced by mangy factors and the precise mechanisms were not completely clear. Herein, our study also confirmed the diversity in EMT state in esophageal cancer cells which were cultured in CAFs medium. Some cancer cells had the epithelial marker E-cadherin down-regulated while the mesenchymal marker vimentin was not changed. Although E-cadherin was down-regulated in cancer cells, the level of vimentin showed different degrees of increase. There results suggested that there were several EMT states in esophageal cancer cells which are cultured in CAFs medium.
Chunyu Zhang et al reported that CAFs from ESCC patients had many deregulated genes which were responsible for cell proliferation, extracellular matrix remodeling and immune response in comparison with their matched NFs [29]. Their study observed that CAFs promoted the growth, migration and invasion of esophageal cancer cells in vitro. However, the underlying mechanisms by which CAFs influenced tumor progression were not explored in their study. Previous studies showed that CAFs modulated EMT phenotype of cancer cells by regulating the expressions of EMT-related genes [30–32]. There were many genes deregulated such as TGF-β1 in CAFs compared with in NFs in ESCC [29]. Our previous study found that either intracellular or extracellular TGFβ1 signaling was increased in CAFs compared with in NFs in ESCC [13]. When TGFβ1 signaling was blocked with its receptor inhibitor LY2157299, CAFs-induced EMT phenotype of esophageal cancer cells was significantly reversed. TGFβ1 could activate and phosphorylate Smad 2 and Smad 3 upon binding to its receptor [33]. Activated Smad2 and Smad3 form a complex with Smad 4 and then translocate into the nucleus and regulate the transcription of target genes such as snail, slug and twist, which are inducers of EMT by repressing the transcription of E-cadherin by binding to the E-box sequences within the promoter region. Our data showed that CAFs-secreted TGFβ1 signaling promoted the phosphorylation of Smad 2 and Smad 3 and up-regulated slug to induce EMT of esophageal cancer cells. However, the other EMT-inducers such as snail and twist were not changed. This difference is maybe due to that the Smad complex had low affinity with the promoters of snail and twist. Therefore, the transcription of snail and twist was not initiated in esophageal cancer cells in response to CAFs medium. By induction of EMT, CAFs promoted the migration and invasion, drug resistance and cancer stemness traits of esophageal cancer cells. Therefore, the reversal of EMT would inhibit CAFs-induced malignant phenotype of esophageal cancer cells.
It is known that activated NF-κB signaling was capable of initiating the transcription of over 400 target genes which were involved in immunoregulation, growth regulation, inflammation, carcinogenesis, and cell survival [24]. Several studies demonstrated that the activation of NF-κB signaling diminished the apoptotic action of chemotherapeutic agents and ionizing radiation and thus was an important target to reverse therapy resistance of human cancers [34]. Julie G. Izzo et al. reported that NF-κB signaling was constitutively activated in ESCC patients and significantly associated with poor treatment outcome after chemoradiotherapy [25]. Bin Li et al confirmed that the repression of NF-κB signaling significantly inhibited esophageal cancer cell proliferation and promoted apoptosis and enhanced cell sensitivity to 5-fluorouracil and cisplatin [35]. In our study, we uncovered novel molecular mechanisms of NF-κB signaling activation. We found CAFs-secreted TGFβ1 signaling significantly enhanced the activity of NF-κB transcription factor by degradation of its inhibitor IκВα in esophageal cancer cells. Furthermore, CAFs-secreted TGFβ1 signaling also changed the expressions of target genes downstream of NF-κB signaling with anti-apoptotic Bcl-2 and survivin up-regulated and proapoptotic Bax down-regulated. Recently, Shumei Song et al reported that AT101, an inhibitor of Bcl-2, demonstrated potent anti-tumour activity by targeting cancer stem cells pathways in gastroesophageal carcinoma [36]. These results highlighted Bcl-2 as an attractive anti-cancer target in gastroesophageal carcinoma. TGFβ1 activated NF-κB signaling pathway in a Smad-dependent or independent manner [37]. In response to TGFβ1, Smad3 can initiate the activation of NF-κB signaling pathway by physically interacting with NF-κB or its activator IKKα. Furthermore, TGFβ1 can induce the acetylation of p65/ RelA to enhance the activation of the DNA-binding and transcription activity of NF-kB in a Smad-dependent manner. TGFβ1 can also activate NF-kB in a Smad-independent manner. In a subset of head and neck cancers, TGFβ1 activated NF-kB signaling pathway by TAK1 [38]. Once activated by TGFβ1, TAK1 proceeded to phosphorylate IKKα, leading to NF-kB signaling activation. Our study demonstrated CAFs-secreted TGFβ1 did not alter TAK1 in esophageal cancer cells. We therefore proposed that CAFs-secreted TGFβ1 induced the activation of NF-kB signaling activation in a Smad-dependent manner in esophageal cancer.
As demonstrated in our previous study, the co-culture of CAFs and esophageal cancer cells induced the activation of TGFβ1 signaling via an autocrine/paracrine loop which may further promote the EMT phenotype of esophageal cancer cells. Therefore, we proposed that TGFβ1 signaling was an important target to reverse the tumor-promoting effect of CAFs in esophageal cancer. However, it still needs to be explored how TGFβ1 signaling was activated in our future study. Only the complete understanding of CAFs’ action mechanisms would accelerate the development of therapeutic strategies against esophageal cancer.