| Introduction: Gastric cancer(GC), the second leading cause of cancer-related deaths worldwide [1], is considered to result from Helicobacter pylori(H. pylori) infection and subsequent inflammation [2, 3]. By inducing cytokines including IL-6 and IL-8, H. pylori plays important roles in GC development [4], including tumour initiation [5], promotion [6] and metastasis [7]. Generally, IL-6 that activates STAT3 have particularly important roles in the malignant transformation of gastric and intestinal epithelial cells [8]. As a point of convergence for numerous oncogenic signalling pathways, STAT3 activation in diverse human cancers has been shown to increase proliferation, survival, angiogenesis and metastasis and to inhibit anti-tumour immunity [9-11]. This transcription factor is constitutively activated both in tumour cells and in immune cells, and is involved in oncogenesis and inhibition of apoptosis; however, the cause of constitutively active STAT3 in cancer cells has not been fully explored. Cyp B belongs to cyclophilins, which are intracellular receptors for cyclosporin A [12] and possess peptidyl-prolyl isomerase activities that accelerate protein folding [13]. Interestingly, Cyp B is essential for STAT3 activation in cancer cells. Depletion and pharmacologic inhibition of Cyp B caused death through the loss of Janus-activated kinase(JAK)/STAT3 signalling [14]. Besides, Cyp B overexpression was observed in several types of cancer [15-17]. However, the mechanisms underlying dysregulation of Cyp B are not well understood.Micro RNAs(mi RNAs) are important small non-coding RNAs that either inhibit the translation of or trigger the degradation of target m RNAs through binding to the 3-untranslated regions(3’-UTRs) [18, 19]. Our results showed that mi R-520d-5p was a bioinformatically predicted Cyp B regulator and that mi R-520d-5p and Cyp B expression were inversely correlated in GC cell lines and tissues. We thus set out to testify the hypothesis that mi R-520d-5p may partially account for Cyp B upregulation in cancer and inhibit GC growth by suppressing Cyp B.Here we present evidence that a Cyp B/STAT3/mi R-520d-5p feedback loop triggered by IL-6 regulates gastric carcinogenesis. Our study found that Cyp B is required for STAT3 activation in cancer cells and that mi R-520d-5p, an important inhibitory factor of Cyp B, is transcriptionally suppressed by STAT3, thus potentially explaining H. pylori infection and IL-6 stimulation-triggered constitutive STAT3 activation in cancer cells.Materials and Methods: A GC tissue microarray containing 90 cases of GC and paired adjacent non-cancerous tissue was purchased from Shanghai Outdo Biotech. Blood samples from 100 GC patients(without overlap to the cases of tissue array) and 50 healthy volunteers were collected from Xijing Hospital, Xi’an, China. Expression vectors encoding Cyp B were constructed by cloning the open reading frames and downstream 3’-UTR into the pc DNA 3.1 vector(Invitrogen) between Hind III and Eco RI sites for expression driven by the CMV promoter. The sh RNA sequences of Cyp B and STAT3 were amplified and cloned into the GV115 vector(Gene Chem) between Age I and Ecor I sites for expression driven by the CMV promoter. Synthetic mi R-520d-5p mimic, mi R-520d-5p inhibitor, mi R-520d-5p agomir, mi R-520d-5p antagomir and their negative control oligonucleotides were purchased fromRiBoBio(Guangzhou, China). BGC823 and SGC7901 cells were transfected with indicated constructs or oligonucleotides. Target cells were then seeded in 96-well plates at 48 h post-transfection. XTT assays and colony formation assays were conducted to determine the cell growth according to the manufacturer’s instructions. Besides, The Annexin V-FITC Apoptosis Detection Kit was used for apoptosis assays. Total proteins were prepared from fresh frozen tissue or cultured cell samples by complete cell lysis followed by western blotting. Subcutaneous xenograft models were used to determine the in vivo effects of Cyp B/STAT3/mi R-520d-5p feedback loop. Luciferase reporter assays and Ch-IP assays were performed to test if mi R-520d-5p directly bind to the 3’-UTR of Cyp B and if STAT3 binds promoter region of mi R-520 d. Immunofluorescence were used to determine the location of Cyp B and STAT3 upon restoration of mi R-520d-5p after IL-6 treatment. Finally, IHC and ISH were performed on the same tissue microarray to determine the correlation between expression of mi R-520d-5p and Cyp B/STAT3. Results:Cyp B is upregulated in GC tissues and patient sera and is correlated with GC progression. Cyp B was significantly upregulated in GC tissues compared with adjacent non-cancerous gastric tissues. Correlation analysis revealed that high-level Cyp B expression in GC tissues was significantly associated with a more aggressive tumour phenotype. Kaplan–Meier analysis further revealed that high-level Cyp B expression was associated with shorter disease-free survival time for GC patients. Cox regression analysis also indicated that high Cyp B expression was an independent prognostic factor for poor survival in GC patients. We further found that the Cyp B concentrations in serum samples from GC patients were significantly increased compared with those from volunteers. Cyp B silencing inhibits GC cell growth in vitro and in vivo. Western blotting revealed that Cyp B expression was greater in MKN45, SGC7901 and BGC823 cells compared with GES-1 cells. BGC823 and SGC7901 cells were then infected with sh RNA against Cyp B or a control. XTT assays and colony formation assays revealed that cell growth was significantly reduced by Cyp B downregulation compared with the control. Moreover, cell cycle assays showed that silencing Cyp B increased the G0/G1 population and reduced theS and G2/M population compared with control cells. Apoptosis assays revealed that Cyp B inhibition led to an increased percentage of apoptotic GC cells. Furthermore, in vivo analysis showed that silencing Cyp B in BGC823 and SGC7901 cells caused dramatic reductions in tumour weight and volume in nude mice. Together, these results suggest that Cyp B may play an important role in GC cell growth in vitro and in vivo. Cyp B regulates GC growth through activation of STAT3 p STAT3(Tyr705) levels were then determined in the tissue array that was used for Cyp B expression analysis by IHC). Compared with normal tissues, both Cyp B and nuclear p STAT3 levels were increased in GC tissues. Interestingly, after treatment with IL-6, Cyp B redistributed to a cytoplasmic and nuclear localization in approximately 50% or more of the cell population. Besides, Cyp B sh RNA led to a significant reduction in the phosphorylation of JAK2/STAT3 pathway-related proteins without affecting their expression, while Cyp B restoration increased JAK2 and STAT3 phosphorylation. BGC823 cells were co-infected with lentiviral vectors encoding Cyp B and STAT3 sh RNA,colony formation and XTT assays demonstrated that knockdown of STAT3 abrogated the Cyp B-upregulation-induced GC growth promotion. mi R-520d-5p downregulates Cyp B expression by directly binding its 3-UTR. Several mimics of micro RNAs which were previously reported to have tumor suppressive roles in cancer were transfected to BGC823 cells and western blotting analysis revealed that transfection with mi R-520d-5p mimics reduced Cyp B expression. To determine whether mi R-520d-5p represses Cyp B by targeting the potential binding site, PCR products containing either wild-type or mutant Cyp B 3-UTR sequences were cloned downstream of the luciferase open reading frame. mi R-520d-5p overexpression suppressed luciferase activities of the Cyp B 3’-UTR reporter constructs, whereas the effect was abolished when mutation were introduced into its seed sequences. Furthermore, q RT-PCR and western blotting analysis revealed that ectopic mi R-520d-5p expression reduced Cyp B m RNA and protein levels, while mi R-520d-5p knockdown increased Cyp B expression. Together, these results suggest that mi R-520d-5p reduces Cyp B expression by directly targeting the Cyp B 3-UTR.mi R-520d-5p inhibits GC growth in vitro and in vivo by targeting Cyp B. To investigate mi R-520d-5p function with regards to GC cell growth, BGC823 and SGC7901 cells were transfected with mi R-520d-5p mimics, inhibitors, agomir or antagomir, and mi R-520d-5p expression was confirmed by q RT-PCR. Both XTT and colony formation assays indicated that mi R-520d-5p upregulation significantly inhibited GC cell growth. Cell cycle analysis showed that restoration of mi R-520d-5p induced G1-phase arrest, whereas mi R-520d-5p inhibition reduced the proportion of cells in G1. Furthermore, mi R-520d-5p overexpression increased the proportion of cells undergoing apoptosis, while mi R-520d-5p knockdown reduced the number of apoptotic cell. Similar changes were also observed in SGC7901 cell. The effects of mi R-520d-5p on GC were also studied in vivo: BGC823 cells transfected with agomir or control were subcutaneously injected into the right and left flanks of the same nude mice, respectively. At 21 days post-injection, the mean xenograft tumour volume and weight was significantly lower for mi R-520d-5p-BGC823 xenografts than for mi R-control-BGC823 xenografts. mi R-520d-5p modulates STAT3 phosphorylation and nuclear translocation through Cyp B. We further examined the phosphorylation and subcellular localization of STAT3. mi R-520d-5p restoration were found to induce significant reduction in the phosphorylation of JAK2/STAT3 pathway-related proteins, while silencing mi R-520d-5p elicited increased JAK2 and STAT3 phosphorylation. Moreover, we found that the IL-6-induced time- and dose-dependent increase of STAT3 phosphorylation was also blocked by restoration of mi R-520d-5p. We further observed that Cyp B co-localized with STAT3 in the nucleus following IL-6 treatment as expected, while upon transfection of mi R-520d-5p, the percentage of cells with nuclear distribution of STAT3 was significantly decreased. To investigate whether STAT3 mediates the function of mi R-520d-5p in cancer growth, BGC823 cells were infected with STAT3 sh RNA and then transfected with mi R-520d-5p agomir. Functional study found that STAT3 knockdown abrogated the mi R-520d-5p-downregulation-induced GC growth promotion, as well as changes of cell cycle distribution and apoptotic percentage. Together, these data suggest that mi R-520d-5p modulates STAT3 activation and regulates cancer growth through a Cyp B/STAT3 axis.STAT3 directly suppresses mi R-520d-5p expression in GC cells. To further elucidate the connection between Cyp B and STAT3 activation, GES-1 cells that do not express significant amounts of Cyp B were treated with IL-6 and assayed for the expected STAT3 activation by western blotting. Interestingly, Cyp B expression was also induced by IL-6 treatment, and this expression can be blocked by STAT3 knockdown. Restoration of mi R-520d-5p attenuated the IL-6-induced upregulation of Cyp B. mi R-520d-5p expression were found to decrease in a time-dependent manner in cells treated with IL-6. Moreover, sh RNA-mediated downregulation of STAT3 prevented the repression of mi R-520d-5p after IL-6 treatment. We thus tested if STAT3 targets mi R-520d-5p directly. We generated a series of mi R-520d-5p promoter truncation mutants and determined whether STAT3 transcriptionally suppresses mi R-520d-5p. Luciferase assay after IL-6 treatment showed that the regulatory region might be between-1329 and-722 bp,which contains two STAT3 binding sites(-1215 to-1205 bp and-733 to-723 bp). Reporter genes containing the mi R-520d-5p promoter or various binding site mutations were then transfected into BGC823 and SGC7901 cells and then treated with IL-6. This analysis revealed that STAT3-based mi R-520d-5p regulation was lost when the region between-733 and-723 bp was mutated. Ch-IP assays further confirmed that STAT3 binds to the same site of the promoter of mi R-520 d. Together, these results indicate that STAT3 suppresses mi R-520d-5p transcription in GC cells. The Cyp B/STAT3/mi R-520d-5p feedback loop is characteristic of primary GC tissues. Finally, to test whether the regulation described above in GC cell lines is clinically relevant, in situ hybridisation(ISH) for mi R-520d-5p were performed on the 90 GC tissue cohort that was used for Cyp B and p STAT3 analysis. Compared with normal tissues, mi R-520d-5p levels were reduced in GC tissues. We found that patients with low mi R-520d-5p expression had significantly poorer prognoses than those with high mi R-520d-5p expression. Cox regression analysis also indicated that low mi R-520d-5p expression was an independent prognostic factor for poor survival in GC patients. In addition, we found that reduced mi R-520d-5p levels tended to increase Cyp B and nuclear p STAT3 expression. The 90 GC patient cases were then divided into groups with relatively high or low levels of mi R-520d-5p, Cyp B and p STAT3. From this analysis, we observed an inverse pattern of mi R-520d-5p and Cyp B or p STAT3 expression. In Summary, these results showed that the Cyp B/STAT3/mi R-520d-5p feedback loop is active in primary human gastric carcinogenesis.Conclusion: In conclusion, we elucidated the schematic model of GC development. H. pylori-associated gastritis induces gastric epithelial cells and macrophages to produce IL-6, activating the JAK2/STAT3 pathway and transcriptionally decreasing mi R-520d-5p, leading to a dramatic increase in Cyp B. Cyp B then aids STAT3 phosphorylation and nuclear translocation, resulting in JAK2/STAT3 pathway activation and increased GC cell proliferation and survival. In this view, the loss of mi R-520d-5p in GC cells and overexpression of Cyp B may be one of the reasons that STAT3 is constitutively activated in cancer cells and this new Cyp B/STAT3/mi R-520d-5p feedback loop may contribute to an improved understanding of inflammatory signalling in gastric carcinogenesis. |
PDF Full Text Request