The Expression And Function Of Periostin Identified By Proteomics Analysis Of Prostate Cancer

Part I The differential proteomics analysis of prostate cancerObjective:To screen differentially expressed proteins and identify more biomarkers involved in oncogenesis and progression of prostate cancer(PCa) by the proteomics analysis using isobaric tags for relative and absolute quantification(iTRAQ). Then, the function and regulation of the differentially expressed proteins would be understood by bioinformatics analysis.Materials and Methods:The patients undergoing prostate biopsies were classified in 3 groups based on the pathological results of biopsies:benign prostate hyperplasia (BPH, n=20), PCa(n=20) and prostatic intraepithelial neoplasm(PIN, n=10). After quantification and enzymolysis of the protein extract from the specimens of the 3 groups, the iTRAQ regents 114,116 and 117 were used to label the peptides from BPH, PCa and PIN respectively. Then, the mixture of the peptides was analyzed by two-dimensional liquid chromatography-tandem mass spectrometry (2DLC-MS/MS). The MS/MS data were searched against the International Swissprot (090210,Human) using the Protein Pilot software (version 3.0) for peptide identification and quantification. The fold change cut-off ratio>1.50 or<0.66(116:114 or 116:117) was selected to designate the proteins of differential expression (P<0.05). In this study, the differentially expressed proteins of PCa compared with BPH(116:114) were mainly discussed. Several differentially expressed proteins such as Tumor protein D52(TPD52), Prohibitin-2, Elongation factor Tu(EF-Tu) and Decorin were verified by Western-blot assay. The MetaCore(GeneGo) software was used for bioinformatics analysis of the differentially expressed proteins including GO analysis, transcription regulation networks and transcription factor networks.Results:A total of 760 proteins were identified from the International Swissprot (090210, Human). Among the proteins, prostate specific antigen(PSA) and prostatic acid phosphatase(PAP) were well-known proteins enjoying clinical application. Based on the condition of screening differentially expressed proteins,20 proteins were significantly up-regulated (>1.5-fold) and 26 were significantly down-regulated in PCa(<0.66-fold) compared with BPH. Many of them such as mitochondrial proteins and extracellular matrix proteins have been previously reported as differentially expressed in PCa and closely related to oncogenesis and cancer progression. TPD52, Prohibitin-2, EF-Tu were significantly up-regulated (4.25fold,4.57fold and 3.02fold, respectively) and Decorin was down-regulated (0.43fold) in PCa. The results of Western-blot indicating that TPD52, Prohibitin-2, EF-Tu were overexpressed in PCa but Decorin was overexpressed in BPH were consistent with the proteomic changes of these 4 proteins. Periostin as an interesting protein will be mainly discussed in part?. In term of GO database, the top 10 components for cellular component (CC), molecular function (MF) and biological process(BP) of the differentially expressed proteins were picked out according to the enrichment of the these proteins. The results indicated that the differentially expressed proteins in our study mainly located in extracellular matrix with the function of binding were involved in muscle related BP, cytoskeleton organization, anti-apoptosis and cell adhesion. Additionally, according to the enrichment of the differentially expressed proteins, the top 5 regulation networks regulated by SP1, p53, YY1, androgen receptor(AR), c-Myc, respectively and the top 1 regulation factor network focusing on Slug were picked out. These transcription factors regulating the differentially expressed proteins were involved in BP such as anti-apoptosis, regulation of cell adhesion and regulation of cell cycle.Conclusion:The iTRAQ technology provides a good platform for us to understand the differentially expressed proteins and identify more significant biomarkers involved in oncogenesis and progression of PCa. The differentially expressed proteins of PCa identified by proteomic analysis using iTRAQ are reliable in our study and it is revealed that mitochondrial proteins and extracellular matrix proteins may play an important role in oncogenesis and progression of PCa. The bioinformatics analysis can broaden our mind and help us understand the global function and regulation of the differentially expressed proteins. The transcription factors such as SP1, p53, YY1, AR, c-Myc and Slug regulating the differentially expressed proteins screened by bioinformatics analysis are involved in the regulation of many BP in oncogenesis and progression of PCa.Part II The expression and function of Periostin in prostate cancerObjective:To investigate the expression of Periostin in the tissues and cell lines of prostate cancer (PCa). Furthermore, the effects of sliencing Periostin expression by RNA interference(RNAi) on proliferation?invasion and apoptosis of PCa LNCap cell line were studied.Materials and Methods:Western-blot assay and immunohistochemical staining were used to analyze the expression of Periostin in the tissues of PCa.Then, the protein and mRNA expression of Periostin in different PCa cell lines:LNCaP, DU145,22RV1 and PC3 were detected by immunofluorescence staining, western-blot assay and reverse transcription PCR(RT-PCR) analysis. The LNCap cells with Periostin expression were used for the study of RNAi. The LNCap cells were classified in 3 groups:normal LNCap cells, GFP-LNCap cells transfected with control GFP lentiviral particles and shRNA-Periostin-LNCap cells transfected with shRNA-Periostin lentiviral particles. The efficiency of shRNA-Periostin lentiviral particles transduction was evaluated by immunofluorescence. The expression of Periostin in shRNA-Periostin-LNCap cells was detected by real-time PCR analysis and western-blot assay. Additionally, the effects of sliencing Periostin expression by RNAi on proliferation, invasion and apoptosis of LNCap cells were analyzed by MTT assay, flow cytometry and transwell chambers. Furthermore, the subcutaneous xenograft models in nude mice of PCa were established and classified in 3 groups:normal LNCap cells group, GFP-LNCap cells group and shRNA-Periostin-LNCap cells group. The effects of silencing Periostin expression by RNAi on the tumor formation of LNCap cells in vivo were studied and the microvessel density(MVD) in the xenografts of three groups was analyzed by immunohistochemical staining of CD31.Results:The results of western-blot revealed a significant increase of Periostin amount in PCa compared to BPH. Furthermore, immunohistochemical staining was performed to evaluate Periostin expression in the stroma or epithelial cells of prostate. Although, there was no statistical significance for the epithelial expression of Periostin between PCa and BPH (P>0.05). Statistical significance was observed for the stromal expression of Periostin between PCa and BPH (P<0.01). The results of immunofluorescence staining, western-blot and RT-PCR indicated that both the protein and mRNA of Periostin were only expressed in PCa LNCap cell line. The effective rate of shRNA-Periostin lentiviral particles transfecting LNCap cells after 48h was more than 90%under fluorescence microscope. The LNCap cells stablly expressing shRNA-Periostin were obtained by transfecting shRNA-Periostin lentiviral particles. As was expected, both the protein level and mRNA level of Periostin were significantly reduced by shRNA-periostin lentiviral particles transfection. The MTT results illustrated that compared with normal LNCap cells and GFP-LNCap cells, shRNA-Periostin-LNCap cells started to grow slowly since the third day. There was the most significant difference of growth rates in vitro at 6 day (P<0.01). The results of flow cytometry indicated that the apoptotic rate of shRNA-Periostin-LNCap cells was 18.1%. But, the apoptotic rates of normal LNCap cells and GFP-LNCap cells were 4.4%and 4.9%(P>0.05), respectively. The apoptosis of LNCap cells was promoted by sliencing Periostin expression in vitro (P<0.05). The number of migrated cells on the underside of the chambers were calculated by microscope. For the nomal LNCap cells and GFP-LNCap cells, the number of migrated cells was 37.38±5.53, 35.38±6.57 (P>0.05), respectively. For shRNA-Periostin-LNCap cells, the number of migrated cells was 20.25±6.71. Our results indicated that sliencing Periostin expression significantly inhibited invasion of LNCap cells in vitro (P<0.05). Similar to the growth inhibition of LNCap cells in vitro, the growth of the xenografts in shRNA-Periostin-LNCap cells group was obviously inhibited. Sliencing Periostin expression of LNCap cells also resulted in a significant decrease in the tumor burden of shRNA-Periostin-LNCap cells group. The immunohistochemical staining of CD31 indicated that MVD in the xenografts of nomal LNCap cells group, GFP-LNCap cells group, and shRNA-Periostin-LNCap cells group were 32.1±9.4,30.2±11.5,15.5±5.6, respectively. There was a significant decrease in MVD of shRNA-Periostin-LNCap cells group (P<0.05).Conclusion:Periostin is overexpressed in PCa matrix and only expressed in PCa LNCap cell line. Sliencing Periostin expression in PCa LNCap cell line by RNAi cannot only inhibit proliferation and invasion of LNCap cells but also promote apopotosis of LNCap cells in vitro. The growth inhibition of LNCap cells in vivo by sliencing Periostin expression may be related to the inhibition of tumor angiogenesis. So, Periostin may be a promising biomarker and a novel target of therapeutical intervention for PCa in future.