| Objective:Prostate cancer(PCa)threatens the health of millions of men worldwide.PCa is the second most common cancer in male cancer patients after lung cancer and accounts for 7%of new male cancers diagnosed worldwide(15%in developed regions).In addition,there are more than 1.2 million new cases diagnosed and more than 350,000 PCa-related deaths worldwide each year,making PCa one of the leading causes of cancer-related deaths in men.Therefore,the diagnosis and prevention of PCa has always been an important work in human health.In this study,patients with PCa were divided into subtypes with different risk levels through transcriptome data,and then the risk differences among these subtypes were verified and explained using multiple omics data.Based on this,machine learning was used to construct clinical prognostic risk model and proliferating cell nuclear antigen clamp associated factor(PCLAF)was identified as the potential research target.PCLAF is involved in biological processes such as DNA synthesis and repair,cell cycle regulation and so on.Moreover,more and more studies have found that PCLAF is significantly related to the prognosis,occurrence and development of a variety of tumors.A study published in European Urology found that PCLAF expression decreased early in patients with PCa after rapid androgen castration,and the PCLAF gene promoter region was found to have an AR binding site.However,the role of PCLAF in PCa and its molecular mechanism are still unclear at present,which requires relevant exploration and research.Therefore,this study would carry out further studies from three levels:clinical tissue,in vitro cell line experiment and in vivo experiment in nude mouse model,in order to provide a research basis for the role of PCLAF in PCa and its molecular mechanism.Methods:Part Ⅰ:484 patients from the prostate adenocarcinoma(PRAD)cohort of The Cancer Genome Atlas(TCGA)database were screened into this study.Gene sets of androgen response,E2F transcriptional activation,fatty acid metabolism,glycolysis,hypoxia pathway,c-MYC transcriptional activation(V1 and V2 gene sets),oxidative phosphorylation,PI3K/AKT/mTOR pathway were obtained from Molecular Signatures Database v7.2.Single-sample gene set enrichment analysis(ssGSEA)was performed on 484 patients with these nine gene sets as background,and the results were used as input for consensus clustering.Thus,484 patients were divided into subtypes with different characteristics.The prognoses of these patients were assessed by survival analysis to determine risk levels among different subtypes.Clinical data,proliferation scores,published subtypes,immunoinfiltration deconvolution analysis,single nucleotide variants(SNV)data and copy number variants(CNV)of the 484 patients were then used to verify and define the risk characteristics among different subtypes.Then,the weighted correlation network analysis was used to find genes significantly associated with the high-risk subtype and the potential research target PCLAF.And genes related to the high-risk subtype were used as the input to construct one clinical prognostic risk model through least absolute shrinkage and selection operator.The validity of the model was tested in TCGA-PRAD cohort,GSE70769 dataset,DKF2018 dataset and MSKCC2010 dataset.Ridge regression model was used to predict sensitive drugs in high-risk patients identified by clinical prognostic risk models.Data from public database(TCGA-PRAD cohort,DKFZ2018 and MSKCC2010)were used again to detect the clinical correlation between PCLAF and PCa.Finally,pathologically discarded tissues from 35 patients in the Department of Urology,Shengjing Hospital of China Medical University were selected.These included 11 benign prostate hyperplasia tissues,24 PCa tissues,and 6 PCa-paired paracancer tissues.Immunohistochemistry(IHC)and Western blot(WB)were used to assess PCLAF expression patterns in PCa.Part Ⅱ:The mRNA and protein expression levels of PCLAF were detected by quantitative real-time polymerase chain reaction(qPCR)and WB assay in three PCa cell lines LNCaP,PC-3,DU 145 and a normal prostate stromal cell line WPMY-1.PCLAF was subsequently knocked down and overexpressed in LNCaP and PC-3 cell lines using shRNA lentiviruses and overexpressed lentiviruses,respectively.CCK-8 assay and colony formation assay were used to detect the proliferation of PCa cells in different groups.The change of migration ability of PCa cells was detected by wound healing assay.The invasion ability of PCa cells was detected by Transwell assay.Finally,subcutaneous tumor models of nude mice were established to detect the changes in tumor-forming ability of PCa cells in vivo.Part Ⅲ:Gene set enrichment analysis(GSEA)was used to search for signaling pathways related to PCLAF expression.Subsequently,cell lines with stable knockdown and overexpression of PCLAF were constructed in LNCaP and PC-3 cell lines using shRNA lentiviruses and overexpressed lentiviruses.WB assay was used to detect whether cMYC protein expression was changed after PCLAF expression was changed.Thus,the regulatory relationship between PCLAF and c-MYC was determined.Finally,cell lines with stable knockdown and overexpression of PCLAF were treated with 10074-G5,a c-MYC inhibitor.Colony formation assay,woud healing assay and Transwell assay were conducted again to test whether the inhibitory effect of knocking down PCLAF on PCa cell proliferation,migration and invasion phenotype could be further inhibited by 10074-G5 or whether the promoting effect of overexpressing PCLAF on PCa cell proliferation,migration and invasion phenotype could be effectively rescued by 10074-G5.Thus,the mediating role of c-MYC on PCLAF function could be determined.Finally,IHC assay was used to detect the regulatory relationship between PCLAF and c-MYC in subcutaneous tumor tissues of nude mice.Results:Part I:1.The 484 PCa patients in TCGA database could be classified into high-risk(Risk_H)subtype,median-risk(Risk_M)subtype,and low-risk(Risk_L)subtype according to the enrichment patterns of androgen response,E2F transcriptional activation,fatty acid metabolism,glycolysis,hypoxia pathway,c-MYC transcriptional activation(V1 and V2 gene sets),oxidative phosphorylation,and PI3K/AKT/mTOR pathway.2.Tumor proliferation of patients was more active in Risk_H subtype.3.SPOP mutations associated with poor prognosis in PCa were positively correlated with Risk_H subtype.The mutation frequency of SPOP in each subtype decreased successively:Risk_H>Risk_M>Risk_L.In addition,the results of our previous subtypes also support the worse prognosis of Risk_H.4.The tumor tissue of Risk_H was more pure and had fewer infiltrated immune cells.In terms of immune cell infiltration,the active natural killer cells and Treg cells were reduced in Risk_H tumor tissue.5.SNVs of SPOP and CNVs of RHOBTB2 and TNFRSF10C were more common in Risk_H subtype.Furthermore,the low expression of RHOBTB2 and TNFRSF10C in Risk_H may be associated with their single copy number deletion events.6.In this study,we found a group of genes that can effectively represent Risk_H subtype,and explored a potential target,PCLAF.7.A prognostic risk model consisting of 12 genes was constructed,and the model showed robust global expression and high accuracy in multiple data sets.8.According to the model calculation,for patients with high risk value,the sensitivity of leptomycin B,LY2606368,and vincristine is high.9.PCLAF has a significant clinical correlation with the occurrence and development of PCa at both transcription and translation levels,and the study of its mechanism is of further value.Part Ⅱ:1.PCLAF had higher mRNA and protein expression levels in PCa cells.2.After PCLAF was knocked out,the absorbance in CCK-8 experiment decreased significantly(one-way ANOVA test,P<0.05)and the number of clones in colony formation assay decreased significantly(independent-samples T Test,P<0.05).3.After PCLAF was knocked out,cell mobility in wound healing assay decreased significantly(independent-samples T Test,P<0.001).4.After PCLAF was knocked out,the number of cells passing through the Trans well chamber decreased significantly(independent-samples T Test,P<0.001).5.After PCLAF was knockout,the volume of subcutaneous tumors in nude mice decreased significantly(one-way ANOVA,P<0.01),the subcutaneous tumors weights decreased significantly in nude mice(Wilcoxon test,P<0.05).6.After overexpression of PCLAF,the absorbance of CCK-8 increased significantly(one-way ANOVA,P<0.05)and the number of clones in colony formation assay increased significantly(independent-samples T Test,P<0.01).7.Overexpression of PCLAF significantly increased cell mobility in wound healing assay(independent-samples T Test,P<0.001).8.After overexpression of PCLAF,the number of cells passing through the Trans well chamber increased significantly(independent-samples T Test,P<0.001).9.After overexpression of PCLAF,the volume of subcutaneous tumors in nude mice increased significantly(one-way ANOVA,P<0.01),subcutaneous tumors weights significantly increased in nude mice(Wilcoxon test,P<0.01).Part Ⅲ:1.According to GSEA,E2F pathway,c-MYC pathway,G2M checkpoint and oxidative phosphorylation have a high positive correlation with the expression of PCLAF,and the relationship between c-MYC and PCLAF is more valuable for further study.2.After PCLAF was knocked out,the expression level of c-MYC was also significantly reduced(independent-samples T Test,P<0.01).3.Knockdown of PCLAF showed synergistic effect of 10074-G5 on c-MYC expression(independent-samples T Test,P<0.01).4.After overexpression of PCLAF,the expression level of c-MYC was also significantly increased(independent-samples T Test,P<0.001).5.10074-G5 can effectively rescue the promoting effect of PCLAF on c-MYC expression(independent-samples T Test,P<0.01).6.After PCLAF was knocked down,the number of clones in colony formation assay decreased,and 10074-G5 could further reduce the number of cell clones(independent-samples T Test,P<0.01).7.The number of clones increased after overexpression of PCLAF,and 10074-G5 effectively rescued the increased number of cell clones(independent-samples T Test,P<0.01).8.After PCLAF was knocked out,the mobility of cells in wound healing assay decreased significantly,and 10074-G5 could further reduce the mobility of cells(independent-samples T Test,P<0.001).9.After overexpression of PCLAF,cell mobility increased significantly in wound healing assay,and 10074-G5 effectively rescued the increased mobility(independent-samples T Test,P<0.001).10.After PCLAF was knocked out,the number of cells passing through the Transwell chamber decreased significantly,and 10074-G5 further reduced the number of cells passing through the Trans well chamber(independent-samples T Test,P<0.001).11.After overexpression of PCLAF,the number of cells crossing the Transwell chamber was significantly increased,and 10074-G5 effectively rescued the increased number of cells crossing the Transwell chamber(independent-samples T Test,P<0.001).12.IHC experiment indicated that c-MYC expression was also decreased after the decrease of PCLAF expression in subcutaneous tumor tissues of nude mice(PCLAF was knocked down).13.IHC experiment suggested that the expression of c-MYC would increase with the increase of PCLAF expression in subcutaneous tumor tissues of nude mice(PCLAF was overexpressed).Conclusion:1.PCa patients were innovatively divided into three subtypes with different risk levels based on key biological processes such as androgen response,E2F transcriptional activation,fatty acid metabolism,glycolysis,hypoxia pathway,c-MYC transcriptional activation(V1 and V2 gene sets),oxidative phosphorylation,and PI3K/AKT/mTOR pathway.2.There were significant differences among three risk subtypes in clinical relevance,proliferation score,published subtypes,immune infiltration,SNV and CNV.3.This study also constructed and verified an effective prognostic risk model based on genes related to the high-risk subtype,so as to quantify the prognosis.Effective drugs were also predicted for patients with high-risk values.4.A potential research target PCLAF was identified from the Risk_H subtype related genes,and PCLAF was experimentally verified to have significant clinical correlation with the occurrence and development of PCa at both transcription and translation levels.5.After knocking down PCLAF,the proliferation,migration,invasion and tumorigenicity of PCa cells were decreased.6.The proliferation,migration,invasion and tumorigenicity of PCa cells were increased after overexpression of PCLAF.7.PCLAF has a positive regulation effect on c-MYC expression.8.PCLAF can promote the development of PCa through the mediation of c-MYC. |
PDF Full Text Request