2013-08-012024-05-18https://scholars.lib.ntu.edu.tw/handle/123456789/701826摘要：本團隊先前的研究中使用比較基因體技術(array-CGH)挑選出新穎致癌基因 ZNF322A。研究結果顯示ZNF322A 能夠促進癌細胞的生長及轉移侵襲能力，此外ZNF322A 為一個轉錄因子，促進cyclin D1但抑制p53基因轉錄活性；然而，目前對於 ZNF322A 於轉錄調控的機制及在腫瘤生成及轉移作用中的角色仍不清楚。由於ZNF322A 為一個鋅指 (Zinc finger) 蛋白，此鋅指結構可與DNA結合，也可與蛋白質結合，以形成多蛋白之轉錄調控複合體，並且已知其他鋅指蛋白活性可受轉譯後修飾所調控。因此本研究假設ZNF322A的轉錄調控複合體蛋白，可調控更多新穎的癌症相關基因，其ZNF322A蛋白本身也受轉譯後修飾如磷酸化的訊息傳遞機制。 近年來，將癌細胞特定基因轉殖及剔除後以新世代定序法分析mRNA-seq、ChIP-seq，並結合定量蛋白體技術及蛋白質體關聯性分析，探討轉錄調控蛋白的全基因體目標基因及特定DNA結合序列，並建立轉錄調控複合體蛋白及其轉譯後修飾的訊息傳遞，為機制研究的新趨勢。目前對於ZNF322A致癌蛋白相關研究僅有本團隊所發表的一篇報導，所以建立ZNF322A致癌蛋白的mRNA-seq、ChIP-seq、proteomics與interactomics資料庫，並整合上述四種型錄所建立的ZNF322A轉錄調控網路及關連蛋白系統生物研究，具有跨領域研究創新性；此ZNF322A調控網路系統生物研究也可以成為轉錄因子促進癌症機轉研究的基礎，成為全世界獨一無二的研究。因此本跨領域整合型計畫成立了三個子計畫進行整合研究，茲將研究內容說明如下：子計畫一將建構ZNF322A的mRNA-seq與ChIP-seq資料庫，與子計畫二以ZNF322A基因轉殖及剔除後的proteomics與interactomics資料庫配合，由子計畫三結合子計畫一、二之資料庫以及其他如signaling pathway、protein-protein interaction network等重要生物資料庫，在生物資訊技術上建構整合全基因體與蛋白體資料庫，並運用圖形探勘 (graph-based mining)、文字探勘 (text mining)、資訊檢索 (information retrieval) 及整合式探勘 (integrated mining) 等技術發展多種新穎之資料及文字探勘模組。 本跨領域整合型計畫的目標如下：(1) 以RNA-seq建立ZNF322A所影響之全基因體mRNA表現型錄，並驗證其新穎之正向與負向轉錄調控目標基因、(2) 以ChIP-seq建立ZNF322A所結合之全基因體區域，並定義出ZNF322A直接結合之DNA序列 (bone fide DNA binding elements)，及其DNA結合序列是否包含其他已知轉錄調控因子、(3) 以DNA affinity precipitation assay結合蛋白質譜分析法 (DAPA/MS) 及electrophoretic mobility shift assay結合蛋白質譜分析法 (EMSA/MS) 來揭露與ZNF322A 共同結合DNA之蛋白，以鑑定ZNF322A的協同轉錄調控因子 (transcriptional factors and co-regulators)、(4) 以定量蛋白質體技術 (isobaric tagging for relative and absolute quantitation, iTRAQ及LC-MS/MS) 及蛋白微型晶片 (proteome microarray) 建立ZNF322A所影響之全蛋白體型錄、(5) 驗證此ZNF322A全蛋白體型錄之訊息傳遞路徑蛋白，以釐清ZNF322A所參與之生物系統與癌症相關路徑、(6) 以磷酸化蛋白質體技術鑑定出 ZNF322A 的磷酸化位點及尋找可能調控 ZNF322A 的蛋白質激&#37238;，以進一步鑑定其訊息傳遞機轉、(7) 以分子模擬與結構生物學技術，探討 ZNF322A 的功能性片段進而了解ZNF322A如何與核酸及蛋白質進行結合、(8) 利用基於圖形理論的探勘方法(graph-based mining)、分類模型 (classification modeling) 與整合式探勘方法 (integrated mining methods) 建置各種新穎性資料/文字探勘模組進而找尋與預測重要轉錄目標基因 (transcriptional target genes) 以及ZNF322A互動蛋白、(9) 結合資料庫與上述資料/文字探勘模組至一個整合性高通量綜合分析平台，提供ZNF322A重要生物標誌發掘及相關延伸研究一個高效率及人性化之執行介面及分析環境。 本團隊在申請此跨領域整合計畫之前半年，每一子計畫已針對其設定目標進行技術平台的建立並得到初步研究結果：子計畫一由ZNF322A剔除及過度表現的細胞樣本（該樣本也交與子計畫二進行蛋白質體分析），以RNA-seq 及 ChIP-seq建構一個ZNF322A所影響之全基因體mRNA表現型錄及全基因體DNA結合型錄，在子計畫三的協助下定義出ZNF322A正向與負向轉錄調控目標基因群，其中5,373 ChIP 結合DNA區域可對應至 3,350 特定 RefSeq 基因，根據基因資料庫比對發現，ZNF322A結合之DNA序列有30.73%座落於目標基因之近端區域（promoter、5’UTR、coding region、3’UTR），並以生物資訊軟體分析發現，這些基因最為相關的調控路徑皆與細胞重要生理作用有關。利用motif 搜尋軟體初步定義出六個可能的ZNF322A直接結合之DNA序列，並發現其DNA結合序列含其他已知轉錄調控因子如AP-1、CREB1、FOXP1、TP53，其中也包含了及他的鋅指蛋白Klf4及協同轉錄調控因子Egr1。目前已進行驗證的基因其正確驗證比例高達60%以上。子計畫二利用iTRAQ定量蛋白質體技術，已建立ZNF322A所影響之全蛋白體型錄共1,108個蛋白，在子計畫三的協助下，定義出許多與子計畫一所建構的ZNF322A全基因體mRNA表現型錄相類似的細胞重要生理作用如：細胞凋亡、細胞週期、囊胞運輸、能量代謝調控路徑等；目前已有十個以上蛋白以西方轉漬法驗證。以磷酸化蛋白質體技術初步鑑定出二個 ZNF322A 的磷酸化位點，並預測出可能的蛋白質激&#37238;，目前正在以激&#37238;抑制劑與點突變方式進行驗證；同時也建構好ZNF322A全長與數個片段缺失的重組蛋白，將用於蛋白微型晶片結合分析，並與子計畫一、三合作探討 ZNF322A 的功能性片段，進而了解ZNF322A如何與核酸及蛋白質進行結合。子計畫三已利用一些新穎性技術，包含基於圖形理論的探勘方法 (graph-based mining)、文字探勘 (text mining)、資訊檢索 (information retrieval)、與整合式探勘方法 (integrated mining methods) 發展出一系列資料/文字探勘之基礎模組，以找尋與預測重要ZNF322A的轉錄目標基因與ZN<br> Abstract: ZNF322A was revealed as a novel oncogene in lung cancer patients using an array-comparative genomic hybridization in our previous study. We also found that ZNF322A is a transcription factor which activated cyclin D1 but repressed p53 transcription activities. Additionally, ZNF322A promoted tumor growth and metastasis. However, the precise mechanism of transcriptional regulation of ZNF322A and its role in tumorigenesis remain unclear. Since ZNF322A is an oncogenic zinc finger transcription factor, we hypothesized that other target genes, in addition to cyclin D1 and p53, could be transcriptionally regulated by ZNF322A. In addition, ZNF322A may interact with other proteins to regulation gene expression and oncogenesis. New generation sequencing-derived mRNA expression and chromatin-binding profiles are useful for transcription network establishment and application for mechanism study. Proteomics and interactomics studies can identify the interacting proteins (such as members of its transcriptional complex and its putative protein kinase). Thus far, there is no report on database integration and mining to identify transcriptional target genes and regulation networks for ZNF322A. To this end, we form an interdisciplinary program project grant (PPG) and the aims and approaches of each component project are described below: Project 1 will establish RNA-seq and ChIP-seq databases and Project 2 will establish proteomics and interactomics databases in cells with ZNF322A overexpression or knockdown. The four databases mentioned above will be combined by Project 3 as a meta-analysis-oriented platform together with useful biological datasets such as signaling pathways and protein-protein interaction networks. This PPG aims to (1) establish a transcriptional network of ZNF322A target genes by RNA-seq analyses, (2) define the intrinsic DNA sequence specificity or neighboring DNA elements of positive and negative regulated target promoters of ZNF322A by ChIP-seq, (3) uncover the cooperative transcriptional factors and co-regulators of ZNF322A by DNA affinity precipitation assay-Mass spectrometry (DAPA/MS) or electrophoretic mobility shift assay (EMSA)/MS, (4) construct the protein interaction network of ZNF322A by isobaric tagging for relative and absolute quantitation (iTRAQ) labeling and LC-MS/MS analysis as well as human proteome microarray, (5) define downstream signal pathways of ZNF322A, (6) identify key phosphorylation sites in ZNF322A and their putative protein kinases and further investigate the regulatory mechanism for ZNF322A function using phosphoproteomic and pathway analyses, (7) map the DNA and protein binding domain of ZNF322A by molecular modeling and structure biology, (8) develop various novel data/text mining modules to discover and predict important transcriptional target genes and interacting proteins of ZNF322A using graph-based mining, classification modeling and integrated mining methods, and (9) integrate the databases and data/text mining modules mentioned above into an integrative and high-throughput meta-analysis platform with a user-friendly interface for studies on ZNF322A and further related biomarkers. The representative preliminary data/tools are described below: Project 1 had performed RNA-seq and ChIP-seq to examine the differentially expressed genes and ZNF322A binding gene elements upon ZNF322A knockdown and overexpression. With the help from Project 3, the ZNF322A positively- and negatively-regulated genes were found in various cancers in disease biomarker analysis. A total of 5,373 ChIP regions corresponding 3,350 unique RefSeq genes were identified. We found that ZNF322A binging sites located in promoter, 5’UTR, coding region, and 3’UTR of the transcription unit, which accounted for 30.73% of total reads. To identify the bone fide DNA binding elements of ZNF322A and the potential ZNF322A-interacting transcription factors that may mediate the preferential recruitment of ZNF322A, we performed de novo motif analysis and found DNA-binding motifs for known transcription factors in which several have shown to interact with other ZNF proteins, supporting our hypothesis. We have conducted qRT-PCR and ChIP-qPCR for 20 more genes so far and the validation rate for ZNF322A downstream targets was >60%. Project 2 conducted iTRAQ quantitative proteomics to reveal the differential expressed proteins and identified 1,108 potential proteins regulated by ZNF322A. Using GO analysis by home-made script designed by Project 3, cancer-related functions such as the regulation of cell death, cell cycle, vesicle targeting and generation of energy were revealed, which similar to gene sets identified by Project 1. More than ten proteins had been verified by immunoblotting. To investigate experimentally the phosphorylation sites of ZNF322A, we made HA-ZNF322A overexpressing cells and performed LC-MS/MS analyses. We identified several novel ZNF322A phosphorylation sites. To understand the interaction between ZNF322A and partner protein or DNA binding element, full-length and several truncated forms of ZNF322A recombinant proteins were designed, produced and purified. These recombinant proteins will be used to identify DNA binding region of ZNF322A by EMSA and DAPA assays and interacting protein by proteome array. Project 3 has developed various novel data/text mining modules for discovering and predicting important transcriptional target genes and interacting proteins of ZNF322A by using techniques including graph-based mining, text mining, information retrieval and integrated mining techniques. The developed data mining modules will also be combined into a seamless and high-throughput meta-analysis platform. A preliminary analysis platform that could perform normalization, patterns analysis, clustering, gene ontology query, and pathway query has also been constructed, which had been included in the Bioinformatics Core of National Cheng Kung University and successfully applied to identify disease biomarkers for liver cancer, bladder cancer, and colon cancer. Such a seamless and high-throughput analysis platform will be optimized further in this project for ZNF322A study based on the data derived from Project 1 and Project 2. Together, most of the collaboration has been established or is working in progress in our team. The preliminary data will be used as the groundwork for extension to all sub-projects. The long-term goal is to establish regulation networks of ZNF322A and apply for oncogenesis studies. The platforms developed in our study can be used for other transcription factors and their disease models.ZNF322A轉錄調控因子轉錄調控網路蛋白質交互作用網路生物資訊整合型資料庫文字探勘平台訊息傳遞路徑癌症分子指標。ZNF322Atranscription factortranscriptional networkprotein interaction networkintegrated data