Cervical cancer is the third leading cause of cancer death in women worldwide. Although cervical cancer is commonly infected with human papillomavirus (HPV), HPV infection alone is insufficient to induce malignant changes. Many characteristic genetic and epigenetic alterations have been identified in invasive cervical carcinomas but relatively little is known about the specific genetic and molecular alterations that allow pre-invasive epithelial cells to acquire the ability to progress to invasive squamous cell carcinomas. Recently, a family of small non-coding RNAs termed microRNAs (miRNAs) with specific inhibitory functions on target gene expression has been suggested to play an important role in the pathogenesis of human cancers including lung and breast cancer but remain undefined in cervical cancer. / Genome wide chromosomal copy number changes in cervical cancer by Agilent high-density array Comparative Genomic Hybridization demonstrated that only a very limited number of genomic imbalances have an impact on the miRNA profile in cervical cancer cells, although a high proportion of genomic loci containing miRNA genes exhibited DNA copy number alterations in other cancers. The impact of the genomic aberration on their mRNA expression was then confirmed by Aligent Whole Human Genome gene expression array. This suggests that the regulation of miRNA and mRNA expression may be different in cervical cancer. / In conclusion, our global miRNA profiling identified the common differentially expressed and genomic aberration independent miRNAs in cervical cancer. We further revealed the inhibition of hsa-miR-182 reduced tumor cell growth in vitro and in vivo through apoptosis and cell cycle mechanism. This provides new evidence that hsa-miR-182 may contribute to the pathogenesis of cervical cancer. / Keywords. MicroRNA, Cervical Cancer, Tumor Growth / To identify microRNA(s) associated with the tumorigenesis of cervical cancer, we firstly used the TaqMan MicroRNA Assays to survey and quantify a panel of 157 known human miRNAs in cervical cancer cell fines and micro-dissected normal cervical epithelium cells. We identified 2 microRNAs that were differentially up-regulated (fold change > 2, p < 0.05) and 9 differentially down-regulated (fold change > 2, p < 0.05) in cervical cancer cell lines comparing with normal cervical epithelium. Further investigation in tumor samples confirmed these two up-regulated miRNAs (hsa-miR-182 and -183 ) and 3 down-regulated miRNA (hsa-miR-145, 150, 195) from 4 investigated downregulated miRNAs (hsa-miR-145, 150, 195 and 328). / To investigate the biological function of those aberrantly expressed microRNAs, we chose one of the most aberrantly up-regulated microRNA ( hsa-miR-182, fold change > 10) for further investigation. Inhibition of hsa-miR-182 by antisense oligonucleotides inhibited HeLa cervical cancer cell growth in vitro and reduced tumor cell volume in vivo. Gene expression array analysis of HeLa cells with hsa-miR-182 knockdown and over-expression showed specific hsa-miR-182 targeting pathway in apoptosis and cell cycle. It indicated the roles of hsa-miR-182 in cervical cancer growth through apoptosis and cell cycle functions. / Tang, Tao. / Adviser: Richard K W Choy. / Source: Dissertation Abstracts International, Volume: 70-06, Section: B, page: 3446. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (leaves 155-169). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

Gastric cancer is the second leading cause of cancer death worldwide and in China. The mechanism of gastric carcinogenesis is not fully understood. Epigenetic studies indicated that inactivation of tumor suppressor genes by DNA hypermethylation plays a crucial role in the progression of gastric cancer. Epigenetic inactivation of secreted frizzled-related protein (SFRP 1) by methylation plays a pivotal role on the development of various cancers. However, the role of SFRP family genes in gastric cancer remains largely unknown. We aimed to characterize the epigenetic abnormalities and discover novel biomarkers for early detection of gastric cancer. We investigated the epigenetic alterations in gastric adenocarcinoma by microarray based analysis and gene promoter hypermethylation. Based of the microarray data, we determined the functional significance and frequency of SFRP family genes hypermethylation in human gastric cancer. We screened the mRNA expression and methylation status of the SFRP family members in human gastric cancer cell lines and primary gastric cancer samples. Demethylation study of SFRP family genes were done by treating gastric cancer cell lines with 5'Aza. The biological effects of SFRP were analyzed by flow cytometry, cell viability assay and tumor growth in nude mice. SFRP1, 2, 4 and 5 were undetectable in 100% (7/7), 100% (7/7), 42.8% (3/7) and 85.7% (6/7) of gastric cancer cell lines, respectively. However, only SFRP2 showed significant down-regulation in gastric cancer compared with adjacent non-cancer samples (P<0.01). Treatment with demethylation agent, 5'-Aza, restored the expression of SFRP2 in all 7 cancer cell lines. Promoter hypermethylation of SFRP2 was detected in 73.3% of primary gastric cancer samples and 20% of adjacent non-cancer tissue (P<0.01). Bisulfite sequencing confirmed the density of promoter methylation in cell line, primary gastric cancer tissue and their adjacent non-cancer tissue. Transfection of SFRP2 induced cell apoptosis, inhibited proliferation in vitro and suppressed tumor growth in vivo. Furthermore, SFRP2 methylation was detected in 37.5% of samples showing intestinal metaplasia. Methylated SFRP2 was also detected in 66.7% of serum samples from cancer patients but not in normal controls. Epigenetic inactivation of SFRP2, but not SFRP1, SFRP4 and SFRP5 is a common and early event of carcinogenesis. Hence, detection of SFRP2 methylation in serum may have diagnostic value in gastric cancer patients. / by Cheng, Yuen Yee. / Adviser: FKL Chan. / Source: Dissertation Abstracts International, Volume: 69-02, Section: B, page: 0803. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (leaves 165-179). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese. / School code: 1307.

Background. B-lymphoma Moloney murine leukemia virus insertion region-1 (Bmi-1) acts as an oncogene in various cancer such as non-small cell lung cancer, colon cancer, gastric cancer, bladder cancer and nasopharyngeal cancer (NPC). / Methods. Immunohistochemistry was performed to evaluate Bmi-1 expression in 252 breast cancer samples. The correlations were analyzed between Bmi-1 expression and clinicopathologic parameters, including age, tumor size, lymph nodal involvement, distant metastasis, clinical stages, hormone receptor (ER, PR) and Human Epidermal Growth Factor Receptor 2 (HER-2). The overall survivals were compared by Kaplan-Meier analysis based on Bmi-1 expression. / Results. Bmi-1 expression was significantly increased in primary cancer tissues than in matched adjacent non-cancerous tissues ( P<0.001). Only 35.9% (14 of 39) of adjacent non-cancerous tissues displayed high expression compared with 72.2% (182 of 252) in primary cancer tissues. Among adjacent non-cancerous tissues, no Bmi-1 staining signal was detected in 30.8% (12 in 39) samples. Only 28.2% (11 in 39) samples showed nucleus staining and the remaining 41.0% (16 in 39) samples exhibited cytoplasm staining. Of those cancer tissues, however, 75.4% (190 in 252) was stained in the nucleus and 24.6% (62 in 252) located in the cytoplasm. The elevated Bmi-1 expression was correlated with advanced clinicopathologic classifications (T, N, M) and clinical stages (P<0.001, respectively). A high level of Bmi-1 expression displayed unfavorable overall survival ( P<0.001). The overall survival rate, assessed by the Kaplan-Meier method, was 85.1% (57 in 67) in low Bmi-1 expression group, whereas it was only 59.9% (103 in 172) in high Bmi-1 expression group. In addition, Bmi-1 serves as a high risk for breast cancer and the relative risk increased almost four fold in patients with high Bmi-1 expression compared with that with low Bmi-1 expression by univariate Cox regression analyses. After the adjustment of the confounding factors, Bmi-1 was still found to predict the poor survival (P=0.042), which indicated Bmi-1 was an independent prognostic factor. The overexpression of Bmi-1 increased the mobility and invasiveness in 76N-TERT and MCF-10A, concurrent EMT-like molecular changes, the stabilization of Snail protein and the activation of Akt/GSK3beta pathway. Consistent with these observations, the repression of Bmi-1 in MDA-MB-435S remarkably attenuated the cellular mobility, invasiveness and transformation, as well as tumorigenesis and spontaneous lung metastases in nude mice. In addition, the repression of Bmi-1 reversed the EMT markers and inhibited the Akt/GSK3beta/Snail pathway. However, ectopic Bmi-1 alone was not able to lead to the phenotype of HMECs. Additionally, discordant mRNA expression levels of Bmi-1 and E-cadherin were detected between primary cancer tissues and matched adjacent non-cancerous tissues. The mRNA level of Bmi-1 was strongly up-regulated in breast cancer tissues compared with paired non-cancerous tissues ( P=0.001), whereas the mRNA level of E-cadherin was markedly down-regulated (P=0.042). Furthermore, there was a converse correlation between Bmi-1 and E-cadherin expression at the transcriptional level ( P=0.041). (Abstract shortened by UMI.) / Guo, Baohong. / Adviser: Kung, Hsiang Fu. / Source: Dissertation Abstracts International, Volume: 73-02, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2010. / Includes bibliographical references (leaves 161-183). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

Array comparative genomic hybridization (array-CGH) was used in this study to analyze the chromosomal aberrations in 9 gastric cancer cell lines. Our results showed good concordance with those of conventional CGH. We correlated the results from array-CGH with expression profiling and found some novel and independent target genes which deserved further confirmation. / Gastric cancer is one of the most common malignancies worldwide and is the second most frequent cause of cancer related death. A variety of genetic and epigenetic aberrations underlie development abnormality of gastric cancer. / Taken together, our findings supported YAP1 is a functional oncogene in gastric cancer. We provided the first evidence that YAP1 exerted the oncogenic function by enhancing the capacity to activate the early response gene pathway. YAP1 could be a prognostic biomarker and potential therapeutic target for gastric cancer. / The study was focused on the putative oncogene Yes-associated Protein 1 (YAP1) located in 11q22.1. Up-regulation of YAP1 was observed in 92.3% of gastric cancer by immunohistochemistry (IHe) on gastric cancer tissue microarrays. YAP1 nuclear accumulation correlated with cancer specific survival. In addition, multivariate Cox regression showed that YAP1 was an independent predictor of short disease specific survival time for patients with early stage gastric cancer (P=0.042) in addition to T stage ( P=O.038). Knockdown YAP1 in gastric cancer cell lines MKN1 and AGS resulted in a significant reduction in proliferation, anchorage-dependent colony formation, cell invasion and cell motility. Ectopic YAP1 expression in MKN45 cells promoted anchorage-independent colony formation, induced a more invasive phenotype and accelerated cell growth both in vitro and in vivo. Microarray analysis highlighted the alteration of MAPK pathway by YAP1. We confirmed a constitutive activation of RAF/MEKJERK in YAP1-expressing MKN45 cells and further demonstrated that YAP1 enhanced serum/EGF induced c-Fos expression in gastric cancer cells. Furthermore, we demonstrated that ectopic MST1 promoted phosphorylation and cytoplasmic translocation of YAP1 and subsequently quenched the oncogenic function of YAP1 in the nucleus. / Kang, Wei. / "December 2010." / Adviser: To Kai-fai. / Source: Dissertation Abstracts International, Volume: 73-04, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2011. / Includes bibliographical references (leaves 175-185). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

Guang Fu Chen. / "May 2002." / Thesis (Ph.D.)--Chinese University of Hong Kong, 2002. / Includes bibliographical references (p. 194-221). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

Chan Wing Yan Michael. / "August 2002." / Thesis (Ph.D.)--Chinese University of Hong Kong, 2002. / Includes bibliographical references (p. 168-200). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

研究背景及目的：EB病毒(EBV)相關性胃癌的發病率約占胃癌的10%。近年來，越來越多的研究表明， EBV相關性胃癌的腫瘤抑制基因發生異常甲基化。然而，EBV的感染對全基因組DNA甲基化的影響尚不清楚。本研究通過分析EBV感染的細胞中全基因組DNA甲基化的情況，篩選出因EBV感染而發生甲基化的基因，並闡明靶基因在胃癌發生過程的作用。 / 方法：本研究應用穩定轉染EBV的胃癌細胞AGS (AGS-EBV)和無EBV轉染的AGS細胞為模型。採用高解析度的甲基化DNA免疫共沉澱晶片技術(MeDIP-chip)比較AGS-EBV 和AGS全基因組DNA甲基化的變化，並根據基因本體論(GO)，對EBV誘導的甲基化基因進行分類。採用RT-PCR，去甲基化處理及亞硫酸氫鈉測序(BGS)等方法驗證EBV誘導的甲基化基因。同時，採用基因敲除和過表達方法體外研究靶基因的生物學功能：通過細胞活力實驗和集落形成實驗判斷靶基因對細胞增殖的影響；通过流式细胞技术、伤口愈合实验及侵袭实验研究筛选到的靶基因生长抑素受体1（SSTR1）的功能；此外，还通過腫瘤通路基因PCR晶片分析靶基因調控的下游腫瘤相關基因。 / 结果：EBV編碼的小RNA(EBER)原位雜交方法和EBV潛伏期膜蛋白(LMP2A)的表達均證實AGS-EBV細胞中確實存在EBV的感染。和AGS細胞相比，發現AGS-EBV細胞中DNA甲基轉移酶3b(DNMT3b)的表達和活性顯著增加。AGS細胞中， LMP2A過表達後，DNMT3b的表達和活性也顯著增加。通過MeDIP-chip篩選出AGS-EBV中1,065甲基化有差異的基因，其中886基因為高甲基化。GO分析結果表明這些高甲基化基因參與KEGG信號通路。其中，六個新的高甲基化基因(MDGA2, IL15RA, SCARF2, EPHB6, SSTR1 和 REC8)在AGS-EBV細胞中的表達低於AGS；經過去甲基化處理之後，這些基因的表達水準有顯著增加。 / 通過深入研究生長抑素受體1(SSTR1)的生物學功能，發現：敲除SSTR1 能促進胃癌細胞的增殖和集落形成；通過调节G1/S期的調節因子，加快細胞進入S期，顯著增加S期的細胞數目。此外，胰腺癌細胞PANC1細胞中，SSTR1的過表達，也進一步證實SSTR1確實是一種腫瘤抑制基因。腫瘤通路基因PCR晶片結果顯示SSTR1通過促進細胞週期抑制因數(包括p15，p16，p21和p27)的表達，同時抑制CDC25A 和Myc的表達，發揮抑制增殖作用，導致細胞週期停滯在G1期，減少細胞增殖。SSTR1也通過減少凋亡相關基因的表達參與細胞凋亡的過程。此外，SSTR1還能顯著下調遷移相關基因的表達。這些結果表明，在EB病毒相關胃癌的發生過程中，SSTR1通過調節細胞週期、凋亡及遷移的有關基因，進而抑制細胞增殖，減少細胞的遷移和轉移。 / 结论：AGS感染EBV後，通過LMP2A促進DNMT3b的表達，激活DNMT3b的活性，導致886個腫瘤相關基因发生甲基化。SSTR1是一種EBV誘導的新的甲基化基因，在胃癌中具有抑制腫瘤的特性。研究表明， 由EBV誘導的SSTR1所具有的表觀遺傳學抑制作用參與EB病毒相關性胃癌的發病機制。 / Background and Aims: Epstein-Barr virus (EBV)-associated gastric cancer (GC) accounts for about 10% of all GCs. Accumulating evidences revealed aberrant rmethylation of tumor suppressor genes in EBV-associated GCs. However, the effect of EBV infection on the genome-wide aberrant DNA methylation remains unclear. We aim to profile the genome-wide EBV-associated hypermethylation in EBV-infected cells, to identify EBV-associated methylated genes and to elucidate their function in gastric carcinogenesis. / Methods: The cell model of gastric cancer AGS cells with or without stable EBV infection was used in this study. Genome-wide DNA methylation profiles were compared between AGS-EBV and AGS cells by high resolution Methyl-DNA immunoprecipitation microarry (MeDIP-chip) assay. EBV-associated methylated genes were classified according to gene ontology (GO). The novel EBV-associated methylated candidates were validated using bisulfite genomic sequencing (BGS), RT-PCR, and demethylation treatment. Biological function of one of the candidate genes (Somatostatin Receptor 1, SSTR1) was studied in vitro using gene knockdown and over-expression approaches simultaneously. Effects of SSTR1 expression on gastric cancer cell was measured by cell viability assay, colony formation assay, flow cytometry, wound-healing assay and invasion assay. Gene modulation by SSTR1 in human cancer pathways was assessed by cancer pathway PCR array. / Results: EBV infection was confirmed by EBER in situ hybridization. LMP2A expression was detected in AGS-EBV cells but not in EBV negative AGS cells. Expression and activity of DNMT3b was found to be significantly increased in AGS-EBV cells compared to AGS cells. Ectopic expression of LMP2A in AGS enhanced the expression and activity of DNMT3b. MeDIP-chip profiling identified a total of 1,065 genes differentially methylated by EBV infection (fold changes ≥2, P < 0.05) (fold-changes 2.4365.2). Gene ontology analysis indicated the enrichment of hypermethylated genes involving in important KEGG pathways. Notably, in addition to higher methylation levels confirmed by BGS, six novel hypermethylated genes (MDGA2, IL15RA, SCARF2, EPHB6, SSTR1 and REC8) were down-regulated in AGS-EBV cells as compared with AGS cells. Furthermore, demethylation treatment increased transcription levels of the six genes in AGS-EBV cells. / The biological function of SSTR1 gene was further investigated. Knockdown of SSTR1 in GC cells increased cell proliferation (P < 0.05) and colony formation ability (P < 0.01), and markedly increased cells in S phase through regulating G1/S phase mediators. Overexpression of SSTR1 in PANC1 cell line further confirmed that SSTR1 indeed was a tumor suppressor gene. Analysis of SSTR1 regulation of cancer pathway demonstrated that SSTR1 exerted antiproliferative effect by inducing cyclin-dependent inhibitors (p15, p16, p21 and p27) and inhibiting cell divison cycle 25 homolog A (CDC25A) and Myc, resulting in cell cycle arrest in G1 phase and reduction of cell proliferation. SSTR1 also took part in proliferation by decreasing expression of apoptosis regulators. Moreover, SSTR1 significantly downregulated the expression of migration-related genes, including ITGA1, ITGA2, ITGA3, ITGB5, IL8, MMP1 and PLAUR. These findings suggest that SSTR1 inhibits proliferation and reduces cell migration/invasion in gastric cancer by deregulating genes invloved in the regulation of cell cycle, survival/apoptosis and migration. / Conclusions: EBV infection in AGS cells induces genome-wide aberrant hypermethylation of 886 genes which involved in important cancer-related pathways. EBV-associated methylation is mediated by activation of DNMT3b through LMP2A. We identified and functionally characterized a novel EBV-associated methylated gene SSTR1 which exerted anti-tumor properties in GC. Epigenetic silencing of SSTR1 associated with EBV infection contributes to the pathogenesis of EBV-associated GC. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Zhao, Junhong. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 129-141). / Abstract also in Chinese. / ABSTRACT --- p.i / 摘 要 --- p.iv / Acknowledgements --- p.vi / Publications --- p.vii / Research articles --- p.vii / Conference abstracts --- p.viii / Table of contents --- p.x / list of tables --- p.xiii / list of figures --- p.xiv / list of abbreviations --- p.xvi / Chapter Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- General introduction --- p.1 / Chapter 1.2 --- Gastric Cancer --- p.3 / Chapter 1.2.1 --- Eipdemiology of Gastric Cancer --- p.4 / Chapter 1.2.2 --- Pathology of Gastric Cancer --- p.8 / Chapter 1.2.3 --- Risk Factors for Gastric Cancers --- p.11 / Chapter 1.3 --- Epstein-Barr Virus-associated Gastric Cancer (EBVaGC) --- p.15 / Chapter 1.3.1 --- Historical Discovery and Harm of EBV --- p.15 / Chapter 1.3.2 --- Molecular Biology of EBV --- p.16 / Chapter 1.3.3 --- Latent and Lytic Infection of EBV --- p.18 / Chapter 1.3.4 --- EBV Products --- p.18 / Chapter 1.3.5 --- EBV-associated gastric cancer (EBVaGC) --- p.28 / Chapter 1.4 --- EBV-induced Epigenetic Alteration in Gastric Carcinogenesis --- p.36 / Chapter 1.4.1 --- Cytosine Methylation and CpG Island --- p.36 / Chapter 1.4.2 --- DNA Methylation in Gastric Cancer --- p.39 / Chapter 1.5 --- How to identify EBV-induced promoter methylation in gastric cancer --- p.45 / Chapter 1.5.1 --- Methylated DNA Immunoprecipitation (MeDIP) --- p.48 / Chapter 1.5.2 --- Combined Bisulfite Restriction Analysis (COBRA) --- p.49 / Chapter 1.5.3 --- Bisulfite Genomic Sequencing --- p.49 / Chapter 1.5.4 --- Pyrosequencing --- p.49 / Chapter Chapter 2 --- Materials and Methods --- p.51 / Chapter 2.1 --- Materials --- p.51 / Chapter 2.1.1 --- Cancer Cell Lines and Culture Condition --- p.51 / Chapter 2.1.2 --- Primary GC Samples --- p.52 / Chapter 2.2 --- EBV Encoded Nuclear RNA (EBER) in situ Hybridization (EBER-ISH) --- p.52 / Chapter 2.3 --- Western Blot Analysis --- p.53 / Chapter 2.4 --- Plasmid and Transfection --- p.56 / Chapter 2.4.1 --- Plasmid Construction and Extraction --- p.56 / Chapter 2.4.2 --- Plasmid Transfection --- p.59 / Chapter 2.5 --- Gene Expression Analysis --- p.59 / Chapter 2.5.1 --- Purification of Total RNA (RNeasy Kit, Qiagen) --- p.59 / Chapter 2.5.2 --- cDNA Reverse Transcription --- p.60 / Chapter 2.5.3 --- Semi-Quantitative PCR --- p.62 / Chapter 2.5.4 --- Quantitative Real-Time PCR (qRT-PCR) --- p.64 / Chapter 2.6 --- DNMT1 and 3b Activity Assay --- p.64 / Chapter 2.7 --- DNA Methylation Analysis --- p.64 / Chapter 2.7.1 --- Genomic DNA Extraction --- p.64 / Chapter 2.7.2 --- Genome-wide Profiling of EBV-associated DNA Methylation by MeDIP-chip --- p.65 / Chapter 2.7.3 --- Bioinformatics Analysis --- p.66 / Chapter 2.7.4 --- CpG Island Prediction and Analysis of the Targets’ Promoter Region --- p.66 / Chapter 2.7.5 --- DNA Sodium Bisulfite Modification --- p.67 / Chapter 2.7.6 --- Target Gene Methylation in GC Cell Lines --- p.67 / Chapter 2.7.7 --- Bisulfite Pyrosequencing Analysis in GC Tissue Samples --- p.70 / Chapter 2.8 --- Biological Function Analysis of SSTR1 --- p.72 / Chapter 2.8.1 --- Cell Proliferation Assay for Stable Transfection --- p.72 / Chapter 2.8.2 --- Colony Formation Assay --- p.72 / Chapter 2.8.3 --- Cell Cycle Analysis Assay --- p.72 / Chapter 2.8.4 --- Cell Migration Analysis --- p.73 / Chapter 2.8.5 --- Invasion Analysis --- p.73 / Chapter 2.8.6 --- Human Cancer Pathway Finder RT2 Profiler PCR Array Analysis --- p.74 / Chapter 2.9 --- Statistical Analysis --- p.76 / Chapter Charpter 3 --- results --- p.77 / Chapter 3.1 --- EBV Infection in AGS-EBV Cell Model --- p.77 / Chapter 3.2 --- Activation of DNMT3b in AGS-EBV Cells --- p.80 / Chapter 3.3 --- LMP2A Induced DNMT3b Activity in AGS Cells --- p.82 / Chapter 3.4 --- Genome-wide Profiling of DNA Methylation Associated with EBV Infection Using MeDIP-chip --- p.84 / Chapter 3.5 --- EBV-associated Cancer Pathways Defined by EBV-associated Promoter Methylated Genes --- p.86 / Chapter 3.6 --- CpG Hypermethylation and Transcriptional Silencing of EBV-associated Methylated Genes in AGS-EBV Cells --- p.88 / Chapter 3.7 --- Bioinformatics Analysis of SSTR1 Using University of California Santa Cruz Genome Bioinformatics (UCSC) Database and CpG Island Searcher --- p.93 / Chapter 3.8 --- COBRA Analysis of SSTR1 Promoter Methylation in GC Cell Lines --- p.93 / Chapter 3.9 --- Frequent SSTR1 Hypermethylation was Associated with EBV Positive Primary Gastric Cancer --- p.96 / Chapter 3.10 --- SSTR1 was Down-regulated in GC Cell Lines through RNA Interference --- p.103 / Chapter 3.11 --- SSTR1 Knockdown Induced Cell Proliferation in GC Cell Lines --- p.105 / Chapter 3.12 --- SSTR1 Knock-down Promoted Cells to Enter into S Phase --- p.108 / Chapter 3.13 --- SSTR1 Knock-down Increased the Migration Ability of GC --- p.110 / Chapter 3.14 --- SSTR1 Knock-down Promoted Cell Invasion --- p.112 / Chapter 3.15 --- Ectopic Expression of SSTR1 Inhibited Proliferation and Clonogenicity in PANC1 Cancer Cells --- p.114 / Chapter 3.16 --- Identification of Genes Modulated by SSTR1 --- p.116 / Chapter Chapter 4 --- Discussion --- p.119 / Chapter Chapter 5 --- Limitation of the study --- p.127 / ConclusionS --- p.128 / Reference --- p.129

雖然錯誤調控微小核糖核酸 (miRNA) 引起肝細胞癌 (HCC) 發生發展的生物途徑得到了廣泛的研究，但是對於上游的調控機制卻知之甚少。以往的研究表明，組蛋白甲基化轉移酶 (EZH2) 介導的組蛋白3上27位賴氨酸三甲基化（H3K27me3）是一類通過沉默腫瘤抑制基因而誘發癌症的機制，並且與脫氧核糖核酸 (DNA) 啟動子甲基化機制獨立存在。另一方面，基因抑制也與 H3K27和DNA甲基化相關聯。盡管如此，miRNA沉默機制，特別是在肝癌中，仍然是知之甚少。 / 在這項研究中，我們使用整合全基因組定位和表達分析方法，以探討在肝癌細胞中miRNA表達的表觀遺傳和轉錄控制。通過染色質免疫沉澱偶聯人類啟動子芯片的方法，我們發現在Hep3B和HKCI - 8肝癌細胞中分別有8.4和12.2%的審問miRNA有豐富的H3K27me3。另一方面，在甲基結合域捕捉偶聯芯片實驗中，我們發現在Hep3B和HKCI-8肝癌細胞中分別有15.5和14.7% 的miRNA出現DNA超甲基化。與以往的蛋白質編碼基因結果相同，大多數 H3K27me3豐富的miRNA沒有被檢測到DNA超甲基化，並且反之亦然。 敲除EZH2基因引起H3K27me3水平廣泛下降,並且恢復 H3K27me3抑制的 miRNA表達，而DNA去甲基化劑5-氮雜 -2'-脫氧胞苷 (5-aza-dC) 卻不能重新啟動他們的表達，進一步顯示EZH2基因介導的H3K27me3引發miRNA沉默的機制是獨立存在的。然而，一些過往研究證明有腫瘤抑制功能的miRNA，包括 miR-9-1，miR-9-2和 miR-9-3 被發現同時被 H3K27me3和DNA甲基化調節。我們進一步發現，在肝瘤細胞中，miR-9 特異性調控致癌性的基因結合核因 (NF-κB) 信號通路，並且與配對的非腫瘤肝組織相比，miR-9 的表達在大約一半的原發性肝癌腫瘤（五十九分之三十零）中顯著被壓抑。 / 為了調查在H3K27me3介導的miRNA基因沉默中參與的轉錄因子，我們應用轉錄因子結合位點分析的方法檢查H3K27me3結合蛋白編碼和miRNA基因的調控區域。在包括miR-9亞型的miRNA中，滎陽 1（YY1）的結合位點在這些調控區域中反覆出現並有很高的代表性。定量芯片聯合聚合酶鏈反應結果顯示，在Hep3B細胞中，敲除YY1不僅大大降低了自身的結合力，同時在 miR-9-1，miR-9-2和 miR-9-3 的啟動子中，EZH2基因和H3K27me3結合也大大降低了。尤其重要的是，敲除YY1可以顯著地重新激活他們的表達，表明在肝癌細胞中YY1在EZH2基因介導的的miR-9 表觀遺傳沉默中發揮重要作用。功能研究證明，下調YY1能夠抑制肝癌細胞的增殖，增加細胞凋亡和減少體內的腫瘤生長。定量實時聚合酶鏈反應進一步證實在miR-9 被下調的子集肝癌腫瘤中，有超過85的樣本顯示YY1和EZH2基因同時過量表達，表明我們的研究結果具有臨床相關性。 / 總之，我們完整的分析表明miRNA的調控在肝癌上的獨特表觀遺傳模式。 H3K27me3介導的miRNA沉默可由擁有致癌功能的YY1誘發，它亦可能代表一個可能公認的肝癌癌基因。綜合表觀遺傳和miRNA表達的轉錄控制的結果，能夠提高我們對肝癌發生發展的認識和闡明利用表觀遺傳方法針對性治療肝癌的新的發展方向。 / Although the biological pathways by which mis-regulated microRNAs (miRNAs) contribute to the development of hepatocellular carcinoma (HCC) have been extensively investigated, little is known about the upstream regulatory mechanisms. Previous studies demonstrated that EZH2-mediated histone H3 lysine 27 trimethylation (H3K27me3) is a mechanism of tumor-suppressor gene silencing in cancer that is potentially independent of promoter DNA methylation. On the other hand, gene repression can be associated with both H3K27 and DNA methylation. However, the mechanisms underlying miRNA silencing, particularly in HCC, are poorly understood. / In this study, we used an integrated genome-wide location and expression analysis to investigate the epigenetic and transcriptional controls of miRNA expression in HCC cells. Chromatin immunoprecipitation (ChIP) coupled with human promoter microarrays revealed that 8.4 and 12.2% of the interrogated miRNA were enriched with H3K27me3 in Hep3B and HKCI-8 HCC cells, respectively. On the other hand, Methyl-binding domain capture coupled with microarray (MethylCap-chip) uncovered that 15.5 and 14.7% of miRNA were hypermethylated in Hep3B and HKCI-8 HCC cells, respectively. Consistent with previous observation on protein-coding genes, most of the miRNAs enriched with H3K27me3 had no detectable DNA hypermethylation and vice versa. Knockdown of EZH2 decreased global H3K27me3 level and restored expression of the H3K27me3-targeted miRNAs while the DNA demethylating agent 5-aza-2’-deoxycytidine (5-aza-dC) did not reactivate their expression, further suggesting the independence of EZH2-mediated H3K27me3 in miRNA silencing. Nevertheless, a few miRNAs reported to exhibit tumor-suppressive functions including miR-9-1, miR-9-2 and miR-9-3 were found to be regulated by both H3K27me3 and DNA methylation. We further found that miR-9 targeted the oncogenic NF-κB signaling pathway in HCC cells and were significantly repressed in approximately half of the primary HCC tumors (30/59) compared to the paired non-tumor liver tissues. / To investigate the involvement of transcription factors in H3K27me3-mediated gene silencing of miRNAs, the regulatory regions of H3K27me3-bound protein-coding and miRNA genes were submitted to transcription factor binding site analysis. The binding sites for Ying Yang 1 (YY1) were recurrently over-represented in these loci including the miR-9 isoforms. Quantitative ChIP-PCR demonstrated that knockdown of YY1 in Hep3B cells not only significantly reduced its own binding, but also the EZH2 and H3K27me3 promoter occupancy at miR-9-1, miR-9-2 and miR-9-3. Importantly, their expression levels were significantly reactivated by YY1 knockdown, suggesting that YY1 plays part in the EZH2-mediated epigenetic silencing of miR-9 in HCC cells. Functionally, down-regulation of YY1 was shown to inhibit HCC cell proliferation, increase cell apoptosis and reduce tumor growth in vivo. Quantitative RT-PCR further demonstrated that YY1 and EZH2 were concordantly over-expressed in over 85% of the same subset of HCC tumors that exhibited miR-9 down-regulation, demonstrating the clinical relevance of our findings. / In conclusion, our integrated analysis demonstrated differential epigenetic patterns of miRNA regulation in HCC. H3K27me3-mediated silencing of miRNAs may be initiated by YY1, which possesses oncogenic functions and may represent a putative HCC oncogene. The findings of combinatorial epigenetic and transcriptional control of miRNA expression enhance our understanding of hepatocarcinogenesis and shed light on the development of novel epigenetic targeted therapy of HCC. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Tsang, Pui Fong. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 111-121). / Abstracts also in Chinese. / Abstract (English version) --- p.i / Abstract (Chinese version) --- p.iv / Acknowledgements --- p.vi / table of contents --- p.vii / List of tables --- p.x / List of Figures --- p.xi / Abbreviations --- p.xiii / Chapter CHAPTER 1 --- INTRODUCTION4 / Chapter 1.1 --- Hepatocellular carcinoma --- p.1 / Chapter 1.1.1 --- Epidemiology --- p.1 / Chapter 1.1.2 --- Etiology --- p.2 / Chapter 1.2 --- Epigenetic mechanisms --- p.3 / Chapter 1.2.1 --- Epigenetic silencing by DNA hypermethylation --- p.3 / Chapter 1.2.2 --- Epigenetic silencing by Polycomb group protein --- p.5 / Chapter 1.2.3 --- Interplay between H3K27me3 and DNA hypermethylation --- p.7 / Chapter 1.3 --- microRNA --- p.10 / Chapter 1.3.1 --- Transcriptional gene silencing by miRNA --- p.11 / Chapter 1.3.2 --- miRNA and cancers --- p.12 / Chapter 1.3.3 --- miRNA and liver cancer --- p.13 / Chapter 1.4 --- Signal transduction pathway and cancers --- p.14 / Chapter 1.5 --- Aims of study --- p.15 / Chapter CHAPTER 2 --- MATERIALS AND METHODS / Chapter 2.1 --- Cell lines --- p.16 / Chapter 2.2 --- Clinical samples --- p.16 / Chapter 2.3 --- Plasmid DNA transfection --- p.16 / Chapter 2.4 --- Small interfering RNA transfection --- p.17 / Chapter 2.5 --- Extraction of total RNA --- p.19 / Chapter 2.6 --- Western blot analysis --- p.19 / Chapter 2.7 --- Quantitative RT-PCR --- p.20 / Chapter 2.8 --- miRNA Real Time RT-PCR --- p.22 / Chapter 2.9 --- ChIP-chip assay --- p.24 / Chapter 2.10 --- MethylCap-chip --- p.27 / Chapter 2.11 --- miRNA microarray --- p.28 / Chapter 2.12 --- ChIP Assay and Quantitative ChIP-PCR Assay --- p.28 / Chapter 2.13 --- Colony formation assay --- p.33 / Chapter 2.14 --- Cell proliferation assay --- p.33 / Chapter 2.15 --- Annexin V apoptosis assay --- p.34 / Chapter 2.16 --- Cancer 10-pathway reporter array --- p.34 / Chapter 2.16.1 --- Transfection of siEZH2 with 5-aza-dC treatment --- p.34 / Chapter 2.16.2 --- Transfection of siYY1 and pcDNA3-miR9 plasmid --- p.35 / Chapter 2.16.3 --- Luciferase reporter array --- p.35 / Chapter 2.17 --- Animal Studies --- p.36 / Chapter 2.18 --- Statistical Analysis --- p.36 / Chapter CHAPTER 3 --- Results / Chapter 3.1 --- Occupancy of miRNA genes by epigenetic marks in HCC cells --- p.37 / Chapter 3.1.1 --- Identification of H3K27me3-occupied miRNAs --- p.37 / Chapter 3.1.2 --- Identification of DNA methylation-occupied miRNAs --- p.41 / Chapter 3.1.3 --- Relationship between H3K27me3 and DNA methylation occupancy of miRNAs in HCC cells --- p.45 / Chapter 3.2 --- Regulation of miRNA expression by H3K27me3 and DNA methylation in HCC cells --- p.51 / Chapter 3.3 --- Epigenetic regulation and molecular function of miR-9 in HCC cells --- p.56 / Chapter 3.3.1 --- Confirmation of H3K27me3 and DNA methylation occupancy in miR-9 genes --- p.59 / Chapter 3.3.2 --- Synergistic reactivation of miR-9 upon removal of epigenetic marks --- p.62 / Chapter 3.3.3 --- Effect of miR-9 re-expression on NFKB1 (p50) expression and NF-κB signaling in HCC cells --- p.66 / Chapter 3.4 --- Role of the transcription factor YY1 in the epigenetic regulation of miR-9 --- p.72 / Chapter 3.4.1 --- Identification of transcription factors involved in the regulation of H3K27me3-bound genes --- p.72 / Chapter 3.4.2 --- Occupancy of YY1 on miR-9 in HCC cells --- p.75 / Chapter 3.4.3 --- Effects of YY1 on EZH2/H3K27me3 occupancy and expression of miR-9 --- p.78 / Chapter 3.4.4 --- Effects of YY1 on p50/p65 expression and NF-κB signaling in HCC cells --- p.81 / Chapter 3.5 --- Functional significance of YY1 in HCC --- p.84 / Chapter 3.5.1 --- Effect of YY1 on HCC cell growth --- p.84 / Chapter 3.5.2 --- Effect of YY1 on HCC cell apoptosis --- p.87 / Chapter 3.5.3 --- Effect of YY1 on HCC cell growth in vivo --- p.90 / Chapter 3.5.4 --- Expressions of YY1, EZH2 and miR-9 on clinical HCC samples --- p.92 / Chapter CHAPTER 4 --- DISCUSSION / Chapter 4.1 --- Independence of EHZ2-mediated H3K27me3 and DNA methylation --- p.97 / Chapter 4.2 --- Concordant H3K27 and DNA methylation-mediated silencing of miR-9 --- p.101 / Chapter 4.3 --- Ectopic expression of miR-9 inhibits NF-kB signaling in HCC cells --- p.102 / Chapter 4.4 --- YY1 is involved in the regulation of H3K27me3-bound genes --- p.103 / Chapter 4.5 --- Knockdown of YY1 inhibits NF-kB signaling in HCC --- p.105 / Chapter 4.6 --- Clinical relevance and therapeutic significance of miR-9 silencing by YY1-mediated recruitment of EZH2 --- p.106 / Chapter 4.7 --- Limitations and future studies --- p.109 / REFERENCES --- p.111 / PUBLICATION --- p.122

Pang, Pei Shin. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2009. / Includes bibliographical references (leaves 96-109). / Abstract also in Chinese. / Acknowledgement --- p.i / Abstract (English) --- p.ii / Abstract (Chinese) --- p.iv / Table of Content --- p.vi / List of Figures --- p.x / List of Tables --- p.xiii / Abbreviations --- p.xiv / Chapter CHAPTER ONE: --- INTRODUCTION --- p.1 / Chapter 1.1 --- Incidence --- p.2 / Chapter 1.2 --- Etiological Factors --- p.7 / Chapter 1.2.1 --- Viral Hepatitis Infection --- p.7 / Chapter 1.2.1.1 --- Hepatitis B Virus --- p.10 / Chapter 1.2.1.2 --- Hepatitis C Virus --- p.11 / Chapter 1.2.2 --- Cirrhosis and Chronic Inflammation --- p.14 / Chapter 1.2.3 --- Dietary Aflatoxin Contamination --- p.16 / Chapter 1.2.4 --- Obesity --- p.16 / Chapter 1.3 --- Genomic Aberrations of HCC --- p.17 / Chapter 1.3.1 --- Chromosomal Imbalances --- p.17 / Chapter 1.3.2 --- Candidate Tumour Suppressor Genes and Oncogenes in HCC --- p.18 / Chapter 1.3.2.1 --- Chr 1q21-q22 gain --- p.18 / Chapter 1.3.2.2 --- Chr 8p21-p23 loss and 8q21-q24 gain --- p.18 / Chapter 1.3.2.3 --- Chr 13ql2-ql4 loss --- p.19 / Chapter 1.3.2.4 --- Chr 17pl3 loss --- p.20 / Chapter 1.3.3 --- Chromosomal Rearrangement --- p.21 / Chapter 1.4 --- Cell Lines as In-vitro Study Models --- p.22 / Chapter 1.5 --- Aim of study --- p.23 / Chapter 1.5.1 --- Objectives --- p.24 / Chapter CHAPTER TWO: --- MATERIALS AND METHODS --- p.25 / Chapter 2.1 --- Materials --- p.26 / Chapter 2.2 --- Cell Lines and Cell Culture --- p.29 / Chapter 2.2.1 --- Cell Lines --- p.29 / Chapter 2.2.2 --- Cell Culture --- p.29 / Chapter 2.3 --- Comparative Genomic Hybridization (CGH) --- p.30 / Chapter 2.4 --- Spectral Karyotyping (SKY) --- p.31 / Chapter 2.5 --- Expression Profiling --- p.34 / Chapter 2.6 --- Functional Investigations --- p.37 / Chapter 2.6.1 --- Growth Kinetics --- p.37 / Chapter 2.6.2 --- Cytotoxic Assay --- p.38 / Chapter CHAPTER THREE: --- RESULTS --- p.39 / Chapter 3.1 --- Molecular Cytogenetic Analysis --- p.40 / Chapter 3.1.1 --- Comparative Genomic Hybridization (CGH) --- p.40 / Chapter 3.1.1.1 --- Introduction --- p.40 / Chapter 3.1.1.2 --- CGH Results --- p.41 / Chapter 3.1.1.3 --- Clustering Analysis of CGH Data --- p.49 / Chapter 3.1.2 --- Spectral Karyotyping (SKY) --- p.51 / Chapter 3.1.2.1 --- Introduction --- p.51 / Chapter 3.1.2.2 --- Ploidy Status --- p.51 / Chapter 3.1.2.3 --- Structural Rearrangements --- p.54 / Chapter 3.1.2.4 --- Chromosomes Susceptible to Further Rearrangements --- p.63 / Chapter 3.1.2.5 --- Maintained Common HCC Translocations --- p.65 / Chapter 3.2 --- Expression Profiling --- p.67 / Chapter 3.2.1 --- Introduction --- p.67 / Chapter 3.2.2 --- Gene Expression Profiling --- p.69 / Chapter 3.2.3 --- The Ontologies of Deregulated Genes --- p.71 / Chapter 3.2.4 --- Maintained Biological Pathways --- p.74 / Chapter 3.3 --- Functional Investigation --- p.76 / Chapter 3.3.1 --- Introduction --- p.76 / Chapter 3.3.2 --- Cell Morphology --- p.76 / Chapter 3.3.3 --- Growth Kinetics --- p.79 / Chapter 3.3.4 --- Cytotoxic Assay --- p.83 / Chapter CHAPTER FOUR: --- DISCUSSION --- p.86 / Chapter 4.1 --- Introduction --- p.87 / Chapter 4.2 --- Molecular Cytogenetic Analysis --- p.87 / Chapter 4.3 --- Expression Profiling --- p.91 / Chapter 4.4 --- Conclusion --- p.94 / REFERENCES --- p.95

Array-CGH analysis of the gastric cancer cell lines suggested that TXNIP loci were intact, suggesting that allelic loss might not be the major mechanism responsible for the downregulation of TXNIP in these cells. Furthermore, our data suggested that promoter hypermethylation of TXNIP may not be an important epigenetic mechanism that regulate the silencing of this gene. Chromatin immunoprecipitation (ChIP) assay revealed that SAHA induced hyperacetylation of histone H3 and H4 at the 5' flanking region of TXNIP gene, suggesting SAHA could promote TXNIP gene transcription via modification of histones located at the promoter region. Our data revealed that the loss or reduced expression of TXNIP in gastric cancer cells is associated with epigenetic histone acetylation mechanism. / Gastric cancer is a common cancer especially in Asian countries and is associated with high morbidity and mortality. Epigenetic inactivation of tumor suppressor is a common mechanism involved in carcinogenesis of a variety of human cancers and recent evidence suggested that targeting epigenetic modifications may be an approach to combat cancer. Our group and others have demonstrated frequent promoter methylation of cancer related genes in gastric cancer. In this study, we aim to identify cancer associated genes regulated by another important epigenetic mechanism, namely histone acetylation. / In addition, we demonstrated that over-expression of TXNIP significantly reduced cell migration ability and inhibited cell invasiveness in gastric cancer cells. Furthermore, absence or reduced expression of TXNIP in gastric cancer was associated with diffuse-type gastric cancer, advanced stage disease and predicted a poor disease specific survival. The findings supported that TXNIP is a functional tumor suppressor gene and may be a potential biomarker in gastric cancer. / We analyzed 25 paired gastric cancer and non-cancer gastric mucosa and found that expression of TXNIP mRNA level was reduced in 84% of gastric cancer and was significantly downregulated as compared to the paired non-cancer gastric tissues (p=0.002). Expression of TXNIP protein by western blot was down-regulated in 3 out of 5 cases. Furthermore, by immunohistochemical staining of TXNIP in tissue array containing 150 cases of gastric cancer also showed frequent down-regulation of TXNIP expression and &sim;26% with complete lack of TXNIP expression. / We first showed that suberoylanilide hydroxamic acid (SAHA), a well known histone deacetylase inhibitor, has anti-proliferative effect in a panel of gastric cancer cell lines (MKN1, MKN7, MKN28, MKN45, SNU1, SNU16, AGS, N87 and KatoIII cells). We compared gene expression profiles of SAHA treated vs control AGS cells to identify a set of genes that were differentially upregulated by SAHA treatment. Based on our microarray analysis in nine gastric cancer cell lines (MKN1, MKN7, MKN28, MKN45, SNU1, SNU16, AGS, N87 and KatoIII) and normal gastric tissues, a set of commonly downregulated genes in gastric cancer cells was elucidated. Analysis of these data sets with subsequent confirmation using real-time PCR analysis, genes that were downregulated in gastric cancer cells but upregulated upon SAHA treatment were identified. Among these selected genes, Thioredoxin Interacting Protein (also known as VDUP-1/TBP2/TXNIP ) was down-regulated in all cancer cell lines tested, and its protein expression was significantly induced by SAHA treatment in a numbers of gastric cancer cell lines including AGS, MKN1, MKN45, N87 and KatoIII. Thus, we focused on the TXNIP in the subsequent studies. / Tang, Angie. / Adviser: To Ka Fai. / Source: Dissertation Abstracts International, Volume: 72-04, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2010. / Includes bibliographical references (leaves 180-202). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.