XENOESTROGEN-SPECIFIC MECHANISMS OF DEVELOPMENTAL REPROGRAMMING CORRELATE WITH GENE EXPRESSION AND TUMOR DEVELOPMENT

Environmental exposures during sensitive windows of development can reprogram normal physiological responses and alter disease susceptibility later in life in a process known as developmental reprogramming. We have shown that neonatal exposure to the xenoestrogen diethylstilbestrol (DES) can developmentally reprogram the reproductive tract in genetically susceptible Eker rats giving rise to complete penetrance of uterine leiomyoma. Based on this, we hypothesized that xenoestrogens, including genistein (GEN) and bisphenol A (BPA), reprogram estrogen-responsive gene expression in the myometrium and promote the development of uterine leiomyoma. We proposed the mechanism that is responsible for the developmental reprogramming of gene expression was through estrogen (E2)/ xenoestrogen inducedrapid ER signaling, which modifies the histone methyltransferase Enhancer of Zeste homolog 2 (EZH2) via activation of the PI3K/AKT pathway. We further hypothesized that there is a xenostrogen-specific effect on this pathway altering patterns of histone modification, DNA methylation and gene expression. In addition to our novel finding that E2/DES-induced phosphorylation of EZH2 by AKT reduces the levels of H3K27me3 in vitro and in vivo, this work demonstrates in vivo that a brief neonatal exposure to GEN, in contrast to BPA, activates the PI3K/AKT pathway to regulate EZH2 and decreases H3K27me3 levels in the neonatal uterus. Given that H3K27me3 is a repressive mark that has been shown to result in DNA methylation and gene silencing we investigated the methylation of developmentally reprogrammed genes. In support of this evidence, we show that neonatal DES exposure in comparison to VEH, leads to hypomethylation of the promoter of a developmentally reprogrammed gene, Gria2, that become hyper-responsive to estrogen in the adult myometrium indicating vi that DES exposure alter gene expression via chromatin remodeling and loss of DNA methylation. In the adult uterus, GEN and BPA exposure developmentally reprogrammed expression of estrogen-responsive genes in a manner opposite of one another, correlating with our previous data. Furthermore, the ability of GEN and BPA to developmental reprogram gene expression correlated with tumor incidence and multiplicity. These data show that xenoestrogens have unique effects on the activation of non-genomic signaling in the developing uterus that promotes epigenetic and genetic alterations, which are predictive of developmental reprogramming and correlate with their ability to modulate hormone-dependent tumor development.

An Sp1/Sp3 site polymorphism associated with hypermethylation of the candidate tumor suppressor gene RIL in cancer

Gene silencing due to promoter methylation is an alternative to mutations and deletions, which inactivate tumor suppressor genes (TSG) in cancer. We identified RIL by Methylated CpG Island Amplification technique as a novel aberrantly methylated gene. RIL is expressed in normal tissues and maps to the 5q31 region, frequently deleted in leukemias. We found methylation of RIL in 55/80 (69%) cancer cell lines, with highest methylation in leukemia and colon. We also observed methylation in 46/80 (58%) primary tumors, whereas normal tissues showed substantially lower degrees of methylation. RIL expression was lost in 13/16 cancer cell lines and was restored by demethylating agent. Screening of 38 cell lines and 13 primary cancers by SSCP revealed no mutations in RIL, suggesting that methylation and LOH are the primary inactivation mechanisms. Stable transfection of RIL into colorectal cancer cells resulted in reduction in cell growth, clonogenicity, and increased apoptosis upon UVC treatment, suggesting that RIL is a good candidate TSG. ^ In searching for a cause of RIL hypermethylation, we identified a 12-bp polymorphic sequence around the transcription start site of the gene that creates a long allele containing 3CTC repeat. Evolutionary studies suggested that the long allele appeared late in evolution due to insertion. Using bisulfite sequencing, in cancers heterozygous for RIL, we found that the short allele is 4.4-fold more methylated than the long allele (P = 0.003). EMSA results suggested binding of factor(s) to the inserted region of the long allele, but not to the short. EMSA mutagenesis and competition studies, as well as supershifts using nuclear extracts or recombinant Sp1 strongly indicated that those DNA binding proteins are Sp1 and Sp3. Transient transfections of RIL allele-specific expression constructs showed less than 2-fold differences in luciferase activity, suggesting no major effects of the additional Sp1 site on transcription. However, stable transfection resulted in 3-fold lower levels of transcription from the short allele 60 days post-transfection, consistent with the concept that the polymorphic Sp1 site protects against time-dependent silencing. Thus, an insertional polymorphism in the RIL promoter creates an additional Sp1/Sp3 site, which appears to protect it from silencing and methylation in cancer. ^

RNA-SEQUENCING APPLICATIONS: GENE EXPRESSION QUANTIFICATION AND METHYLATOR PHENOTYPE IDENTIFICATION

My dissertation focuses on two aspects of RNA sequencing technology. The first is the methodology for modeling the overdispersion inherent in RNA-seq data for differential expression analysis. This aspect is addressed in three sections. The second aspect is the application of RNA-seq data to identify the CpG island methylator phenotype (CIMP) by integrating datasets of mRNA expression level and DNA methylation status. Section 1: The cost of DNA sequencing has reduced dramatically in the past decade. Consequently, genomic research increasingly depends on sequencing technology. However it remains elusive how the sequencing capacity influences the accuracy of mRNA expression measurement. We observe that accuracy improves along with the increasing sequencing depth. To model the overdispersion, we use the beta-binomial distribution with a new parameter indicating the dependency between overdispersion and sequencing depth. Our modified beta-binomial model performs better than the binomial or the pure beta-binomial model with a lower false discovery rate. Section 2: Although a number of methods have been proposed in order to accurately analyze differential RNA expression on the gene level, modeling on the base pair level is required. Here, we find that the overdispersion rate decreases as the sequencing depth increases on the base pair level. Also, we propose four models and compare them with each other. As expected, our beta binomial model with a dynamic overdispersion rate is shown to be superior. Section 3: We investigate biases in RNA-seq by exploring the measurement of the external control, spike-in RNA. This study is based on two datasets with spike-in controls obtained from a recent study. We observe an undiscovered bias in the measurement of the spike-in transcripts that arises from the influence of the sample transcripts in RNA-seq. Also, we find that this influence is related to the local sequence of the random hexamer that is used in priming. We suggest a model of the inequality between samples and to correct this type of bias. Section 4: The expression of a gene can be turned off when its promoter is highly methylated. Several studies have reported that a clear threshold effect exists in gene silencing that is mediated by DNA methylation. It is reasonable to assume the thresholds are specific for each gene. It is also intriguing to investigate genes that are largely controlled by DNA methylation. These genes are called “L-shaped” genes. We develop a method to determine the DNA methylation threshold and identify a new CIMP of BRCA. In conclusion, we provide a detailed understanding of the relationship between the overdispersion rate and sequencing depth. And we reveal a new bias in RNA-seq and provide a detailed understanding of the relationship between this new bias and the local sequence. Also we develop a powerful method to dichotomize methylation status and consequently we identify a new CIMP of breast cancer with a distinct classification of molecular characteristics and clinical features.

Therapeutic Targeting of ATP7B in Ovarian Carcinoma.

PURPOSE: Resistance to platinum chemotherapy remains a significant problem in ovarian carcinoma. Here, we examined the biological mechanisms and therapeutic potential of targeting a critical platinum resistance gene, ATP7B, using both in vitro and in vivo models. EXPERIMENTAL DESIGN: Expression of ATP7A and ATP7B was examined in ovarian cancer cell lines by real-time reverse transcription-PCR and Western blot analysis. ATP7A and ATP7B gene silencing was achieved with targeted small interfering RNA (siRNA) and its effects on cell viability and DNA adduct formation were examined. For in vivo therapy experiments, siRNA was incorporated into the neutral nanoliposome 1,2-dioleoyl-sn-glycero-3-phosphatidylcholine (DOPC). RESULTS: ATP7A and ATP7B genes were expressed at higher levels in platinum-resistant cells compared with sensitive cells; however, only differences in ATP7B reached statistical significance. ATP7A gene silencing had no significant effect on the sensitivity of resistant cells to cisplatin, but ATP7B silencing resulted in 2.5-fold reduction of cisplatin IC(50) levels and increased DNA adduct formation in cisplatin-resistant cells (A2780-CP20 and RMG2). Cisplatin was found to bind to the NH(2)-terminal copper-binding domain of ATP7B, which might be a contributing factor to cisplatin resistance. For in vivo therapy experiments, ATP7B siRNA was incorporated into DOPC and was highly effective in reducing tumor growth in combination with cisplatin (70-88% reduction in both models compared with controls). This reduction in tumor growth was accompanied by reduced proliferation, increased tumor cell apoptosis, and reduced angiogenesis. CONCLUSION: These data provide a new understanding of cisplatin resistance in cancer cells and may have implications for therapeutic reversal of drug resistance.

Mammalian miRNA RISC recruits CAF1 and PABP to affect PABP-dependent deadenylation.

MicroRNAs (miRNAs) inhibit mRNA expression in general by base pairing to the 3'UTR of target mRNAs and consequently inhibiting translation and/or initiating poly(A) tail deadenylation and mRNA destabilization. Here we examine the mechanism and kinetics of miRNA-mediated deadenylation in mouse Krebs-2 ascites extract. We demonstrate that miRNA-mediated mRNA deadenylation occurs subsequent to initial translational inhibition, indicating a two-step mechanism of miRNA action, which serves to consolidate repression. We show that a let-7 miRNA-loaded RNA-induced silencing complex (miRISC) interacts with the poly(A)-binding protein (PABP) and the CAF1 and CCR4 deadenylases. In addition, we demonstrate that miRNA-mediated deadenylation is dependent upon CAF1 activity and PABP, which serves as a bona fide miRNA coactivator. Importantly, we present evidence that GW182, a core component of the miRISC, directly interacts with PABP via its C-terminal region and that this interaction is required for miRNA-mediated deadenylation.

Attenuation of fibrosis in vitro and in vivo with SPARC siRNA.

INTRODUCTION: SPARC is a matricellular protein, which, along with other extracellular matrix components including collagens, is commonly over-expressed in fibrotic diseases. The purpose of this study was to examine whether inhibition of SPARC can regulate collagen expression in vitro and in vivo, and subsequently attenuate fibrotic stimulation by bleomycin in mouse skin and lungs. METHODS: In in vitro studies, skin fibroblasts obtained from a Tgfbr1 knock-in mouse (TBR1CA; Cre-ER) were transfected with SPARC siRNA. Gene and protein expressions of the Col1a2 and the Ctgf were examined by real-time RT-PCR and Western blotting, respectively. In in vivo studies, C57BL/6 mice were induced for skin and lung fibrosis by bleomycin and followed by SPARC siRNA treatment through subcutaneous injection and intratracheal instillation, respectively. The pathological changes of skin and lungs were assessed by hematoxylin and eosin and Masson's trichrome stains. The expression changes of collagen in the tissues were assessed by real-time RT-PCR and non-crosslinked fibrillar collagen content assays. RESULTS: SPARC siRNA significantly reduced gene and protein expression of collagen type 1 in fibroblasts obtained from the TBR1CA; Cre-ER mouse that was induced for constitutively active TGF-beta receptor I. Skin and lung fibrosis induced by bleomycin was markedly reduced by treatment with SPARC siRNA. The anti-fibrotic effect of SPARC siRNA in vivo was accompanied by an inhibition of Ctgf expression in these same tissues. CONCLUSIONS: Specific inhibition of SPARC effectively reduced fibrotic changes in vitro and in vivo. SPARC inhibition may represent a potential therapeutic approach to fibrotic diseases.

PREDICTION OF DNA METHYLATION BASED ON GENOMIC ARCHITECTURE AND APPLICATIONS OF POSITIONAL WEIGHT MATRICES

Gene silencing due to epigenetic mechanisms shows evidence of significant contributions to cancer development. We hypothesis that the genetic architecture based on retrotransposon elements surrounding the transcription start site, plays an important role in the suppression and promotion of DNA methylation. In our investigation we found a high rate of SINE and LINEs retrotransposon elements near the transcription start site of unmethylated genes when compared to methylated genes. The presence of these elements were positively associated with promoter methylation, contrary to logical expectations, due to the malicious effects of retrotransposon elements which insert themselves randomly into the genome causing possible loss of gene function. In our genome wide analysis of human genes, results suggested that 22% of the genes in cancer were predicted to be methylation-prone; in cancer these genes are generally down-regulated and function in the development process. In summary, our investigation validated our hypothesis and showed that these widespread genomic elements in cancer are highly associated with promoter DNA methylation and may further participate in influencing epigenetic regulation.

The effect of DNA methylation in carcinogenesis

CpG island methylation within single gene promoters can silence expression of associated genes. We first extended these studies to bidirectional gene pairs controlled by single promoters. We showed that hypermethylation of bidirectional promoter-associated CpG island silences gene pairs (WNT9A/CD558500, CTDSPL/BC040563, and KCNK15/BF 195580) simultaneously. Hypomethylation of these promoters by 5-aza-2'-deoxycytidine treatment reactivated or enhanced gene expression bidirectionally. These results were further confirmed by luciferase assays. Methylation of WNT9A/CD558500 and CTDSPL/BC040563 promoters occurs frequently in primary colon cancers and acute lymphoid leukemia, respectively. ^ Next we sought to understand the origins of hypermethylation in cancer. CpG islands associated with tumor suppressor genes are normally free from methylation, but can be hypermethylated in cancer. It remains poorly understood how these genes are protected from methylation in normal tissues. In our studies, we aimed to determine if cis-acting elements in these genes are responsible for this protection, using the tumor suppressor gene p16 as a model. We found that Alu repeats located both upstream and downstream of the p16 promoter become hypermethylated with age. In colon cancer samples, the methylation level is particularly high, and the promoter can also be affected. Therefore, the protection in the promoter against methylation spreading could fail during tumorigenesis. This methylation pattern in p16 was also observed in cell lines of different tissue origins, and their methylation levels were found to be inversely correlated with that of active histone modification markers (H3K4-3me and H3K9-Ac). To identify the mechanism of protection against methylation spreading, we constructed serial deletions of the p16 protected region and used silencing of a neomycin reporter gene to evaluate the protective effects of these fragments. A 126 bp element was identified within the region which exerts bidirectional protection against DNA methylation, independently of its transcriptional activity. The protective strength of this element is comparable to that of the HS4 insulator. During long-term culture, the presence of this element significantly slowed methylation spreading. In conclusion, we have found that an element located in the p16 promoter is responsible for protection against DNA methylation spreading in normal tissues. The failure of protective cis-elements may be a general feature of tumor-suppressor gene silencing during tumorigenesis. ^

CREB trans-activation of disruptor of telomeric silencing-1 mediates forskolin inhibition of CTGF transcription in mesangial cells.

Connective tissue growth factor (CTGF) participates in diverse fibrotic processes including glomerulosclerosis. The adenylyl cyclase agonist forskolin inhibits CTGF expression in mesangial cells by unclear mechanisms. We recently reported that the histone H3K79 methyltransferase disruptor of telomeric silencing-1 (Dot1) suppresses CTGF gene expression in collecting duct cells (J Clin Invest 117: 773-783, 2007) and HEK 293 cells (J Biol Chem In press). In the present study, we characterized the involvement of Dot1 in mediating the inhibitory effect of forskolin on CTGF transcription in mouse mesangial cells. Overexpression of Dot1 or treatment with forskolin dramatically suppressed basal CTGF mRNA levels and CTGF promoter-luciferase activity, while hypermethylating H3K79 in chromatin associated with the CTGF promoter. siRNA knockdown of Dot1 abrogated the inhibitory effect of forskolin on CTGF mRNA expression. Analysis of the Dot1 promoter sequence identified a CREB response element (CRE) at -384/-380. Overexpression of CREB enhanced forskolin-stimulated Dot1 promoter activity. A constitutively active CREB mutant (CREB-VP16) strongly induced Dot1 promoter-luciferase activity, whereas overexpression of CREBdLZ-VP16, which lacks the CREB DNA-binding domain, abolished this activation. Mutation of the -384/-380 CRE resulted in 70% lower levels of Dot1 promoter activity. ChIP assays confirmed CREB binding to the Dot1 promoter in chromatin. We conclude that forskolin stimulates CREB-mediated trans-activation of the Dot1 gene, which leads to hypermethylation of histone H3K79 at the CTGF promoter, and inhibition of CTGF transcription. These data are the first to describe regulation of the Dot1 gene, and disclose a complex network of genetic and epigenetic controls on CTGF transcription.

DEFINITION OF THE LANDSCAPE OF CHROMATIN STRUCTURE AT THE FRATAXIN GENE IN FRIEDREICH’S ATAXIA

Friedreich’s ataxia (FRDA) is caused by the transcriptional silencing of the frataxin (FXN) gene. FRDA patients have expansion of GAA repeats in intron 1 of the FXN gene in both alleles. A number of studies demonstrated that specific histone deacetylase inhibitors (HDACi) affect either histone modifications at the FXN gene or FXN expression in FRDA cells, indicating that the hyperexpanded GAA repeat may facilitate heterochromatin formation. However, the correlation between chromatin structure and transcription at the FXN gene is currently limited due to a lack of more detailed analysis. Therefore, I analyzed the effects of the hyperexpanded GAA repeats on transcription status and chromatin structure using lymphoid cell lines derived from FRDA patients. Using chromatin immunoprecipitation and quantitative PCR, I observed significant changes in the landscape of histone modifications in the vicinity of the GAA tract in FRDA cells relative to control cells. Similar epigenetic changes were observed in GFP reporter construct containing 560 GAA repeats. Further, I detected similar levels of FXN pre-mRNA at a region upstream of hyperexpanded GAA repeats in FRDA and control cells, indicating similar efficiency of transcription initiation in FRDA cells. I also showed that histone modifications associated with hyperexpanded GAA repeats are independent of transcription progression using the GFP reporter system. My data strongly support evidence that FXN deficiency in FRDA patients is consequence of defective transition from initiation to elongation of FXN transcription due to heterochromatin-like structures formed in the proximity of the hyperexpanded GAAs.

Mechanism of methylation gene reactivation in human cancer cells

Epigenetic silencing of tumor suppressor genes by DNA hypermethylation at promoter regions is a common event in carcinogenesis and tumor progression. Abrogation of methylation and reversal of epigenetic silencing is a very potent way in cancer treatment. However, the reactivation mechanisms are poorly understood. In this study, we first developed a cell line model system named YB5, derived from SW48 cancer cell line, which bears one copy of stably integrated EGFP gene on Chromosome 1p31.1 region. The GFP gene expression is transcriptionally silenced due to the hypermethylated promoter CMV. However, the GFP expression can be restored using demethylating agent 5-aza-2' deoxycytidine (DAC), and detected by FACS and fluorescent microscopy. Using this system, we observed the heterogeneous reactivation induced by DAC treatment. After flow sorting, GFP negative cells exhibited similar level of incomplete demethylation compared to GFP positive cells on repetitive LINE1 element, tumor suppressor genes such as P16, CDH13, and RASSF1a, and CMV promoter as well. However, the local chromatin of CMV-GFP locus altered to an open structure marked by high H3 lysine 9 acetylation and low H3 lysine 27 tri-methylation in GFP positive cells, while the GFP negative cells retained mostly the original repressive marks. Thus, we concluded that DAC induced DNA hypomethylation alone does not directly determine the level of re-expression, and the resetting of the local chromatin structure under hypomethylation environment is required for gene reactivation. Besides, a lentivirus vector-based shRNA screening was performed using the YB5 system. Although it is the rare chance that vector lands in the neighboring region of GFP, we found that the exogenous vector DNA inserted into the upstream region of GFP gene locus led to the promoter demethylation and reactivated the silenced GFP gene. Thus, epigenetic state can be affected by changing of the adjacent nucleic acid sequences. Further, this hypermethylation silenced system was utilized for epigenetic drug screening. We have found that DAC combined with carboplatin would enhance the GFP% yield and increase expression of other tumor suppressor genes than DAC alone, and this synergistic effect may be related to DNA repair process. In summary, these studies reveal that reversing of methylation silencing requires coordinated alterations of DNA methylation, chromatin structure, and local microenvironment. ^

Post-transcriptional Regulation of Mammalian Gene Expression in Non-coding Region of Target RNA

Tumor Suppressor Candidate 2 (TUSC2) is a novel tumor suppressor gene located in the human chromosome 3p21.3 region. TUSC2 mRNA transcripts could be detected on Northern blots in both normal lung and some lung cancer cell lines, but no endogenous TUSC2 protein could be detected in a majority of lung cancer cell lines. Mechanisms regulating TUSC2 protein expression and its inactivation in primary lung cancer cells are largely unknown. We investigated the role of the 5’- and 3’-untranslated regions (UTRs) of the TUSC2 gene in the regulation of TUSC2 protein expression. We found that two small upstream open-reading frames (uORFs) in the 5’UTR of TUSC2 could markedly inhibit the translational initiation of TUSC2 protein by interfering with the “scanning” of the ribosome initiation complexes. Site-specific stem-loop array reverse transcription-polymerase chain reaction (SLA-RT-PCR) verified several micoRNAs (miRNAs) targeted at 3’UTR and directed TUSC2 cleavage and degradation. In addition, we used the established let-7-targeted high mobility group A2 (Hmga2) mRNA as a model system to study the mechanism of regulation of target mRNA by miRNAs in mammalian cells under physiological conditions. There have been no evidence of direct link between mRNA downregulation and mRNA cleavages mediated by miRNAs. Here we showed that the endonucleolytic cleavages on mRNAs were initiated by mammalian miRNA in seed pairing style. Let-7 directed cleavage activities among the eight predicted potential target sites have varied efficiency, which are influenced by the positional and the structural contexts in the UTR. The 5’ cleaved RNA fragments were mostly oligouridylated at their 3’-termini and accumulated for delayed 5’–3’ degradation. RNA fragment oligouridylation played important roles in marking RNA fragments for delayed bulk degradation and in converting RNA degradation mode from 3’–5’ to 5’–3’ with cooperative efforts from both endonucleolytic and non-catalytic miRNA-induced silencing complex (miRISC). Our findings point to a mammalian miRNA-mediated mechanism for the regulation of mRNA that miRNA can decrease target mRNA through target mRNA cleavage and uridine addition