Benjamin Tycko - US grants

I have detected, cloned, and sequenced naturally occurring products of interchromosomal recombination between gamma (gamma-) and delta (delta-) T cell receptor (TCR) genes by polymerase chain reaction (PCR) DNA amplification. These chimeric gene rearrangements (trans-rearrangements), which have the general structures Vgamma-(Ddelta)-Jdelta and Vdelta-Jgamma, are present in normal human thymus at frequencies of greater than 10-4 and in normal human peripheral lymphoid tissues at frequencies greater than 10- 5. Many of the cloned rearrangements show potentially protein-coding open reading frames, and blot hybridization of RNA PCR reaction products suggests the presence of chimeric Vgamma-Cdelta transcripts. I have found that these gamma-delta trans-rearrangements are also present in mouse thymus. As primary objectives I propose to explore both the functional consequence of chimeric gene rearrangements and the mechanism of their occurrence. Strategies for the construction of clonal murine lymphocyte cell lines bearing and expressing trans-rearrangements are outlined. The availability of clonal lines will facilitate studies of the functional properties of predicted chimeric Vgamma-Cdelta and Vdelta-Cgamma TCR proteins. The mechanism of trans-rearrangement in developing murine thymocytes will be investigated by analysis of hypothetical circular DNA excision products and potential intermediate DNa structures. An additional objective is the detection and characterization of Vgamma-Jalpha chimeric rearrangements, theoretically predicted to derive form Vgamma-(Ddelta)- Jdelta rearrangements during thymocyte maturation, and other hypothetical chimeric TCR rearrangements such as Vbeta-Jalpha, Vgamma-(Dbeta)-Jbeta and their corresponding reciprocal products.

Monoallelic gene expression, resulting from genomic imprinting, appears to have important consequences both in normal development and in human genetic disease and cancer. However, to date there have been very few studies of monoallelic gene expression at defined loci. We have shown that the human H19 gene, at chromosome 11p15, is monoallelically expressed. The 11p15 region is of particular interest in that maternal alleles are commonly deleted in childhood tumors and the region is therefore predicted to contain one or more imprinted tumor suppressor genes. We plan to study the mechanism and consequences of monoallelic gene expression in humans, focusing on H19 and flanking genes, and to explore strategies for isolating additional imprinted genes. To determine whether 11p15 imprinting is regional or local we will clone the genes immediately flanking H19 and test them for monoallelic expression. We will test the possible role of DNA methylation in imprinting by analyzing allele-specific methylation in and around the H19 gene and by carrying out functional assays for the effect of methylation on H19 expression. We will search for candidate imprinting "initiator" sequences in H19 and flanking DNA using a transfection assay in embryonal stem cells. We will characterize a novel phenomenon, tissue-specific somatic allele switching, which we have observed for human H19 and which suggests methylation threshold and feedback effects in maintenance of monoallelic gene expression. To explore the biological function of H19, and to determine whether this gene might have tumor suppressor activity, we will examine the ability of inducible H19 expression vectors to revert the phenotype of Wilms' tumor and embryonal rhabdomyosarcoma cells. Finally, to isolate additional imprinted genes, we will pursue a candidate gene approach in humans and, concurrently, develop a general strategy, based on arbitrarily-primed polymerase chain reaction (AP-RCR), to clone imprinted genes from interspecific mouse hybrids.

I have detected, cloned, and sequenced naturally occurring products of interchromosomal recombination between gamma (gamma-) and delta (delta-) T cell receptor (TCR) genes by polymerase chain reaction (PCR) DNA amplification. These chimeric gene rearrangements (trans-rearrangements), which have the general structures Vgamma-(Ddelta)-Jdelta and Vdelta-Jgamma, are present in normal human thymus at frequencies of greater than 10-4 and in normal human peripheral lymphoid tissues at frequencies greater than 10- 5. Many of the cloned rearrangements show potentially protein-coding open reading frames, and blot hybridization of RNA PCR reaction products suggests the presence of chimeric Vgamma-Cdelta transcripts. I have found that these gamma-delta trans-rearrangements are also present in mouse thymus. As primary objectives I propose to explore both the functional consequence of chimeric gene rearrangements and the mechanism of their occurrence. Strategies for the construction of clonal murine lymphocyte cell lines bearing and expressing trans-rearrangements are outlined. The availability of clonal lines will facilitate studies of the functional properties of predicted chimeric Vgamma-Cdelta and Vdelta-Cgamma TCR proteins. The mechanism of trans-rearrangement in developing murine thymocytes will be investigated by analysis of hypothetical circular DNA excision products and potential intermediate DNa structures. An additional objective is the detection and characterization of Vgamma-Jalpha chimeric rearrangements, theoretically predicted to derive form Vgamma-(Ddelta)- Jdelta rearrangements during thymocyte maturation, and other hypothetical chimeric TCR rearrangements such as Vbeta-Jalpha, Vgamma-(Dbeta)-Jbeta and their corresponding reciprocal products.

Contributions of environmental versus genetic factors in Alzheimer's disease (AD) are not well understood. Our goal is to test two hypotheses regarding possible genetic etiologies. First, we will examine the role of mutations in the gene for the beta amyloid precursor protein (APP) in case of AD which appear to be sporadic. Our rationale is that this gene has previously been shown to be mutated in rare familial cases. Mutations with more subtle effects on biological activities of APP could convey a predisposition to AD without manifesting sufficient penetrance to produce a familial pattern. We will use polymerase chain reaction (PCR) amplification followed by single strand conformation polymorphism (SSCP) analysis to rapidly screen for base pair alterations in the APP coding region and promoter region in a large population study. The significance of mutations will be evaluated both by biological experiments in vitro and by a statistical analysis of the association of the specific types of mutations with AD in family members of the probands and in the study population as a whole. Second, we will test the possibility that mitotic nondisjunctions leading to low level mosaicism for trisomy 21 might play a role in AD. The frequency of trisomic cells in peripheral blood from living donors and brains from autopsies will be established by in situ hybridization with chromosome-specific probes. If measurable frequencies of trisomic cells are found, we will examine the correlation of the trisomy index (number of trisomic cells/number of disomic cells) with AD, both in terms of likelihood and age of onset of disease. We will also analyze cases of classical mosaic Down's syndrome to ask whether the trisomy index in these cases correlates with the age of onset of dementia.

Data from genetic epidemiological studies suggest that the involvement of genetic variation at the APOE locus in Alzheimer's disease (AD) susceptibility may not be restricted to the well-studied coding polymorphisms, but may also be influenced by polymorphisms in flanking DNA. In particular, we have recently observed an association of the APOC1 HpaI+ allele with AD in groups of people with the "reference" APOE e3/e3 genotype. In this project we will carry out molecular analyses and experiments to test the hypothesis that APOE flanking sequences influence AD susceptibility is that this effect is through the influence of these polymorphisms on allele-specific APOE mRNA expression. First, to test whether AD susceptibility depends on particular combinations of APOE coding and regulatory polymorphisms located in cis on the chromosome, we will develop a method for determining complete haplotypes of polymorphic markers in the APOE promoter, the APOE coding region and the APOC1 promoter and we will generate haplotype data for AD association analysis in the large multi-ethnic population samples discussed in Project 1. Second, using a transfection assay screen, we will test for the existence of functional sequence polymorphisms in the APOE and APOC1 promoters, we will determine the sequence of functional sequence polymorphisms in the APOE and APOC1 promoters, we will determine the sequence of these polymorphisms in the APOE and APOC1 promoters, we will determine the sequence of these polymorphisms and we will apply them as new markers in the haplotype analyses. Third, we will test for functional significance for the individual polymorphisms and haplotypes in vivo by correlating them with allele-specific APOE mRNA levels in human lymphoblastoid cell lines and liver, kidney and brain tissues. Fourth, using apoCI promoter "knockout" mice, we will test a model in which the APOE and APOC1 engage in enhance-competition as an explanation for the observed association of the APOC1 HpaI+ allele with AD. Using transfections and transgenic mice we will also investigate localization of the putative brain-enhancer of APOE and then search for functional polymorphisms in this region. Results of these studies may reveal a basic mechanism for AD which is accessible to pharmacological intervention and may also shed light on the difference sin apparent AD risk who have been observed in different ethnic groups.

DESCRIPTION (provided by applicant): Major efforts have gone into identifying the targets of oncogenic transcription factors, but data from tissue culture experiments have seldom been vetted in primary cancers. Here we address this problem, using Wilms tumor as an experimental system. Wilms tumors occurring in Denys-Drash and WAGR syndromes carry inactivating mutations of the WT1 tumor suppressor gene. In contrast, WT1 mutations are rare in sporadic Wilms tumors. We have confirmed previous data indicating that mutations in beta-catenin (CTNNB1), a component of the Wnt signaling pathway that acts as an oncogene in many of the most common human malignancies, are restricted to the WT1-null tumors. We also find that these mutations are strikingly clustered at codon Ser45. By expression profiling we have identified a panel of genes that distinguish the WT1-null from WT1-positive tumors. This dataset should be a powerful tool to identify downstream targets of WT1 and beta-catenin/TCF, which we hypothesize are enriched among these differentially expressed genes. Aim 1. We will determine whether the Wnt/beta-catenin signaling axis is universally activated in the WT1- null class of Wilms tumors. Aim 2. To narrow the list of WT1 and beta-catenin target genes, we will manipulate WT1 levels, and components of the beta-catenin pathway in tissue culture, and compare the results with our "gold standard" data from the primary Wilms tumors. Aim 3. To validate the candidate beta-catenin target genes in vivo, we will express mutant beta-catenin to the developing kidney, using a gene knock-in approach. By creating an isogenic series, these mice will also allow us to assay for the relative potency of mutation at Ser45 vs. other phosphorylation sites in affecting the proliferative/oncogenic potency of beta-catenin. We will subsequently cross these mice with Wt1-mutant heterozygotes, possibly generating a mouse model for Wilms tumor.

[unreadable] DESCRIPTION (provided by applicant): This application is for partial funding of a conference called "Cancer Genetics and Epigenetics" to be held under the auspices of the Gordon Research Conference (GRC) in mid-January 2003 at the Sheraton Harbortown, Ventura, California. This will be the fourth biennial GRC conference on this topic during a particularly exciting time for the field. Work in all areas covered by this conference is progressing at an ever-increasing pace. This makes communication between researches ever more essential. At this conference, the attendees (130 maximum) working on selected aspects of cancer research and pertinent basic genetic/epigenetic mechanisms will debate presentations on current topics, critique poster presentations, and develop informal discussions. The meeting will feature nine speaker sessions and daily poster sessions. Poster presentations, as well as more than nine of the speakers, will be chosen from the submitted abstracts. Therefore, this conference will provide a good showcase for younger investigators and late-breaking data, as well as featuring internationally renowned senior scientists. In the tradition of this conference series, there will be emphasis on formal discussion, with 15-20 minutes set aside for debate after each talk. This conference will provide a forum for exchange of information between investigators working on the molecular biology of cancer and those working on basic mechanisms of gene regulation. An outline schedule of scientific talks has been prepared based on the unifying themes of epigenetic gene regulation and the molecular responses to DNA damage in cancers and in cancer precursor tissues. These talks will cover selected aspects of tumor suppressor and oncogene function, gene silencing by DNA methylation, genomic imprinting and chromatin alterations, DNA repair and recombination. There will be one session dealing with these same topics in the setting of molecular epidemiology, and two sessions will deal with these topics in breast cancer and hematological cancers. Recent discoveries continue to uncover mechanisms linking these areas, so we expect that cross talk among investigators will yield synergistic advances. [unreadable] [unreadable]

Genetic and epigenetic aberrations are well studied hallmarks of cancer cells. But in addition, an increasing number of reports have suggested that such changes can also occur in the non-neoplastic stromal cells that surround and intermingle with carcinoma cells. There are also recent functional data suggesting that tumorassociated stromal cells, including myofibroblasts and tumor-associated endothelial cells (TECs), can acquire new phenotypes, different from normal stromal cells, which actively contribute to tumor progression. An obvious challenge is to link these two lines of research, pinpointing the specific genetic and epigenetic changes that might account for the new phenotypic characteristics acquired by tumor-associated stromal cells. In this Project we will carry out a genome-wide analysis of chromosomal and sub-chromosomal aneuploidies (DNA copy number aberrations;CNA), loss of heterozygosity (LOH) and gains and losses of DNA methylation (GOM, LOM) in the myofibroblasts that proliferate in human cirrhotic livers and hepatocellular carcinomas, and in stromal myofibroblasts from gastric cancers. To achieve this objective, we will apply a new method, methylation-sensitive SNP chip analysis (MSNP), that we have recently tested and validated for combined genetic and epigenetic profiling in other types of human cancers and normal tissues. First, we will use MSNP to obtain comprehensive genetic (CNA, LOH) and epigenetic (GOM, LOM) profiles of myofibroblasts from normal human livers, cirrhotic human livers and human hepatocellular carcinomas. Second, we will carry out parallel experiments analyzing stromal myofibroblasts from human gastric cancers, comparing the epigenetic and genetic profiles of these cells to control myofibroblasts from normal human stomach. Third, we will validate the MSNP data for loci that show recurrent genetic or epigenetic changes, using independent molecular methods. We expect that the genetic and epigenetic data from this Project will allow the other Projects to formulate and test new biological hypotheses for the role of stromal cells in human liver and gastric cancers, as well as in pre-neoplastic liver cirrhosis.

[unreadable] DESCRIPTION (provided by applicant): Altered DNA methylation is a hallmark of cancer cells, and epigenetic aberrations including gains and losses of DNA methylation also underlie other medical disorders. Such epigenetic aberrations are thought to act in concert with genetic changes, such as chromosome losses and gains, in cancer initiation and progression. In this technology development project we will develop and extensively validate a cost-effective, high-resolution and high-throughput method for simultaneously characterizing genetic aberrations and alterations in DNA methylation genome wide in human cancers. We expect that this method, which we abbreviate MSNP (Methylation-sensitive SNP array analysis), will be useful both for answering fundamental questions such as the relationship between the DNA sequence and its pattern of methylation (genetic/epigenetic interactions and allele-specific DNA methylation) and the relationship between DNA methylation and gene expression, and for achieving important medical objectives including identifying tumor suppressor genes and proto-oncogenes regulated by DNA methylation and finding new and robust markers for cancer diagnosis and detection. Altered DNA methylation is a hallmark of cancer cells, and gains and losses of DNA methylation also underlie other medical disorders. This project is designed to optimize and validate a new method, called MSNP, for detecting and mapping DNA methylation at high resolution across the entire human genome in both normal and cancerous human tissues. Data from MSNP are expected to shed light on cancer pathogenesis and produce new and robust markers for cancer diagnosis and detection. [unreadable] [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Our project is an epigenomic approach to understand the pathogenesis of Alzheimer's disease (AD), a degenerative disorder of neurons that is the most common cause of age-related dementia. Much prior emphasis in AD research has been on genetic studies, which have uncovered the fundamental involvement of the APP, APOE and PS1 genes. But with the exception of polymorphisms in APOE, known coding variants in these genes are quite rare, and the existing data do not fully account for the heritability of AD in the general population. Additional chromosomal loci that influence AD susceptibility have been uncovered by genome scanning, but while several of these loci have been confirmed in multiple studies, the causative or predisposing genetic variants have not been clearly pinpointed. New approaches are needed to get through this roadblock, and we believe that the novel combined epigenetic-genetic strategy that we propose here can be transformative in solving this problem. The equally important question of why AD is inexorably progressive, at least in our current ignorance of effective treatments, also has no clear answer. It is safe to say that progression to neuronal cell death occurs after a threshold of cellular damage, but what the crucial damage is and how it accumulates over time are largely unknown. A component of this damage may be epigenetic, including altered DNA methylation. In this collaborative and multi-disciplinary project we will bring genome-wide epigenomic profiling methods to bear on 2 working hypotheses. First, we postulate that epigenetic aberrations in the brain are involved in the progression of AD, and that genome-wide analysis of DNA methylation in cerebral cortex from early to mid-stage AD cases, compared to control brains, will pinpoint biologically relevant loci that are recurrently affected by gains or losses of net DNA methylation (affecting both alleles) in this disease. Second, our genome-wide epigenetic analysis will pay off, at no additional cost, in the identification of loci with sequence-dependent allele-specific DNA methylation (ASM) and allele- specific gene expression (ASE), thus pinpointing regulatory polymorphisms that act to establish inborn and stable inter-individual differences in gene expression in brain cells, in part through haplotype- dependent epigenetic modifications. We expect that among these sequence variants will be some that affect inter-individual differences in susceptibility to AD. The list of candidate loci from our epigenomic profiling will be brought forward for narrowly focused and thus cost efficient genetic fine mapping in the valuable well characterized cohorts of AD case and controls that are available through the long- standing collaborative relationship of the 2 PI's. Public Health Relevance: Our project is an epigenomic approach to understand the pathogenesis of Alzheimer's disease (AD), a degenerative disorder of neurons that is the most common cause of age-related dementia. By profiling DNA methylation in normal brains and in brains affected by AD we will determine whether recurrent changes in DNA methylation play a role in the inexorable progression of this disease. From the same epigenomic data we will identify a set of genes with genetically determined differences in DNA methylation, which we will directly test as candidates for causing inter-individual differences in susceptibility to AD, using large family-based and case-control cohorts of elders at risk for this disease.

DESCRIPTION (provided by applicant): Alterations in DNA methylation are hallmarks of cancer cells, and epigenetic markers are increasingly viewed as having great potential for diagnosing, classifying and prognosticating cancers and cancer precursor lesions. Thus, a technical challenge is to develop and apply efficient and high-coverage methods to profile DNA methylation genome-wide in human cancers and in the normal precursor tissues of these cancers. We have developed such a method, called MSNP, to characterize DNA methylation genome-wide using Affymetrix single nucleotide polymorphism DNA microarrays. In addition to profiling gains and losses of net DNA methylation (GOM, LOM), a particular strength of MSNP is that it also queries allele-specific DNA methylation (ASM). Here we hypothesize that MSNP can be further developed and optimized as a high resolution method to reveal differences in methylation patterns not only between cancer and normal tissues, but also between normal tissues and the early atypical or dysplastic precursor tissues which eventually give rise to cancers. In this collaborative R21 proposal, with an experienced team of investigators from the Institute for Cancer Genetics and the Departments of Pathology and Biostatistics, we will advance the methodology and applications of MSNP in several ways. Aim 1 is to optimize MSNP for very high density Affymetrix 1.8M (6.0 array) SNP chips, vetting this method by profiling net and allele-specific DNA methylation in human breast cancers and normal breast epithelium. The results of will be verified by independent assays, including high throughput bisulfite sequencing. In this Aim we will develop bioinformatics approaches for tumor class prediction from MSNP data, and develop formats for data annotation and data sharing. Aim 2 is to miniaturize the MSNP method so that high quality genetic and epigenetic data can be obtained from the small amounts of genomic DNA available from laser-capture microdissection (LCM) or manual microdissection (MM). We will establish conditions allowing ASM and net DNA methylation to be determined using genomic DNA obtained by LCM or MM from normal and cancerous breast epithelium. In this aim we will particularly evaluate breast cancer precursor lesions, namely atypical duct epithelial hyperplasias (ADH), which are associated with a high risk for subsequent breast cancer development. Aim 3 is to correlate MSNP data with expression profiling data, to determine whether MSNP can produce a list of candidate DNA sequences, both promoter-associated and non- promoter-associated, in the human genome that may act as novel methylation-sensitive regulatory elements controlling gene expression in normal and cancer tissues.

DESCRIPTION (provided by applicant): Much prior emphasis in research on Alzheimer's disease (AD) has been on genetic studies, which have uncovered the fundamental involvement of the APP, APOE and PS1 genes. But with the exception of polymorphisms in APOE, known coding variants in these genes are quite rare, and the existing data do not fully account for the heritability of late-onset AD (LOAD) in the general population. Additional chromosomal loci putatively influencing AD susceptibility have been uncovered by genome-wide association studies (GWAS) and related high-throughput genomic scans, but replication has been uneven and the causative or predisposing genes have not been clearly pinpointed. Further it has not been possible to gauge the importance of numerous sub-threshold statistical peaks in these scans. New approaches are needed to get through these roadblocks, and we believe that a combined strategy to first obtain lists of loci with sequence-dependent allele-specific methylation (ASM) and/or allele-specific mRNA expression (ASE) in human cerebral cortex, and then overlap this information with statistical peaks from GWAS, will help to solve this problem. We will screen for ASM/ASE in human cerebral cortex using microarray-based and sequencing-based methods. We will validate the strongest candidate loci for ASM/ASE using independent assays, and overlap the list of such loci with data from existing and ongoing GWAS in LOAD. Loci from the intersection of these 2 datasets will be subjected to highly focused and thus cost efficient genetic fine mapping in 3 independent well characterized cohorts of LOAD case and controls. Our reasoning is that sequence-dependent ASM/ASE in or near specific genes in the cerebral cortex will be a robust indicator for the presence of bona fide cis-acting regulatory polymorphisms (rSNPs, rCNPs) that confer inter-individual differences in the expression of specific genes in the human cerebral cortex, and that within this set of genes will be some that cause inter-individual differences in LOAD susceptibility. Reproducible signals from GWAS and related genome scans for LOAD that overlap with loci that show ASM and/or ASE will thus have a strong functional underpinning and warrant further intensive study as biological contributors and potential therapeutic targets in LOAD.

DESCRIPTION (provided by applicant): Inherited genetic risk is a major component of Type I diabetes (T1D). Major international research efforts have identified a large number of T1D susceptibility genes through genome wide association studies (GWAS). Despite these advances, in nearly all cases GWAS "hits" have not yielded insight into protein coding alterations but rather are presumed to reflect regulatory SNPs or haplotypes that have refractory to detection. To address the intractable problem of missing hereditability in the post-GWAS era we have assembled a uniquely qualified interdisciplinary team, both to extract maximum information from GWAS and to pursue "post-GWAS" pathogenic epigenetic events. Our innovative approach to this question focuses on both aberrations of T cells and end- organ vulnerability. We have assembled both T1D immunologists and beta cell biologists to team with molecular geneticists/epigeneticists and an experienced bioinformatician in a systems approach to identify therapeutically targetable genome wide epigenetic events in the relevant cytotoxic T cells and their targets. Using mouse models and human subjects recruited through the diabetes clinical centers at Columbia/Yale, T1D specific epigenetic hits will be identified with clinical predictive utility and mechanistic importance. Therapeutic opportunities for reversal will be pursued using novel antisense approaches. Our approach has already identified key epigenetically regulated immunological targets modulating tolerance (CD3, PD-1), and has developed pioneering T1D monitoring tools to define the prediabetic window of insulitis (hypomethylated insulin). Novel epigenetic-based therapeutic approaches are proposed and tested that represent an innovative strategy for inducing endogenous anti-inflammatory isoforms, including endogenous soluble cytokine receptors and ligand independent inhibitory costimulatory molecules. In addition, we will use allele-specific mapping of DNA methylation and gene expression to help extract maximum information from GWAS and re-sequencing data for T1D susceptibility loci. In sum this project will identify genetic and epigenetic events responsible for T1D susceptibility and disease progression, and pursue therapeutic approaches directed at known and novel genetic and epigenetic targets. PUBLIC HEALTH RELEVANCE: Heritable risk in Type 1 diabetes has been extensively studied but remains poorly defined. We will use epigenetic tools to bolster the genetic data, and to find genetic variants that are responsible for aberrant activity of killer T cells and the vulnerability of the end organ islet cells. Treatment opportunities to correct genetically and epigenetically dysregulated genes will be pursued with antisense approaches.

DESCRIPTION (provided by applicant): Nearly half of the U.S. population will meet criteria for a psychiatric disorder during their lives, and 1 in 17 has a seriously debilitating illness. Increasingly, these psychopathologies are conceptualized as the late-stage cul- mination of aberrant developmental processes shaped by a complex interplay of genes and experience, includ- ing those occurring in utero. Decades of studies with pregnant animals demonstrate that stress-elicited pertur- bations in maternal biology affect offspring development, leading to a profile characterized by heightened be- havioral and physiologic stress responsivity. Studies of distress in pregnant women, which range from exami- nations of life stress to psychiatric disorder, largely mirror these findings. Despite ample evidence linking ante- natal maternal functioning to offspring outcomes, the mechanistic pathways for this in utero influence on chil- dren's neurodevelopment remain unknown, particularly with human subjects. The burgeoning field of epigenet- ics - the detection of the molecular effects of environmental experience - has only minimally been applied to pregnant women, yet may provide a vital link in understanding the mechanisms involved in the fetal origins of psychiatric disease risk. The goal of this project is to use recent advances in studying epigenetic gene regulation to identify the biological mechanisms mediating the impact of maternal distress on perinatal development. Aim 1: Determine the influence of pregnant women's distress on epigenetic gene regula- tion relevant to perinatal development. Specifically, to establish whether (a) prenatal distress (daily life stress assessed 3x in pregnancy using 24-hour Ecological Momentary Assessment (EMA) via a Personal Digi- tal Assistant (PDA)) and mood symptoms elicited by clinician interviews) predict women's stress hormone lev- els (cortisol (from 3x, 12 salivary samples in 48-hours) and CRH (3x blood draws) and gene expression in her PBL (3x blood draws);(b) the timing and degree of women's altered stress hormone levels and PBL gene ex- pression predict placental gene expression;(c) these mood-dependent biological alterations are associated with the epigenetic mechanism of DNA methylation. Aim 2: Determine perinatal consequences of pregnant women's distress. Specifically whether, (a) women's distress-associated altered HPA-axis hormone levels, PBL and placental gene expression/epigenetic variation, predict fetal cord blood gene expression/epigenetic variation, as well as a neurobehavioral profile characterized by heightened reactivity to novelty (fetal and new- born autonomic and central nervous system regulation in response to stimuli). Aim 3: Establish causal influ- ence of epigenetic modification on offspring neurodevelopment. Specifically, using a rodent model in which brain effects of chronic maternal prenatal stress exposure can be directly assessed, we aim to determine (a) the influence of maternal condition on DNA methylation and gene expression in maternal PBLs, placenta, and in the fetal/infant brain and, (b), the relationship between epigenetic variations in these tissues and the de- velopment of the postnatal ANS and CNS as indexed by behavioral and stress-hormone responsivity. PUBLIC HEALTH RELEVANCE: The proposed research is relevant to public health because mechanistic studies of the impact of pregnant women's distress on perinatal neurobehavioral development will greatly increase understanding of the contri- bution of prenatal experience to long-term risk for psychopathology, as well as risk prediction. Thus, the pro- posed research is relevant to NIH's mission that pertains to the development of fundamental knowledge that will reduce the burdens of human illness.

The presence in carcinomas of large numbers of myofibroblasts (MFs) originally suggested the concept of cancer as a non-healing wound, and experiments have now implicated these cells as contributing to cancer growth, invasion and metastasis. Recently our labs (Tycko collaborating with Wang) showed that cancerassociated myofibroblasts (CAFs) in human gastric carcinomas (GCAs) and in a mouse model of GCA have globally reduced DNA methylation and focal gains of promoter methylation, compared to normal MFs in the stomach. We now have substantial unpublished data indicating that these findings of altered DNA methylation in CAFs extend to pancreatic carcinoma (PnCA), and that the demethylating drug decitabine (5aza-dC) has strong anti-tumor activity in a mouse model of PnCA, mediated in part through its effects on the tumor-supporting activity of CAFs and in part via its direct effects on the malignant epithelial cells. To build on these findings we have 4 current objectives - all centered on DNA methylation as a therapeutic target in both the supportive CAFs and malignant epithelial cells of stroma-rich gastrointestinal cancers. First, we will carry out several types of molecular assays to ask whether CAFs accumulate global and genespecific alterations in DNA methylation in pancreatic and hepatocellular carcinomas. Second, we will use mixing/allografting experiments to ask whether the demethylating drug decitabine inhibits the tumorsupporting function of CAFs, and we will profile gene expression and characterize mesenchymal stem cell markers in response to this drug to gain insights to its mechanisms of action against CAFs. Third, we will use in vivo mouse models of PnCA to test whether decitabine, alone and in combination with other agents, can prevent tumor progression and lead to regression of established PnCA tumors. This objective is linked to a Phase 1 clinical trial of decitabine in humans with PnCA, for which we will carry out the laboratory-based pharmacodynamic studies. Our fourth and last objective is to use a genetic strategy in mice to test whether deletion of the maintenance methyltransferase gene Dnmtl, specifically in MFs, can slow or prevent liver fibrosis and cancer in a chemical carcinogenesis model of hepatocellular carcinoma.

This application is being submitted in response to NOT-OD-19-071. In our MPI R01 parent grant ?Epigenetics of Down syndrome?, we are using mouse models to address the mechanisms of altered epigenetic patterning in Down syndrome (DS), and how the epigenetic changes can affect developmental phenotypes. In this supplement/revision application, we propose to study epigenetic and biological aging and the roles of these processes in co-occurring age-related conditions, using our segmental duplication mouse models of DS, namely Dp(10)1Yey & Dp(16)1Yey, which contain segmental duplications of the regions of mouse chromosomes 10 and 16 (Mm10, Mm16) that show conserved synteny with human chromosome 21 (Hsa21). The medically significant phenotypes to be investigated are age-related immune system alterations, age-dependent hearing loss, and age-dependent cognitive decline (independently of Alzheimer?s disease, AD). We believe that using the Dp(10) and Dp(16) mouse models, and Dp(16) mice with App gene dosage normalized to disomy, we can start to use genetic dissection to help us to answer the following questions: (i) what are the genes on Hsa21 that cause early epigenetic aging in DS? (ii) what are the genes on Hsa21 that cause early biological aging in DS (iii) is epigenetic aging simply a useful ?clock?, or do epigenetic changes functionally influence biological aging? and (iv) does early biological aging contribute to the early onset of immune system deficits, hearing loss, and cognitive decline (independently of AD) in DS? We have two specific aims: (1) By methyl-seq & RNA-seq of brain and immune system cells of the above models at defined ages, we will ask whether early epigenetic aging occurs in these segmental models, and whether this phenomenon requires duplication of both chromosomal regions. (2) By analyzing the aforementioned clinical phenotypes in the same mouse models at each of the same ages as in Aim 1, we will determine whether the duplications of Hsa21 syntenic regions on Mmu10 and Mmu16 affect these age-dependent phenotypes. Like DS in humans, App gene triplication is necessary for DS mice to exhibit Alzheimer-type neurodegeneration. By normalizing App gene dosage to two copies in the segmental duplication mice, we will determine the impact of early aging on cognitive decline, independently of AD pathology, by using synaptic density analysis, hippocampal long-term potentiation and behavioral paradigms of learning and memory at successive ages. Attaining these objectives will set the stage for identifying the Hsa21 gene ortholog(s) that cause early epigenetic and biological aging in DS and thereby affect the age-of-onset of immune system alterations, hearing loss, and AD-independent cognitive decline. These results will be relevant both for understanding and ameliorating co-occurring conditions in people with DS and for ameliorating these same problems in the general (euploid) population.

Project Summary/Abstract Well powered genome-wide association studies (GWAS) for celiac disease and Crohn's disease have identified numerous non-HLA susceptibility loci. Genetic fine-mapping and expression quantitative trait loci (eQTL) can narrow in on which genes probably account for the GWAS signals, but even after these approaches the locations of the causal nucleotide changes often remain unknown. This gap in understanding is a roadblock to targeted prevention and therapy. As developed in our prior work, mapping haplotype-dependent allele-specific CpG methylation (hap-ASM) and methylation quantitative trait loci (mQTLs) in cells relevant to a given disease, and then overlapping these epigenetic maps with GWAS data, can help to hone in on the DNA regulatory sequences that causally underlie the GWAS signals. Our hypothesis is that this combined genetic-epigenetic mapping strategy, followed by functional assays, will be able to identify regulatory DNA sequences that contribute to celiac disease and Crohn's disease susceptibility and pathogenesis. First, we will carry out Methyl-Seq of CD4+ and CD8+ T cells and peripheral blood monocytes, to map hap-ASM genome-wide. These data will pinpoint hap- ASM differentially methylated regions (DMRs) in the same haplotype blocks as GWAS peaks for celiac and Crohn's. Since hap-ASM often reflects allele-specific transcription factor binding site (TFBS) or insulator occupancies, we will cross-validate our findings using an independent method, Assay for Transposase- Accessible Chromatin Sequencing (ATAC-Seq). In our second aim, we will rank and prioritize the hap-ASM DMRs based on the strength of the allelic asymmetries, ATAC-Seq overlaps, and linkage disequilibrium (LD) with GWAS peak SNPs, and perform high-throughput targeted bisulfite sequencing to fine-map the top-ranked DMRs, both in the blood-derived cells and in mucosal T cells from celiac and Crohn's patients. In our third aim, to definitively test the functional roles of specific TFBS in the DMRs, we will use CRISPR to delete these sequences in Jurkat cells and in normal T cells. We will score effects of the deletions on local methylation patterns and mRNA levels of each of the nearby genes. These data will clarify our fundamental understanding of susceptibility and pathogenesis of celiac disease and Crohn's disease.

ABSTRACT DNA repair is a crucial mechanism for maintaining genomic stability in cells. Defects in the DNA repair machinery increase cell vulnerability to DNA-damaging agents and accumulation of mutations in the genome, and lead to the development of various disorders including cancers. Studies that have measured DNA repair capacity (DRC), including our own, have estimated a much higher risk of breast cancer (BC) (3-15-fold) than most other established risk factors for BC, with the exception of highly penetrant mutations in genes like BRCA1 and BRCA2, genes critical to DNA repair. Despite the strength of this association, no large-scale prospective studies of BC exist. Even though some BC risk models include known mutations in DNA repair genes, genotype only partially explains phenotype, and BC risk models currently do not include phenotypic DNA repair measures. The lack of inclusion of a major risk factor ? DRC ? is likely the major reason that clinical BC risk models have only modest performance - which makes it very challenging to target effective primary prevention options (e.g., chemoprevention) for the majority of women who are not known mutation carriers. Further, secondary prevention options (e.g., onset, frequency, and method of BC screening by mammography or other supplemental methods) could be targeted more efficiently if more accurate risk assessment existed. The main limitation of use of DRC for targeted prevention has been the lack of a high-throughput DRC assay, in particular a phenotypic DRC assay, for integration into cancer risk assessment. We have overcome this major gap by adapting our high-throughput, fully-automated ?-H2AX assay system which was originally designed for assaying DNA double strand breaks (DSB) in freshly-drawn blood for use with archival blood samples. We propose one of the largest prospective studies estimating the effect of DSB repair using an enriched cohort (n=12,563) that spans the spectrum of absolute BC risk. Using a nested case-control design within this cohort (699 cases, 1:1 match), we will measure DSB-DRC in archival biospecimens collected at baseline (Aim 1a). We will optimize the assay protocol for measuring DSB-DRC using fresh fingerstick blood and measure longitudinal changes in DSB-DRC in young women (age <40 years) (Aim 1b) (n=100, 1-2 years apart). We will then comprehensively assess the independent contribution of DSB-DRC over genetic and epigenetic alterations in DSB repair genes, and assess and whether genetic and epigenetic changes interact with DSB-DRC in increasing BC risk (Aim 2). We will investigate the clinical utility of DSB-DRC by quantifying the improvement in standard BC risk model performance from its inclusion (Aim 3a), and evaluating the association between DSB-DRC and 5 year survival after BC diagnosis (Aim 3b). Our study will provide essential empirical evidence from integrating functional assays into population studies to accelerate targeted prevention options linked to aberrant responses to DNA damage. This research will be led by a team of established investigators in the fields of BC epidemiology, molecular epidemiology, high-throughput DNA repair capacity assessment, and biostatistics.