The analysis of individual parts of the genome enables a more comprehensive understanding of how the components work together in the broader context of disease. The following projects are described with respect to continuation of previous years' projects and represent integrative analyses of independent genomic data types to address the genome as a complex regulatory system. The analysis areas include cis-acting, trans-acting and epigenetic regulators of the human genome. Comparing genome-wide methylation patterns in subtypes of ovarian tumors and mouse models: Altered DNA methylation in promoter regions can distinguish genes that are relevant to ovarian tumor pathology (Kolbe et al, PLoS ONE 2012). Given the somatic nature of 90% of ovarian cancers, disruption of normal gene regulation is a likely contributor to disease etiology. Methylation patterns at 25,475 unique loci in 43 samples of ovarian, endometrial or metastatic tumors, along with normal fallopian tube and normal endometrium showed that methylation mirrors histopathological subdivisions and discriminates tumor types with finer granularity and greater reproducibility than published gene expression assays. The extensive differences we showed between tumor and normal samples are the first report of a methylator phenotype in ovarian endometrioid tumors, analogous to the methylator phenotype identified in colorectal cancer and glioblastoma. We have expanded these studies to show that methylator phenotypes can be identified in a majority of tumor types (Sanchez-Vega et al. 2015). Furthermore, we developed a method to classify tumor DNA as methylator phenotype, or not, which works on 11 cancer types from TCGA, as well as numerous distinct cancer types represented by cancer cell lines. We have shown that commonalities across CIMP samples from different tissue implicate a common underlying etiology (Miller at el, 2016), however, the causal event remains elusive. Nevertheless, in a detailed study of tumors from the gastrointestinal tract we showed similarities and differences among CIMP tumors from esophageal, gastric, and colorectal adenocarcinomas (Snchez-Vega et al. 2017), which could shed light on shared targets for cancer treatments. Related projects include methods development for assessing significant sites of epigenetic modifications, including DNA methylation, in myriad ChIP-seq samples (Lichtenberg 2017). Profiling common epigenetic features in solid human epithelial tumors: The study of aberrant DNA methylation in cancer holds the key to the discovery of novel biological markers for diagnostics and can help to delineate important mechanisms of disease. We have identified 12 loci that are differentially methylated in serous ovarian cancers and endometrioid ovarian and endometrial cancers with respect to normal controls. The strongest signal showed hypermethylation in tumors at a CpG island within the ZNF154 promoter (Sanchez-Vega et al. 2013). We show that hypermethylation of this locus is recurrent across solid human epithelial tumor samples for 15 of 16 distinct cancer types from TCGA. Furthermore, ZNF154 hypermethylation is strikingly present across a diverse panel of ENCODE cell lines, and unique to cell lines derived from tumor cells. By extending our analysis from the Illumina 27K Infinium platform to the 450K platform, followed by PCR amplification of target region sin bisulfite treated DNA, we demonstrate that hypermethylation extends across the breadth of the ZNF154 CpG island. We have also identified recurrent hypomethylation in two genomic regions associated with CASP8 and VHL. These three genes exhibit significant negative correlation between methylation and gene expression across many cancer types, as well as patterns of DNaseI hypersensitivity and histone marks that reflect different chromatin accessibility in cancer vs. normal cell lines. Our findings emphasize hypermethylation of ZNF154 as a biological marker of relevance for tumor diagnostics. Epigenetic modifications affecting the promoters of ZNF154, CASP8 and VHL are shared across a vast array of tumor types and may therefore be important for understanding the genomic landscape of cancer. We recently began testing methylation of ZNF154 as a marker for development of diagnostic tests of myriad epithelial cancers. These tests use DNA methylation at the ZNF154 locus in DNA that is collected from plasma of patients. We implemented the test using a semi-quantitative PCR assay that uses melting curve temperatures to assess DNA methylation in the amplified products. The test is highly sensitive and specific for circulating tumor DNA. The findings of our first report on ovarian cancer are being drafted. Developing a comprehensive association map for the cancer methylome and driver mutations: Many cancers are currently defined by characteristics that correlate with the presence of a particular driver mutation. Having studied the reproducible patterns of DNA methylation in myriad cancer types, we hypothesized that the epigenetic landscape of DNA methylation would correlate with the presence of specific driver mutations. We measured this relationship using principal component analyses and methylation-mutation associations applied at the nucleotide level and with respect to genome-wide trends. We found that a few mutated driver genes were associated with genome-wide patterns of aberrant hypomethylation or CpG island hypermethylation in specific cancer types. In addition, we identified associations between 737 mutated driver genes and site-specific methylation changes. Moreover, using these mutation-methylation associations, we were able to distinguish between two uterine and two thyroid cancer subtypes. The driver gene mutation-associated methylation differences between the thyroid cancer subtypes were linked to differential gene expression in JAK-STAT signaling, NADPH oxidation, and other cancer-related pathways. These results establish that driver gene mutations are associated with methylation alterations capable of shaping regulatory network functions. In addition, the methodology presented here can be used to subdivide tumors into more homogeneous subsets corresponding to underlying molecular characteristics, which could improve treatment efficacy (Chen et al. 2017). Functional Elements: It is readily apparent through genome sequencing studies that ultraconserved regions of the genome are under strong purifying selection, although the necessity of such adaptive pressure remains unclear. I recently reviewed the importance of ultraconserved elements in the human genome with respect to their role in neurodevelopmental disorders. Together, the data support the idea that loss of UCEs and other strongly evolutionarily conserved elements can cause defects that may have profound consequences for reproductive success in the wild, but are nevertheless subtle in laboratory mice, due to redundancy of enhancer functions (Elnitski and Ovcharenko 2018). Tying into this concept, we recently reported (Chen et al. 2018) a study of 14 pairs of monozygotic twins disparate for attention deficit hyperactivity disorder. Despite a lack of causal gene mutations, affected twins had a significantly smaller right striatum and thalamus, and a trend toward a larger cerebellum, but did not differ in cerebral cortical volume. Despite results showing no germline mutations, indels or deletions that would explain the disparate disease presence, the affected twins showed significant differences in DNA methylation patterns associated with some enhancer regions of genes expressed in the altered brain regions. Taken together, these reports offer the possibility that subtle effects, such as loss or alteration of enhancer elements in the genome may be associated with discrete neuroanatomical anomalies.
Showing the most recent 10 out of 35 publications
