Dysregulation of Synaptic Signaling Genes Is Involved in Biology of Uterine Leiomyoma

Uterine leiomyomas are tumors, which are hormone driven and originate from the smooth muscle layer of the uterine wall. In addition to known genes in leiomyoma pathogenesis, recent approaches also highlight epigenetic malfunctions as an important mechanism of gene dysregulation. RNA sequencing raw data from pair-matched normal myometrium and fibroid tumors from two independent studies were used as discovery and validation sets and reanalyzed. RNA extracted from normal myometrium and fibroid tumors from 58 Slovenian patients was used as independent confirmation of most significant differentially expressed genes. Subsequently, GWA data from leiomyoma patients were used in order to identify genetic variants at epigenetic marks. Gene Ontology analysis of the overlap of two independent RNA-seq analyses showed that NPTX1, NPTX2, CHRM2, DRD2 and CACNA1A were listed as significant for several enriched GO terms. All five genes were subsequently confirmed in the independent Slovenian cohort. Additional integration and functional analysis showed that genetic variants in these five gene regions are listed at a chromatin structure and state, predicting promoters, enhancers, DNase hypersensitivity and altered transcription factor binding sites. We identified a unique subgroup of dysregulated synaptic signaling genes involved in the biology and pathogenesis of leiomyomas, adding to the complexity of tumor biology.

Dissecting the regulatory activity and sequence content of loci with exceptional numbers of transcription factor associations

DNA associated proteins (DAPs) classically regulate gene expression by binding to regulatory loci such as enhancers or promoters. As expanding catalogs of genome-wide DAP binding maps reveal thousands of loci that, unlike the majority of conventional enhancers and promoters, associate with dozens of different DAPs with apparently little regard for motif preference, an understanding of DAP association and coordination at such regulatory loci is essential to deciphering how these regions contribute to normal development and disease. In this study, we aggregated publicly available ChIP-seq data from 469 human DAPs assayed in three cell lines and integrated these data with an orthogonal dataset of 352 non-redundant, in vitro-derived motifs mapped to the genome within DNase hypersensitivity footprints in an effort to characterize regions of the genome that have exceptionally high numbers of DAP associations. We subsequently performed a massively parallel mutagenesis assay to search for sequence elements driving transcriptional activity at such loci and explored plausible biological mechanisms underlying their formation. We establish a generalizable definition for High Occupancy Target (HOT) loci and identify putative driver DAP motifs in HEPG2 cells, including HNF4A, SP1, SP5, and ETV4, that are highly prevalent and exhibit sequence conservation at HOT loci. The number of different DAPs associated with an element is positively associated with evidence of regulatory activity and, by systematically mutating 245 HOT loci, we localized regulatory activity to a central core region that depends on the motif sequences of our previously nominated driver DAPs. In sum, this work leverages the increasingly large number of DAP motif and ChIP-seq data publicly available to explore how DAP associations contribute to genome-wide transcriptional regulation.

Associations between potentially functional CORIN SNPs and serum corin levels in the Chinese Han population

Background Corin is an important convertase involved in the natriuretic peptide system and may indirectly regulate blood pressure. Genetic factors relate to corin remain unclear. The purpose of the current study was to comprehensively examine the associations among CORIN SNPs, methylations, serum corin levels and hypertension. Results We genotyped 9 tag-SNPs in the CORIN gene and measured serum corin levels in 731 new-onset hypertensive cases and 731 age- and sex-matched controls. DNA methylations were tested in 43 individuals. Mendelian randomization was used to investigate the causal associations. Under additive models, we observed associations of rs2289433 (p.Cys13Tyr), rs6823184, rs10517195, rs2271037 and rs12509275 with serum corin levels after adjustment for covariates (P = 0.0399, 0.0238, 0.0016, 0.0148 and 0.0038, respectively). The tag-SNP rs6823184 and SNPs that are in strong linkage disequilibrium with it, i.e., rs10049713, rs6823698 and rs1866689, were associated with CORIN gene expression (P = 2.38 × 10− 24, 5.94 × 10− 27, 6.31 × 10− 27 and 6.30 × 10− 27, respectively). Neither SNPs nor corin levels was found to be associated with hypertension. SNP rs6823184, which is located in a DNase hypersensitivity cluster, a CpG island and transcription factor binding sites, was significantly associated with cg02955940 methylation levels (P = 1.54 × 10− 7). A putative causal association between cg02955940 methylation and corin levels was detected (P = 0.0011). Conclusion This study identified potentially functional CORIN SNPs that were associated with serum corin level in the Chinese Han population. The effect of CORIN SNPs on corin level may be mediated by DNA methylation.

Poly(ADP-ribose)-dependent chromatin unfolding facilitates the association of DNA-binding proteins with DNA at sites of damage

The addition of poly(ADP-ribose) (PAR) chains along the chromatin fiber due to PARP1 activity regulates the recruitment of multiple factors to sites of DNA damage. In this manuscript, we investigated how, besides direct binding to PAR, early chromatin unfolding events controlled by PAR signaling contribute to recruitment to DNA lesions. We observed that different DNA-binding, but not histone-binding, domains accumulate at damaged chromatin in a PAR-dependent manner, and that this recruitment correlates with their affinity for DNA. Our findings indicate that this recruitment is promoted by early PAR-dependent chromatin remodeling rather than direct interaction with PAR. Moreover, recruitment is not the consequence of reduced molecular crowding at unfolded damaged chromatin but instead originates from facilitated binding to more exposed DNA. These findings are further substantiated by the observation that PAR-dependent chromatin remodeling at DNA lesions underlies increased DNAse hypersensitivity. Finally, the relevance of this new mode of PAR-dependent recruitment to DNA lesions is demonstrated by the observation that reducing the affinity for DNA of both CHD4 and HP1α, two proteins shown to be involved in the DNA-damage response, strongly impairs their recruitment to DNA lesions.

Bioinformatics Discovery of Putative Enhancers within Mouse Odorant Receptor Gene Clusters

Olfactory neuronal function depends on the expression and proper regulation of odorant receptor (OR) genes. Previous studies have identified 54 putative intergenic enhancers within or flanking 40 mouse OR clusters. At least 2 of these putative enhancers have been shown to regulate the expression of a small subset of proximal OR genes. In recognition of the large size of the mouse OR gene family (~1400 OR genes distributed across multiple chromosomal loci), it is likely that there remain many additional not-as-yet discovered OR enhancers. We utilized 23 of the previously identified enhancers as a training set (TS) and designed an algorithm that combines a broad range of epigenetic criteria (histone-3-lysine-4 monomethylation, histone-3-lysine-79 trimethylation, histone-3-lysine-27 acetylation, and DNase hypersensitivity) and genetic criteria (cross-species sequence conservation and transcription-factor binding site enrichment) to more broadly search OR gene clusters for additional candidates. We identified 181 new candidate enhancers located at 58 (of 68) mouse OR loci, including 25 new candidates identified by stringent search criteria whose signal strengths are not significantly different from the 23 previously characterized OR enhancers used as the TS. Additionally, we compared OR enhancer versus generic enhancer features in order to evaluate likelihoods that new enhancer candidates specifically function in OR regulation. We found that features distinguishing OR-specific function are significantly more evident for enhancer candidates located within OR clusters as compared with those in flanking regions.

GRID-seq assisted prediction of transcription factor binding motifs

Experimental validation of computationally predicted transcription factor binding motifs is desirable. Increased RNA levels in the vicinity of predicted protein-chromosomal binding motifs intuitively suggest regulatory activity. With this intuition in mind, the approach presented here juxtaposes publicly available experimentally derived GRID-seq data with binding motif predictions computationally determined by deep learning models. The aim is to demonstrate the feasibility of using RNA-sequencing data to improve binding motif prediction accuracy. Publicly available GRID-seq scores and computed DeepBind scores could be aggregated by chromosomal region and anomalies within the aggregated data could be detected using mahalanobis distance analysis. A mantel’s test of matrices containing pairwise hamming distances showed significant differences between 1) randomly ranked sequences, 2) sequences ranked by non-GRID-seq assisted scores, and 3) sequences ranked by GRID-seq assisted scores. Plots of mahalanobis ranked binding motifs revealed an inversely proportional relationship between GRID-seq scores and DeepBind scores. Data points with high DeepBind scores but low GRID-seq scores had no DNAse hypersensitivity clusters annotated to their respective sequences. However, DNase hypersensitivity was observed for high scoring DeepBind motifs with moderate GRID-seq scores. Binding motifs of interest were recognized by their deviance from the inversely proportional tendency, and the underlying context sequences of these predicted motifs were on occasion associated with DNAse hypersensitivity unlike the most highly ranked motif scores when DeepBind was used in isolation. This article presents a novel combinatory approach to predict functional protein-chromosomal binding motifs. The two underlying methods are based on recent developments in the fields of RNA sequencing and deep learning, respectively. They are shown to be suited for synergistic use, which broadens the scope of their respective applications.

Dilated Convolutions for Modeling Long-Distance Genomic Dependencies

We consider the task of detecting regulatory elements in the human genome directly from raw DNA. Past work has focused on small snippets of DNA, making it difficult to model long-distance dependencies that arise from DNA’s 3-dimensional conformation. In order to study long-distance dependencies, we develop and release a novel dataset for a larger-context modeling task. Using this new data set we model long-distance interactions using dilated convolutional neural networks, and compare them to standard convolutions and recurrent neural networks. We show that dilated convolutions are effective at modeling the locations of regulatory markers in the human genome, such as transcription factor binding sites, histone modifications, and DNAse hypersensitivity sites.