Cap analysis of gene expression reveals alternative promoter usage in a rat model of hypertension

The role of alternative promoter usage in tissue-specific gene expression has been well established; however, its role in complex diseases is poorly understood. We performed cap analysis of gene expression (CAGE) sequencing from the left ventricle of a rat model of hypertension, the spontaneously hypertensive rat (SHR), and a normotensive strain, Brown Norway to understand the role of alternative promoter usage in complex disease. We identified 26,560 CAGE-defined transcription start sites in the rat left ventricle, including 1,970 novel cardiac transcription start sites. We identified 28 genes with alternative promoter usage between SHR and Brown Norway, which could lead to protein isoforms differing at the amino terminus between two strains and 475 promoter switching events altering the length of the 5′ UTR. We found that the shift in Insr promoter usage was significantly associated with insulin levels and blood pressure within a panel of HXB/BXH recombinant inbred rat strains, suggesting that hyperinsulinemia due to insulin resistance might lead to hypertension in SHR. Our study provides a preliminary evidence of alternative promoter usage in complex diseases.

Transcriptional Regulation of the Human IL5RA Gene through Alternative Promoter Usage during Eosinophil Development

Regulation of the IL-5 receptor alpha (IL5RA) gene is complicated, with two known promoters (P1 and P2) driving transcription, and two known isoforms (transmembrane and soluble) dichotomously affecting the signaling potential of the protein products. Here, we sought to determine the patterns of P1 and P2 promoter usage and transcription factor occupancy during primary human eosinophil development from CD34+ hematopoietic stem cell progenitors. We found that during eosinophilopoiesis, both promoters were active but subject to distinct temporal regulation, coincident with combinatorial interactions of transcription factors, including GATA-1, PU.1, and C/EBP family members. P1 displayed a relatively constant level of activity throughout eosinophil development, while P2 activity peaked early and waned thereafter. The soluble IL-5Rα mRNA peaked early and showed the greatest magnitude fold-induction, while the signaling-competent transmembrane isoform peaked moderately. Two human eosinophilic cell lines whose relative use of P1 and P2 were similar to eosinophils differentiated in culture were used to functionally test putative transcription factor binding sites. Transcription factor occupancy was then validated in primary cultures by ChIP. We conclude that IL-5-dependent generation of eosinophils from CD34+ precursors involves complex and dynamic activity including both promoters, several interacting transcription factors, and both signaling and antagonistic protein products.

Cap analysis of gene expression (CAGE) sequencing reveals alternative promoter usage in complex disease

The role of alternative promoter usage in tissue specific gene expression has been well established, however, its role in complex diseases is poorly understood. We performed cap analysis of gene expression (CAGE) tag sequencing from the left ventricle (LV) of a rat model of hypertension, the spontaneously hypertensive rat (SHR), and a normotensive strain, the Brown Norway (BN) to understand role of alternative promoter usage in complex disease. We identified 26,560 CAGE-defined transcription start sites (TSS) in the rat LV, including 1,970 novel cardiac TSS resulting in new transcripts. We identified 27 genes with alternative promoter usage between SHR and BN which could lead to protein isoforms differing at the amino terminus between two strains. Additionally, we identified 475 promoter switching events where a shift in TSS usage was within 100bp between SHR and BN, altering length of the 5 prime UTR. Genomic variants located in the shifting promoter regions showed significant allelic imbalance in F1 crosses, confirming promoter shift. We found that the insulin receptor gene (Insr) showed a switch in promoter usage between SHR and BN in heart and liver. The Insr promoter shift was significantly associated with insulin levels and blood pressure within a panel of BXH/HXB recombinant inbred (RI) rat strains. This suggests that the hyperinsulinemia due to insulin resistance might lead to hypertension in SHR. Our study provides a preliminary evidence of alternative promoter usage in complex diseases.

Profiling of transcribed cis-regulatory elements in single cells

Profiling of cis-regulatory elements (CREs, mostly promoters and enhancers) in single cells allows the interrogation of the cell-type and -state specific contexts of gene regulation and genetic predisposition to diseases. Here we demonstrate single-cell RNA-5′end-sequencing (sc-end5-seq) methods can detect transcribed CREs (tCREs), enabling simultaneous quantification of gene expression and enhancer activities in a single assay with no extra cost. We show enhancer RNAs can be effectively detected using sc-end5-seq methods with either random or oligo(dT) priming. To analyze tCREs in single cells, we developed SCAFE (Single Cell Analysis of Five-prime Ends) to identify genuine tCREs and analyze their activities (https://github.com/chung-lab/scafe). As compared to accessible CRE (aCRE, based on chromatin accessibility), tCREs are more accurate in predicting CRE interactions by co-activity, more sensitive in detecting shifts in alternative promoter usage and more enriched in diseases heritability. Our results highlight additional dimensions within sc-end5-seq data which can be used for interrogating gene regulation and disease heritability.

Difference in potential DNA methylation impact on gene expression between fast- and slow-type myofibers

Skeletal muscles are comprised of two major types of myofibers, fast and slow. It is hypothesized that once myofiber type is determined, muscle fiber-type specificity is maintained by an epigenetic mechanism, however, this remains poorly understood. To address this, we conducted a comprehensive CpG methylation analysis with a reduced representation of bisulfite sequencing (RRBS). Using GFP-myh7 mouse, we visually distinguished and separately pooled slow-type and myh7-negative fast-type fibers for analyses. A total of 31,967 and 26,274 CpGs were hypermethylated by ≥10% difference in the fast- and slow-type fibers, respectively. Notably, the number of promoter-hypermethylated genes with down-regulated expression in the slow-type fibers was 3.5 times higher than that in the fast-type fibers. Gene bodies of the fast-type-specific myofibrillar genes Actn3, Tnnt3, Tnni2, Tnnc2, and Tpm1 were hypermethylated in the slow-type fibers, whereas those of the slow-type-specific genes Myh7, Tnnt1, and Tpm3 were hypermethylated in the fast-type fibers. Each of the instances of gene hypermethylation was associated with the respective down-regulated expression. In particular, a relationship between CpG methylation sites and the transcription variant distribution of Tpm1 was observed, suggesting a regulation of Tpm1 alternative promoter usage by gene body CpG methylation. An association of hypermethylation with the regulation of gene expression was also observed in Wdr70, and transcription factors Sim2 and Tbx1. These results suggest not only a myofiber type-specific regulation of gene expression and alternative promoter usage by gene body CpG methylation, but also a dominant effect of promoter-hypermethylation on the gene expressions in slow myofibers.