Genomewide analysis of enhancerRNA transcription reveals regulatory mechanisms

Genome-wide analysis of enhancer-RNA transcription reveals regulatory mechanisms by an anti-diabetic drugs (TZD) in adipocytes Kyoung-Jae Won, Ph. D. Sep 8 2014 Genomics 2014

Non-coding RNA Annotation in ENSEMBL t. RNA Mt-t. RNA r. RNA sc. RNA sno. RNA mi. RNA Misc_RNA linc. RNA e. RNA (enhancer driven RNA)

e. RNA: Transcripts at enhancers Discussions Kim et al (2010) Nature

Enhancer-driven RNAs (e. RNAs) are an important factor for gene regulation Function of e. RNA : noise ? : activating role ? : scaffold ? At least e. RNAs show enhancer activity Lam et al (2014) Trends in Biochemical Sci.

Transcription: An Understudied Step in Gene Regulation Transcription Steady-state m. RNAs GRO-seq • Global Run On – sequencing • Isolate all newly synthesized RNAs in a cell • Measures active transcription rather than steady state m. RNA levels • Maps the position and orientation of new transcripts • Earliest changes can identity primary target genes rather than secondary targets • Unbiased detection of novel transcripts, including non-coding RNAs • detects enhancer RNAs Core et al, Science, 2008

Thiazolidindiones (TZDs) Activates PPARs (peroxisome proliferator-activated receptor) Effective anti-diabetic drugs: increase insulin sensitivity Side effect: weight gain, cardiac loads, peripheral edema Still widely used for the treatment of diabetes despite its notorious side-effects

Up Regulation of Transcription Due to Rosi CEBPa PPARg Pre. Adipocyte Before Rosi + 10 m Rosi + 30 m Rosi + 1 h Rosi + 3 h Fabp 4 + strand

Up and Down Regulation of Transcription Due to Rosi Log 2 -Fold. Change in Gene Body Cited 2: ↓ Ppargc 1 b: ↑ 1384 Adipoq: ↑ 567 Cav 1: ↓

GROseq Profiles at TSS in Adipocyte

Nascent Transcription Changes Precedes m. RNA Regulation Log 2 -Fold. Change in Gene Body Nascent Gene Transcript Steady-state m. RNA GROseq vs Steady-state m. RNA Log 2 Fold Change Over Rosi + 0 30 m 10 m 30 m 1 h 3 h GROseq

Nascent Transcription Changes Precedes m. RNA Regulation Log 2 -Fold. Change in Gene Body Nascent Gene Transcript 48 h 36 h Steady-state m. RNA 24 h GROseq vs Steady-state m. RNA Log 2 Fold Change Over Rosi + 0 12 h 6 h 2 h 1 h 30 m 10 m 30 m 1 h 3 h GROseq

Nascent Transcription Change Precedes m. RNA Regulation Log 2 -Fold. Change in Gene Body 48 h 36 h Steady-state m. RNA 24 h 12 h 6 h 2 h 1 h 1 h 30 m 10 m 30 m 1 h 3 h Gene Body Transcript

Enhancer RNAS (e. RNAs) Log 2 -Fold. Change in Gene Body CEBPa PPARg Before Rosi + 10 m Rosi + 30 m Rosi + 1 h Rosi + 3 h e. RNA + strand - strand

Rosi-regulated e. RNAs

log 2 FC(Gene Transcript) Regulation of e. RNAs and Nascent Transcription by Rosi log 2 FC(e. RNA)

e. RNA Changes Precede Nascent Gene Transcript Enhancer RNA Nascent Gene Transcript Steady-state m. RNA Correlation between e. RNA and gene regulation 2 3 h − 2 Gene transcript 0 2 0 − 2 2 1 h 30 m 10 m 0 − 2 − 3 0 3 10 m 30 m 1 h 3 h e. RNA 10 m 30 m 1 h e. RNA 3 h

Relative expression e. RNA induction preceeds gene induction Pdk 4 4 3 2 Gene e. RNA 1 0 0 10 20 30 40 50 Rosi treatment (minutes) 60

e. RNAs up-regulation is associated with PPARγ

e. RNAs up-regulation is associated with PPARγ

e. RNAs down-regulation is associated with multiple-factors including C/EBP, Fosl 2, Jun. D and ATF 2 : Repression is not directly mediated by PPARγ

Strength of binding did not change upon rosi treatment Fosl 2 Ch. IP 1 ATF-2 Ch. IP 1. 5 0. 9 Percent Input 0. 8 0. 7 0. 6 0. 5 0. 4 1 0. 5 0. 3 0. 2 0. 1 0 0 1 1 2 3 5 Arbp Ins Jun. D Ch. IP 1. 5 Percent Input 4 1 Ad Ad + 1 h rosi 0. 5 0 1 2 3 4 5 Arbp Ins

The majority of e. RNA are upregulated by 10 minutes : downregulated e. RNAs can be secondary

e. RNA changes are associated with Med 1 change

Upon rosi treatment Med 1 recruitment significantly decreased at downregulated e. RNAs and increased at upregulated e. RNAs relative to unregulated e. RNAs log 2(tag counts[rosi/untreated]) Med 1 Ch. IP-seq ** 3 ** 2 1 0 -1 -2 -3 Down Unchanged e. RNA regulation by rosi Up

Squelching is a biological phenomenon in which a transcriptional activator acts to inhibit the expression of another gene.

Discussions GROseq measures the level of e. RNAs and gene transcript Integrative studies using GROseq revealed the rosi-associated regulatory mechanisms • Gene transcription levels needs time until they reach to steady state levels • e. RNA fires before transcriptional initiation • up-regulation is mediated by PPARγ • Multiple TFs are associated with down-regulation • down-regulation is mediated by co-activator redistribution Step et al. (2014) Genes and Development

Thanks to Nha Nguyen Hee-Woong Lim Shuo Li Inchan Choi UPENN Mitchell Lazar Sonia Stem Patrick Seale David Steger Klaus Kaestner Rinho Kim Supported by • NIH/NIDDK, • Institute for Diabetes Obesity and Metabolism at the University of Pennsylvania • Pharma Foundation

In the lab. . Metabolic Biology Mitchell Lazar, Klaus Keastner Patrick Seale, David Steger Stem Cell Kennett Zaret, Montserrat Aguera Chromatin remodelers, histone variants Benjamin Garcia, Hua-Ying Epigenetic landscapes Wei Wang, Bing Ren (UCSD) Cancer biology Cathryn Wellen, Mariuz Wasik Integrative studies RNAseq, GROseq Ch. IPseq, Ch. IP-exo CLIP-seq m. C, 5 hm. C, Histone modification