Study / S3JPWSTR92019-02-01
CG dinucleotides enhance promoter activity independent of DNA methylation
Dominik Hartl, Arnaud R. Krebs, Ralph S. Grand, Tuncay Baubec, Luke Isbel et al.
About this study
Most mammalian RNA polymerase II initiation events occur at CpG islands, which are rich in CpGs and devoid of DNA methylation. Despite their relevance for gene regulation, it is unknown to what extent the CpG dinucleotide itself actually contributes to promoter activity. To address this question, we determined the transcriptional activity of a large number of chromosomally integrated promoter constructs and monitored binding of transcription factors assumed to play a role in CpG island activity. This revealed that CpG density significantly improves motif-based prediction of transcription factor binding. Our experiments also show that high CpG density alone is insufficient for transcriptional activity, yet results in increased transcriptional output when combined with particular transcription factor motifs. However, this CpG contribution to promoter activity is independent of DNA methyltransferase activity. Together, this refines our understanding of mammalian promoter regulation as it shows that high CpG density within CpG islands directly contributes to an environment permissive for full transcriptional activity.
Full author list & citation
Dominik Hartl, Arnaud R. Krebs, Ralph S. Grand, Tuncay Baubec, Luke Isbel, Christiane Wirbelauer, Lukas Burger, Dirk Schübeler. CG dinucleotides enhance promoter activity independent of DNA methylation. 2019-02-01. https://doi.org/10.1101/gr.241653.118
Experiments 8
E1EEZQ2GR
TrAC-seq measured transcriptional activity for 10 pooled promoter constructs comprising mouse housekeeping-gene promoters and a CMV control in wild-type TC-1 mouse embryonic stem cells. Constructs were integrated into the same beta-globin RMCE locus and quantified by RNA/DNA barcode enrichment.
E1NQ02HD6
TrAC-seq tested all combinations of CpG-mutated windows across the Snx3 promoter in wild-type TC-1 cells and matched Dnmt triple-knockout derivatives. The five-bit construct identifiers encode the promoter windows selected for CpG mutation, and the table retains source qPCR and methylation measurements where available.
E4I0MLLH4
TrAC-seq measured activity for a library of mouse genomic promoter windows selected for Sp1, Sp3, GABPA, and NRF1 motif/binding analysis in wild-type TC-1 embryonic stem cells. The table retains the mm9 coordinates and CpG/GC features supplied with the GEO results.
E785JIGIW
TrAC-seq assayed a Pwp2 promoter library containing 10-bp tiling-window mutants and transcription-factor motif controls in wild-type TC-1 mouse embryonic stem cells. The table provides replicate activity, WT-normalized activity, and DNA-supported barcode counts for each tiled construct.
E7Y1LP5OF
TrAC-seq compared Pwp2 promoter constructs with CpGs removed from low-effect regions and an artificial promoter with CpGs reintroduced, in wild-type TC-1 cells and matched Dnmt triple-knockout derivatives. The table contains sequence-composition features and WT-normalized activity for both genotypes.
E87X95BTM
TrAC-seq compared CpG-depleted mutant versions of mouse housekeeping promoters in wild-type TC-1 cells and matched Dnmt1/Dnmt3a/Dnmt3b triple-knockout derivatives. Each construct is annotated with its promoter coordinates, CpG mutation percentage, sequence composition, and WT/TKO activity.
E9IWQ93G4
TrAC-seq measured a combinatorial Pwp2 promoter mutant library in wild-type TC-1 cells and matched Dnmt triple-knockout derivatives. The library includes CpG-window combinations and targeted GABPA, NRF1, SP1/SP3, and MYC motif mutations, with source-provided WT-normalized activity.
E9U78YZ9A
TrAC-seq measured wild-type activity for E. coli genomic fragments selected for high CpG density, mixed with mouse housekeeping-promoter controls, in TC-1 mouse embryonic stem cells. The assay tests whether CpG density alone is sufficient to drive transcription in chromatin.