Poster Presentation 45th Lorne Genome Conference 2024

The epigenetic reader WDR5 is important for KLF3 genomic localisation and gene regulation (#245)

Manan Shah 1 , Lu Yang 2 , Wooi F Lim 2 , Tanit Chavalit 2 , Annalise M Psaila 1 , Ling Zhong 2 , Mark J Raftery 1 , Joel P Mackay 3 , Jacqueline M Matthews 3 , Marc R Wilkins 1 , Kate GR Quinlan 1 , Merlin Crossley 1
  1. School of Biotechnology and Biomolecular Sciences2, UNSW Sydney, Sydney, NSW, Australia
  2. School of Biotechnology and Biomolecular Sciences, UNSW Sydney, Sydney, NSW, Australia
  3. School of Life and Environmental Sciences, The University of Sydney, Sydney, NSW, Australia

Krüppel-like factor 3 (KLF3) is a member of the SP/KLF family of transcription factors (TFs). They bind GC-rich DNA sequence elements via their 3 zinc fingers near their C-termini. KLF3 is known primarily as a repressor, however, knockout studies have shown it can also activate genes. It represses transcription by recruiting the co-repressor CtBP via its N-terminal or functional domain (FD) but how it activates genes is unknown. We have previously shown that the KLF3 FD and in part its interaction with CtBP is not only critical for gene regulation but also KLF3 genomic localisation.  

In this study, we identify a novel co-factor of the KLF3 FD, WDR5 using co-immunoprecipitation coupled mass spectrometry (CoIP/MS). WDR5 is a member of the WD repeat protein family which can directly bind the histone H3 tail but cannot independently bind DNA. Next, by mutational screening, we identified a KLF3 residue critical for the KLF3/WDR5 interaction. We generated cell lines containing either WT KLF3 or KLF3 with the mutation disrupting the association. ChIP-seq of both KLF3 and WDR5 revealed that up to 40% of both their binding sites were differentially bound when comparing WT and mutant KLF3. Gene expression was also altered with more than 1000 genes significantly deregulated. Correlating binding changes with gene expression revealed that loss of KLF3 and/or WDR5 binding primarily led to gene repression.   

In conclusion, we add to only a handful of examples of a non-DNA-binding cofactor not only affecting gene regulation exerted by a TF but also its genomic localisation. We also demonstrate a potential mechanism by which KLF3 can activate gene expression, which was previously unknown. Furthermore, as WDR5 can recognise specific histone post-translational modifications on the histone H3 tail, it suggests a potential mechanism for how histone PTMs can affect TF binding.