While reduced protein synthesis provides a model to explain

While reduced protein synthesis provides a model to explain the long-term memory defects of kdm5 and kdm5 flies, new translation is not required for short-term memories. The chronic reduction in translation observed in mutant fly strains may, however, lower the levels of proteins necessary for short-term memory formation. Alternatively, defects in distinct pathways may cause the short- and long-term phenotypes. For example, genes required for the production and transmission of octopamine, a neurotransmitter required for short-term memory (Schwaerzel et al., 2003), were downregulated in kdm5 and kdm5 flies. It is also possible that other dysregulated genes in kdm5 and kdm5 flies contribute to both the short- and long-term learning and/or memory defects. Our analyses of two mutations that abolish the demethylase activity of KDM5 have provided key insights into potential mechanisms of transcriptional regulation. Promoter proximal H3K4me3 is found at transcriptionally active genes and is recognized by many proteins, including components of the transcriptional initiation machinery (Vermeulen et al., 2007). Because they demethylate H3K4me3, the enzymatic activity of KDM5 proteins is regarded as one that represses transcription. However, we find that KDM5 can directly activate or repress transcription in a demethylase-dependent manner. This is not due to the enzymatic activity of KDM5 varying at different targets, since levels of promoter-proximal H3K4me3 increase similarly at upregulated, downregulated, and unaffected genes. The observation that levels of promoter H3K4me3 are not a driving force in defining gene ML347 mg levels has been noted previously (Benayoun et al., 2014, Lussi et al., 2016). Because we were unable to find enrichment for specific transcription factors at KDM5-regulated genes, KDM5 may be recruited downstream of other events that occur at its target promoters to facilitate transcriptional activation or repression. Supporting this model, we found that genes activated by KDM5 were often bound also by the transcription factor Myc and by the insulator protein BEAF-32. Indeed, Myc and BEAF-32 binding have previously been observed to correlate in cultured Drosophila Kc cells, although whether these proteins act to coordinate chromatin domains or transcriptional events is not known (Yang et al., 2013). We and others have also observed a physical interaction between KDM5 and Myc proteins in flies and mammalian cells (Li et al., 2010, Outchkourov et al., 2013, Secombe et al., 2007). However, because Myc and KDM5 peaks at co-regulated genes are not always coincident, it is not clear whether this interaction is important in the context of the adult brain. It is interesting to note that mutations in human N-Myc cause the cognitive disorder Feingold syndrome (Cognet et al., 2011), and knockdown of Drosophila Myc results in synaptic defects (Oortveld et al., 2013). Transcriptional cooperation between Myc and KDM5 in neurons may thus contribute to these cognitive phenotypes through the regulation of genes required for translation.
Experimental Procedures
Acknowledgments The authors thank the Secombe laboratory members and Peter Gallant. We are grateful for funding from the March of Dimes (6-FY17-315) and NIH R01GM112783 to J.S., NIH Training Grant T32GM007288 to H.A.M.H., the Developmental Studies Hybridoma Bank (DSHB), the NY Genome Center, and the Cancer Center Support Grant (P30 CA013330).
Introduction Gene regulation in multicellular organisms entails diverse strategies that operate over different distances. Some mechanisms act locally on individual genes, whereas others act over long range to co-regulate expression of numerous genes within sub-chromosomal domains (e.g., imprinted gene clusters, hox clusters) (Andrey et al., 2013, Plasschaert and Bartolomei, 2014, Williamson et al., 2012), or across entire chromosomes (e.g., X chromosome dosage compensation) (da Rocha and Heard, 2017, Kuroda et al., 2016, Meyer, 2010, Strome et al., 2014). Post-translational modifications of histones and three-dimensional (3D) chromosome architecture are prime elements of long-range gene regulation, but their functional interplay in controlling gene expression has been elusive (Dekker and Mirny, 2016, Soshnev et al., 2016).