The KMET-READ proposal encompassed three scientific objectives that were investigated in a multidisciplinary approach.
Objective 1: Functional characterization of the PHDs of the MLL 2 and 3.
Interestingly, KMT2C and KMT2D display high sequence similarity to each other reflected in the organization of their six and seven PHD domain, respectively. We have investigated the triple PHD domains (PHD 1-3 and PHD 4-6 of KMT2D and PHD 1-4 and 5-7 KMT2C) on cellulospot and peptide arrays and discovered interesting new combinatorial readout of histone PTMs that was initially confirmed in pulldown experiments. Further to validate the binding and visualize binding pocket together with the collaborator Prof. Bochtler in International Institute of Molecular and Cell Biology in Warsaw, we aimed for solving the crystal structures of these domains in free and target bound state and to design binding pocket mutants. Altogether 10 binding pocket mutants were generated and tested. Unfortunately, all attempts performed so far using diverse techniques and bioinformatics approaches were unsuccessful in the sense that no strucutre could be obtained and the binding pockets are still undefind. We have expressed investigated domains in the mammalian cellular system and observed a distinctive sub-nuclear distribution that prompted us to characterize the chromatin binding specificity of these domains in mammalian cells by immunostaining and CIDOP-seq. The initial results of this objective were presented at international conferences and are still under investigation.
Objective 2: Characterization of the SUV39H1 CDs.
SUV39H1 is crucial for heterochromatin formation by setting the H3K9me3 repressive mark and possess in the structure chromodomain that bounds to the H3K9me3. Previous data suggested that the chromodomain might affect SUV39H1’s enzymatic activity in placing the mark in the genome. Experiments during the MSC project aimed to evaluate the chromatin binding specificity of the SUV39H1 chromodomain in mammalian cells and elucidate its biological function in setting and spreading the H3K9me3 mark. We developed DOX-inducible dKO Suv39h1/h2 mouse fibroblasts (MEF) cell lines expressing SUV39H1 with wildtype (CD-WT) or a mutant (CD-Y67A) form carrying an amino acid exchange in the chromodomain which disrupts its methyllysine interaction. Interestingly, we observed a significant difference between CD-WT and CD-Y67A in the amount of H3K9me3 in WesternBlot and ChIP-qPCR. Moreover, we have performed ChIP-seq experiments for the H3K9me3 and SUV39H1-CD-WT or –CD-Y67A to observe differences in the amount and localization of H3K9me3 mark and the binding of the enzymes. The samples were sent for sequencing and currently are processed for further analysis.
Objective 3: Identification of new reader domains using a novel screening method.
While tremendous progress has been made in the detection of histone methylation, the identification and function of reader domains especially for non-histone lysine methylation is lagging behind. This project successfully adopted a yeast-3-hybrid method for screening libraries to identify new methyl lysine readers. In our system three proteins are expressed in yeast cells, viz. a bait fused to GAL4- binding domain, a methyltransferase that modifies the bait, and a GAL4- activation domain fused to prey responsible for recognition of methylated bait. For proof of principle, we have used a short H3 construct as bait and screened for H3K9me3 methylation-specific readers. We have used the SET domain of G9a as methyltransferase to catalyze H3K9 trimethylation. As prey, we have adopted the chromodomains of CBX1 and MPP8 which are known to be able to bind H3K9me3. If the methylated bait interacts with the prey, the GAL4 transcription factor is reconstituted and triggers transcription of reporter genes. Control experiments clearly confirmed the applicability of the modified yeast three-hybrid system in the detection of methyllysine dependent protein-protein interactions. We have also performed a screen using a normalized human cDNA library fused with GAL4-AD to identify novel H3K9me3 readers. The identification of already known H3K9me3 readers in the library screen confirmed the functionality of the method. Moreover, we have detected novel readers of H3K9me3 that were evaluated using another method and achieved results strongly suggesting that the newly identified factors are H3K9 methylation readers. Currently, a manuscript describing a developed method and summarizing data achieved in the screening of H3K9me3 readers is submitted for publication.
