Fig. 2

Histone methyltransferase assay. Full-length wild-type GFP-Ezl1^wt and mutant GFP-Ezl1^H526A proteins were immunoprecipitated at 12 h after the onset of sexual events (see Supplementary Fig. 2) from Paramecium transformed cells, in which endogenous Ezl1 protein expression was specifically depleted through RNA interference, and used for in vitro histone methyltransferase reactions (a–c). a Wild-type Xenopus recombinant histone octamers are used as substrates and S-adenosyl-[methyl 3H]-methionine as methyl donor. Affinity-purified recombinant PRC2 complex is used as a positive control. The same reactions with histone octamers only were performed as a negative control. Coomassie stain (CB, bottom panel) shows histones and the autoradiograph (³H, top panel) indicates H3 histone methyltransferase activity. The graphs show relative quantification of histone methyltransferase signals analyzed by scintillation counting after SDS-PAGE. b Histone methyltransferase activity on recombinant histone octamers (rOct) and on oligonucleosomes (rNuc). c Substrate specificity. Histone methyltransferase assays were performed with recombinant histone octamers assembled with wild-type H3 histones or with H3 histones bearing a point mutation on lysine 9 (K9A) or on lysine 27 (K27A) or on both (K9A/K27A) on the same histone H3 tail. The same reaction without addition of histone octamers was performed as a negative control. For Ezl1^wt and Ezl1^H526A, the horizontal bars indicate the mean of three biological replicates (independent nuclear extracts). Circles indicate the individual data points. d Full-length 3xFLAG-HA tagged Ezl1^wt and Ezl1^H526A proteins were immunoprecipitated using a FLAG-HA double-affinity purification from Paramecium transformed cells, depleted for the endogenous Ezl1 protein in the case of the mutant FLAG-HA-Ezl1^H526A, and used in in vitro histone methyltransferase reactions (see Supplementary Fig. 2). Histone methyltransferase assays were performed with recombinant histone tetramers assembled with wild-type H3 from Paramecium or with Paramecium H3 histones bearing point mutations on both K9 and K27 (K9R/K27R) on the same histone tail. The horizontal bars indicate the mean of two or three biological replicates (independent nuclear extracts) for Ezl1^wt and Ezl1^H526A, respectively. e Schematic representation of quantitative label-free mass spectrometry analysis of Paramecium histone H3. See Table 1 and Supplementary Data 1. Owing to the insertion of one residue at position 14 in the Paramecium H3 sequence (see Supplementary Fig. 2), positions in the Xenopus and Paramecium H3 sequences differs by one amino acid gap after that position (Supplementary Fig. 2 and Supplementary Data 1). For simplicity, K24 and K28 in the Paramecium H3 sequence are referred to as K23 and K27 in the schema and in the manuscript. Source data are provided as a Source Data file