Identification of a specific inhibitor of the histone methyltransferase SU(VAR)3-9 Dorothea Greiner 1 , Tiziana Bonaldi 1 , Ragnhild Eskeland 1 , Ernst Roemer 2,3 & Axel Imhof 1 Histone methylation plays a key role in establishing and maintaining stable gene expression patterns during cellular differentiation and embryonic development. Here, we report the characterization of the fungal metabolite chaetocin as the first inhibitor of a lysine-specific histone methyltransferase. Chaetocin is specific for the methyltransferase SU(VAR)3-9 both in vitro and in vivo and may therefore be used to study heterochromatin-mediated gene repression. During the life of every multicellular organism, totipotent cells have to acquire specific functions and maintain their differentiated state. The differentiated state of a cell is determined by its specific pattern of gene expression, which in turn is established and maintained through the differential packaging of DNA into chromatin. The basic unit of chromatin is the nucleosome, a nucleoprotein particle that consists of 147 base pairs of DNA that are wrapped around a proteinaceous core of the four core histones H2A, H2B, H3 and H4 (ref. 1). It is widely accepted that the post-translational modifications of the histone N-terminal tails, as well as modifications within the globular domain, regulate the level of chromatin condensation and are therefore important in regulating gene expression 2 . Inhibitors of histone acet- yltransferases can affect the heritable changes in gene expression of specific genes and are used as drugs for cancer therapy 3 . In addition to histone acetylation, histone methylation has also been shown to be important in establishing stable gene-expression patterns. Some of the known histone methyltransferases (HMTs) are misregulated in tumors 4 , and methyltransferase-induced heterochromatin formation could be involved in neurodegenerative diseases 5 . To find small molecules that affect HMT function, we screened a small library of compounds for their ability to inhibit the activity of recombinant Drosophila melanogaster SU(VAR)3-9 protein. SU(VAR)3-9 is a key player in establishing condensed heterochroma- tin by specifically methylating Lys9 of histone H3 (ref. 6) and is conserved in most higher eukaryotes. The screening was done by pooling eight individual compounds and testing the pooled mixtures in a standard radioactive filter-binding assay (Supplementary Methods online). Every pool that reduced HMT activity by more than 60% was considered inhibitory, and the individual compounds were retested for activity (Fig. 1a). Of the 2,976 compounds screened, 22 reduced HMT activity by more than 80% under standard con- ditions. To verify the activity, and to exclude a possible peptide- specific effect, we further tested the most active inhibitors using intact recombinant H3 molecules as substrates (Fig. 1b). One of the strongest inhibitors was the fungal mycotoxin chaetocin (Fig. 1c, 1), which was initially isolated from the fermentation broth of Chaetomium minutum 7 and belongs to the class of 3-6 epi- dithio-diketopiperazines (ETPs). It has a half-maximal inhibitory Coomassie Autorad DMSO b c Mock 32F04/chaetocin 59C05 57A12 59E05 41A01 40D08 1 2 3 4 5 6 7 8 9 a Pools of eight compounds tested <65% activity Retest of individual compounds Chaetocin (1) 0 20 40 60 80 100 120 140 0 5 10 15 20 E(z) complex PR SET7 dSU(VAR)3-9 SUV39H1 G9a DIM5 SET7/9 Activity (%) Chaetocin (μM) d H H S N O N O N H S HO S N O N O H N S OH CH 3 CH 3 Figure 1 In vitro inhibition of dSU(VAR)3-9. (a) Schematic outline of the inhibitor screen. (b) SDS-PAGE analysis of in vitro methylated H3 in the presence or absence of 10 mg of inhibitor. (c) Structure of chaetocin. The chaetocin used in all reactions was 495% pure (Supplementary Fig. 2 online). (d) Inhibition curves for various recombinant HMTs. Methyltransferase assays were performed as described previously 13 . Reactions were done in duplicate and error bars reflect the standard deviation from at least two different experiments. Published online 17 July 2005; doi:10.1038/nchembio721 1 Adolf Butenandt Institute, Department of Molecular Biology, Histone Modifications Group, Ludwig-Maximillians University of Munich, Schillerstr. 44, 80336 Munich, Germany. 2 Leibniz-Institut fu ¨ r Naturstoff-Forschung und Infektionsbiologie–Hans-Kno ¨ ll-Institut fu ¨ r Naturstoff-Forschung, Abteilung Molekulare Naturstoff-Forschung, Beutenbergstrasse 11a, 07745 Jena, Germany. 3 Present address: Institut fu ¨ r Pflanzenbiochemie, Universita ¨t Halle, Weinberg 3, Halle (Saale) 06120, Germany. Correspondence should be addressed to A.I. (imhof@lmu.de). NATURE CHEMICAL BIOLOGY VOLUME 1 NUMBER 3 AUGUST 2005 143 BRIEF COMMUNICATIONS © 2005 Nature Publishing Group http://www.nature.com/naturechemicalbiology