Gene Regulation by the Polycomb Group pathway
Polycomb Group proteins in plants and animals
Polycomb Group (PcG) protein complexes are conserved in plants and animals and built a system to stably repress genes by local compaction of chromatin. PcG-mediated repression can confer a molecular memory to gene regulation, either because genes that are down-regulated by an initial signal may not require the continued presence of this signal once PcG complexes have been recruited or because repressed genes that are activated despite the presence of PcG-complexes lose their chromatin-mediated repression after transcription induction .
Figure 1. Polycomb Repressive Complex 2 (PRC2) is recruited either by the action of DNA-binding transcription factors, ncRNAs or other mechanisms. PRC2 contains an HMTase that tri-methylates lysine 27 of H3 (H3K27me3). Starting from an H3K27me3 nucleation site, which is linked to a PcG-recruiting region, the H3K27me3 spreads to flanking regions because the PRC2 recognizes its own target modification and is further activated by binding to the modified H3. Polycomb Repressive Complex 1 (PRC1) is recruited by H3K27me3 and further modifies chromatin by ubiquitination of lysine 119 (or related positions) of H2A (H2AK119ub) through two components that contain RING domains. H2AK119ub is important for chromatin compaction and subsequent gene repression.[less]
Figure 1. Polycomb Repressive Complex 2 (PRC2) is recruited either by the action of DNA-binding transcription factors, ncRNAs or other mechanisms. PRC2 contains an HMTase that tri-methylates lysine 27 of H3 (H3K27me3). Starting from an H3K27me3 nucleation site, which is linked to a PcG-recruiting region, the H3K27me3 spreads to flanking regions because the PRC2 recognizes its own target modification and is further activated by binding to the modified H3. Polycomb Repressive Complex 1 (PRC1) is recruited by H3K27me3 and further modifies chromatin by ubiquitination of lysine 119 (or related positions) of H2A (H2AK119ub) through two components that contain RING domains. H2AK119ub is important for chromatin compaction and subsequent gene repression.
PcG proteins form two important complexes that function sequentially to repress target genes. The Polycomb Repressive Complex 2 (PRC2) is recruited to nucleation sites by mechanisms that are not yet fully understood. PRC2 modifies the chromatin by tri-methylating the lysine amino-acid residue located at position 27 of the amino-terminal tail of histone H3. The resulting modification (H3K27me3) acts as label to recruit Polycomb Repressive Complex 1 (PRC1), which contains a subunit that specifically binds to the modified H3 tail. PRC1 also encompasses proteins that ubiquitinate a lysine residue in the body of histone H2A. This modification seems particularly important for chromatin compaction and gene repression.
An important feature of PcG-mediated gene repression is a phenomenon called spreading. The PcG histone modification spreads into regions flanking the nucleation site because the PRC2 complex is also recruited by H3K27me3.
Chromatin compaction, H3K27me3 spreading and PRC2 nucleation are observed at many target genes, but how the system actually represses transcription is still not fully understood and may in fact vary from gene to gene 
Many key genes in plant development depend on Polycomb Group-mediated repression
Approximately 15% of all Arabidopsis genes are marked by H3K27me3 [2-4]. As expected, their expression is usually low or restricted to particular organs or developmental stages [2, 4]. Many PcG target genes play important roles in development [4-6]. Mis-regulation of these key developmental genes explains the abnormal development of Arabidopsis plants that carry mutations in single PcG components.
Figure 2. DPA4 and CUC genes controls leaf margin development. From top clockwise: normal 6th leaf of Arabidopsis, of cuc2 mutants, of cuc2;dpa4 double mutants, of dpa4 single mutants, of dpa4 mutants that overexpress miR164a, of non-mutant plants overexpressing miR164a, of plants overexpressing DPA4 and miR164a and of plants overexpressing DPA4.[less]
Figure 2. DPA4 and CUC genes controls leaf margin development. From top clockwise: normal 6th leaf of Arabidopsis, of cuc2 mutants, of cuc2;dpa4 double mutants, of dpa4 single mutants, of dpa4 mutants that overexpress miR164a, of non-mutant plants overexpressing miR164a, of plants overexpressing DPA4 and miR164a and of plants overexpressing DPA4.
In the past, we have made use of the observations outlined above to identify new genes that play a role in Arabidopsis development by screening genes that are H3K27me3 targets and expressed in the shoot apex . At the shoot apex all aerial organs are formed and important developmental decisions are established. One gene that we have further characterized is DEVELOPMENT-RELATED PcG TARGET IN TH APEX 4 ( DPA4), which controls leaf margin architecture . DPA4 acts together with CUP-SHAPED COTELYDON 2 ( CUC2) to shape leaf development, but it is yet unclear at which step of DPA4 regulation PcG-mediated repression becomes important. For other genes, such as FLOWERING LOCUS T (hyperlink to project 1), chromatin repression plays a structural role and orchestrates the input of many independent transcription factors. Another important PcG-target in plants is FLOWERING LOCUS C, for which PcG repression confers the molecular memory of winter by stabilizing the down-regulation of FLC during vernalization.
Ongoing Research in our group
It is still not quite clear how PcG complexes select their targets. The PRC2 seems to be recruited by non-coding RNAs or by cis-regulatory elements that are bound to DNA-binding proteins, which in interact with PRC2. But also the mode of recruitment of the PRC1 complex is not quite clear. The PRC1 complex interacts with H3K27me3, which would provide an evident signal for PRC1 recruitment. However, the only Arabidopsis PRC1 component that possesses an H3K27me3 interacting chromo-domain is LIKE-HETEROCHROMATIN PROTEIN 1 (LHP1) . If LHP1 were the only protein involved in PRC1 recruitment, one would expect the phenotype of plants carrying mutations in this gene to be very strong. However, compared to mutations that severely affect the H2A ubiquitination function of PRC1, the lhp1 mutants display a mild developmental phenotype .
How PRC2 complexes find their targets
Arabidopsis thaliana as model organism has one of the best annotated and characterized genomes. We therefore compared H3K27me3 target regions in different Arabidopsis accessions. The aim is to infer from sequence differences found at sites differentially labeled with H3K27me3 by which mechanisms these differences can be best explained.
Research in Drosophila has revealed that PRC2 recruitment sites are more strongly bound by PRC2-complexes than the much more wide-spread H3K27me3 target regions in general . We therefore initiated a project to map PRC2 binding sites in Arabidopsis. The perspective is to recognize PRC2 recruiting features by an analysis of many PRC2 binding sites.
Solving the LHP1 mystery
Assuming that H3K27me3 recognition is an important aspect of PRC1 recruitment and function, we postulate that there must be other components that act redundantly with LHP1. We therefore initiated a genetic screen for enhancers of the lhp1 mutation. We identified 20 enhancers so far of which 3 did not show a phenotype in the absence of the lhp1 mutation. In collaboration with the group of Korbinian Schneeberger we use NGS sequencing to identify the genes that are affected in the enhancer mutants and have developed an improved method that allows mapping mutations in an isogenic background . Currently, we are further characterizing the enhancer mutations to place them within the PcG pathway. In this project we collaborate with the groups of Justin Goodrich and Myriam Calonje.
Figure 3. Examples of genetic enhancer mutations in the lhp1 background.
Figure 3. Examples of genetic enhancer mutations in the lhp1 background.
- Farrona, S., G. Coupland, and F. Turck, The impact of chromatin regulation on the floral transition. Semin Cell Dev Biol, 2008. 19(6): p. 560-73.
- Farrona, S., F.L. Thorpe, J. Engelhorn, J. Adrian, X. Dong, L. Sarid-Krebs, J. Goodrich, and F. Turck, Tissue-specific expression of FLOWERING LOCUS T in Arabidopsis is maintained independently of polycomb group protein repression. Plant Cell, 2011. 23(9): p. 3204-14.
- Turck, F., F. Roudier, S. Farrona, M.L. Martin-Magniette, E. Guillaume, N. Buisine, S. Gagnot, R.A. Martienssen, G. Coupland, and V. Colot, Arabidopsis TFL2/LHP1 specifically associates with genes marked by trimethylation of histone H3 lysine 27. PLoS Genet, 2007. 3(6): p. e86.
- Zhang, X., O. Clarenz, S. Cokus, Y.V. Bernatavichute, M. Pellegrini, J. Goodrich, and S.E. Jacobsen, Whole-genome analysis of histone H3 lysine 27 trimethylation in Arabidopsis. PLoS Biol, 2007. 5(5): p. e129.
- Engelhorn, J., J.J. Reimer, I. Leuz, U. Gobel, B. Huettel, S. Farrona, and F. Turck, Development-related PcG target in the apex 4 controls leaf margin architecture in Arabidopsis thaliana. Development, 2012. 139(14): p. 2566-75.
- Engelhorn, J. and F. Turck, Metaanalysis of ChIP-chip data. Methods Mol Biol, 2010. 631: p. 185-207.
- Bratzel, F., G. Lopez-Torrejon, M. Koch, J.C. Del Pozo, and M. Calonje, Keeping cell identity in Arabidopsis requires PRC1 RING-finger homologs that catalyze H2A monoubiquitination. Curr Biol, 2010. 20(20): p. 1853-9.
- Enderle, D., C. Beisel, M.B. Stadler, M. Gerstung, P. Athri, and R. Paro, Polycomb preferentially targets stalled promoters of coding and noncoding transcripts. Genome Res, 2011. 21(2): p. 216-26.
- Hartwig, B., G. Velikkakam James, K. Konrad, K. Schneeberger, and F. Turck, Fast isogenic mapping-by-sequencing of EMS-induced mutant bulks. Plant Physiol, 2012.