RAWUL: a new ubiquitin-like domain in PRC1 ring finger proteins that unveils putative plant and worm PRC1 orthologs

Abstract

Background: Polycomb group (PcG) proteins are a set of chromatin-modifying proteins that play a key role in epigenetic gene regulation. The PcG proteins form large multiprotein complexes with different activities. The two best-characterized PcG complexes are the PcG repressive complex 1 (PRC1) and 2 (PRC2) that respectively possess histone 2A lysine 119 E3 ubiquitin ligase and histone 3 lysine 27 methyltransferase activities. While PRC2-like complexes are conserved throughout the eukaryotic kingdoms, PRC1-like complexes have only been described in Drosophila and vertebrates. Since both complexes are required for the gene silencing mechanism in Drosophila and vertebrates, how PRC1 function is realized in organisms that apparently lack PRC1 such as plants, is so far unknown. In vertebrates, PRC1 includes three proteins, Ring1B, Ring1A, and Bmi-1 that form an E3 ubiquitin ligase complex. These PRC1 proteins have an N-terminally located Ring finger domain associated to a poorly characterized conserved C-terminal region.

Results: We obtained statistically significant evidences of sequence similarity between the C-terminal region of the PRC1 Ring finger proteins and the ubiquitin (Ubq)-like family proteins, thus defining a new Ubq-like domain, the RAWUL domain. In addition, our analysis revealed the existence of plant and worm proteins that display the conserved combination of a Ring finger domain at the N-terminus and a RAWUL domain at the C-terminus.

Conclusion: Analysis of the conserved domain architecture among PRC1 Ring finger proteins revealed the existence of long sought PRC1 protein orthologs in these organisms, suggesting the functional conservation of PRC1 throughout higher eukaryotes.

Figure 1

Representative multiple alignment of the RAWUL domain. The coloring scheme indicates average BLOSUM62 score (correlated to amino acid conservation) in each alignment column: red (greater than 1.8), violet (between 1.8 and 1) and light yellow (between 1 and 0.2). Residues that are part of the hydrofobic core of the Ubq fold [50] are indicated by green ovals above the alignment. The limits of the domains are indicated by the residue positions on each side. The main families and subfamilies are indicated by coloured bars to the left of the alignment; green (PRC1), violet (WD48-p80), red (Bmi1/Mel18) and yellow (Ring1A/B). Structure based sequence alignment [86] of representative members of UBL superfamily were included: pdb_1gnu: human gamma-aminobutyric acid receptor type A receptor-associated protein (GABARAP) [84]; pdb_1m94: yeast Ubl modifier protein homologous to ubiquitin 1 (Hub1) [87]; pdb_1yqb: human ubiquilin 3 [88]; pdb_1mg8: Ubl domain from mouse Parkin protein [89]; pdb_1aar: bovine Ubiquitin [90]. Independent PHD secondary structure predictions [55] for PRC1 Ring and WDR48-p80 families are shown below the family sequences. Consensus secondary structure of the Ubiquitin superfamily is shown below the alignment [54]. Alpha-helices and beta-strands are indicated by green cylinders and arrows, respectively. The sequences are named with their Uniprot identifications, and also, if necessary, with their common species name: Human, Homo sapiens; Drome, Drosophila melanogaster; Caeel, Caenorhabditis elegans; Arath, Arabidopsis thaliana; Orysj, Oryza sativa; Crypv, Cryptosporidium parvum; Schja, Schistosoma japonicum; Aspfu, Aspergillus fumigatus; Neucr, Neurospora crassa; Kluma, Kluyveromyces marxianus; Debha, Debaryomyces hansenii; Schpo, Schizosaccharomyces pombe; Yeast, Saccharomyces cerevisiae. The "est" prefix identifies sequence corrections using consensus sequences manually reconstructed by assembling highly similar expressed sequence tags from identical species (conceptual translations). The numbers in blue within the alignment represent sequence insertions that are not included.

Figure 2

HMM-HMM profile comparison E-Values between the RAWUL and Ubl domains. The numbers correspond to HMM-HMM profile comparisons E-values from global profile search results [49] that connect independently each family with the others. The arrows indicate the profile search direction.

Figure 3

Dendrogram and Domain Architecture for RAWUL domain containing proteins. The sequences are named as described in Figure 1. Vitis vinifera (Vitvi) sequences were included to construct the dendrogram. The main families (PRC1 Ring and WDR48-p80) and subfamilies (Bmi1/Mel18 and Ring1A/B) are indicated by coloured dotted lines, as in Figure 1. Only a representative set of sequences was labelled. The scale bar shows the average number of amino acid substitutions per site (0.1). The main bootstrap values for PRC1 Ring family classification are indicated. Schematics of domain architecture are represented with the localization of the RAWUL, Ring domains and WD40 repeats according to this analysis, Pfam [91] and REP web servers [92], respectively.

Figure 4

Representative full sequence alignment of the PRC1 Ring family. The coloring scheme indicates average BLOSUM62 score (correlated to amino acid conservation) in each alignment column: red (greater than 4), violet (between 4 and 1.5) and light yellow (between 1.5 and 0.5). The sequences are named as described in Figure 1. Lowercase characters represent parts of the sequence that are not evolutionarily conserved. In the PRC1 Ring family the N-terminal Ring domain (blue rectangle) is more conserved than the C-terminal RAWUL domain (green rectangle).