Evolution of lysine-specific demethylase 1 and REST corepressor gene families and their molecular interaction

Abstract

Lysine-specific demethylase 1A (LSD1) binds to the REST corepressor (RCOR) protein family of corepressors to erase transcriptionally active marks on histones. Functional diversity in these complexes depends on the type of RCOR included, which modulates the catalytic activity of the complex. Here, we studied the duplicative history of the RCOR and LSD gene families and analyzed the evolution of their interaction. We found that RCOR genes are the product of the two rounds of whole-genome duplications that occurred early in vertebrate evolution. In contrast, the origin of the LSD genes traces back before to the divergence of animals and plants. Using bioinformatics tools, we show that the RCOR and LSD1 interaction precedes the RCOR repertoire expansion that occurred in the last common ancestor of jawed vertebrates. Overall, we trace LSD1-RCOR complex evolution and propose that animal non-model species offer advantages in addressing questions about the molecular biology of this epigenetic complex.

Fig. 1. Schematic representation of human RCOR1 and LSD1 proteins highlighting their functional domains.

ELM2 (Egl-27 and MTA1 homology 2) domain; SANT1 (Swi3, Ada2, NCoR, and TFIIIB) and SANT2 domains; SWIRM (Swi3, Rsc8, and Moira) domain; AOD-N (N-terminus amino oxidase domain); AOD-C (C-terminus amino oxidase domain).

Fig. 2. Maximum-likelihood tree showing relationships among RCOR genes of vertebrates.

Numbers on the nodes correspond to support from the Shimodaira–Hasegawa approximate likelihood-ratio test, approximate Bayes test, and ultrafast bootstrap values. The scale denotes substitutions per site and colors represent gene lineages. Mitotic deacetylase-associated SANT domain protein (MIDEAS) sequences from human (Homo sapiens), gorilla (Gorilla gorilla), mouse (Mus musculus), and ferret (Mustela putorius furo) were used as outgroup (not shown). For more details of the species included in the analysis, refer to Supplementary Data 1.

Fig. 3. Maximum-likelihood tree showing relationships among RCOR genes of metazoa.

Numbers on the nodes correspond to support from the Shimodaira–Hasegawa approximate likelihood-ratio test, approximate Bayes test, and ultrafast bootstrap values. The scale denotes substitutions per site and colors represent gene lineages. Mitotic deacetylase-associated SANT domain protein (MIDEAS) sequences from human (Homo sapiens), gorilla (Gorilla gorilla), mouse (Mus musculus), and ferret (Mustela putorius furo) were used as outgroup (not shown). For more details on the species included in the analysis, refer to Supplementary Data 2.

Fig. 4. Maximum-likelihood tree showing relationships among LSD genes of vertebrates.

Numbers on the nodes correspond to support from the Shimodaira–Hasegawa approximate likelihood-ratio test, approximate Bayes test, and ultrafast bootstrap values. The scale denotes substitutions per site and colors represent gene lineages. Monoamine-oxidases (MAO-A and MAO-B) sequences from human (Homo sapiens), chicken (Gallus gallus), spotted gar (Lepisosteus oculatus), and coelacanth (Latimeria chalumnae) were used as outgroup (not shown). For more details on the species included in the analysis, refer to Supplementary Data 3.

Fig. 5. Phyletic distributions of RCOR and LSD genes, LSD1 microexon, and neuron-specific splicing regulatory sequences in vertebrates.

a Distribution of the RCOR and LSD genes, LSD1 microexon, and neuron-specific splicing regulatory sequences in main groups of vertebrates. b Dot-plot of pairwise sequence similarity between the RCOR2 gene of the painted turtle (Chrysemys picta) and the corresponding syntenic region in the chicken (Gallus gallus) and New Caledonian crow (Corvus moneduloides). ^aRCOR2 is present only in the bird orders Psittaciformes, Passeriformes, Accipitriformes, and Anseriformes. ^bLampreys have a single copy of the RCOR gene. ^cSome species of birds have a -DTVE- microexon. For more details on the species included in the analysis, refer to Supplementary Data 5.

Fig. 6. LSD1 and RCORs normalized conservation scores of jawed vertebrates.

Orange dots represent the conservation scores of the key amino acids for LSD1-RCOR interaction. The red line in RCORs denotes the linker domain required for interaction with LSD1. The blue line shows a pre-linker highly variable sequence with intrinsically disordered characteristics. Black asterisks indicate conserved sequence patches. Position numbers correspond to human proteins.

Fig. 7. Color-coded 3D amino acid conservation of LSD1-RCOR protein complexes.

a LSD1 tower domain residue conservation projected onto the structure of human LSD1-RCOR1 (PDB: 2V1D). The red arrow shows Lys447 residue (see text). b RCOR1 linker-SANT2 domain residue conservation projected onto the structure of human LSD1-RCOR1 (PDB: 2V1D). Light blue arrows show Arg347, Gln350, and Gln354 residues (see text). c RCOR3 linker-SANT2 domain residue conservation projected onto the structure of human LSD1-RCOR3 (PDB: 4CZZ). Light blue arrows show Gln315, Asn316, and Gln319 residues (see text).

Fig. 8. Predicted molecular interaction between the jawed vertebrate ancestral LSD1 and RCOR proteins.

a Graphical representation of the ancestral RCOR and LSD1 protein sequence reconstructions. b Sequence alignments of the linker region and SANT2 domain of the putative jawed vertebrate ancestral RCOR (AR), with human RCOR1 (R1), RCOR2 (R2), and RCOR3 (R3). c Superposition of the 3D structure of the interacting surfaces of the human RCOR1-LSD1 complex with the predicted 3D structure of the ancestral RCOR-LSD1 complex. I, II and IV highlight key amino acids for interaction in the ancestral complex.