Emergence of the Synucleins

Abstract

This study establishes the origin and evolutionary history of the synuclein genes. A combination of phylogenetic analyses of the synucleins from twenty-two model species, characterization of local synteny similarities among humans, sharks and lampreys, and statistical comparisons among lamprey and human chromosomes, provides conclusive evidence for the current diversity of synuclein genes arising from the whole-genome duplications (WGDs) that occurred in vertebrates. An ancestral synuclein gene was duplicated in a first WGD, predating the diversification of all living vertebrates. The two resulting genes are still present in agnathan vertebrates. The second WGD, specific to the gnathostome lineage, led to the emergence of the three classical synuclein genes, SNCA, SNCB and SNCG, which are present in all jawed vertebrate lineages. Additional WGDs have added new genes in both agnathans and gnathostomes, while some gene losses have occurred in particular species. The emergence of synucleins through WGDs prevented these genes from experiencing dosage effects, thus avoiding the potential detrimental effects associated with individual duplications of genes that encode proteins prone to aggregation. Additional insights into the structural and functional features of synucleins are gained through the analysis of the highly divergent synuclein proteins present in chondrichthyans and agnathans.

Keywords: Parkinson’s disease; agnathans; gnathostomes; synuclein; whole-genome duplication.

Figure 1

Model of synuclein emergence linked to the vertebrate WGDs and predictions derived from that model. A single synuclein-coding gene was present in the ancestor of all vertebrates. WGD1 generated two different genes prior to the agnathan/gnathostome split. In the gnathostome lineage, WGD2 generated four synuclein genes, one of which (“δ-syn”) was lost before gnathostome diversification. Agnathans may have additional duplicates due to the hexaploidization of their genomes.

Figure 2

Maximum-likelihood trees obtained from the analysis of gnathostome synuclein proteins. Species names and accession numbers are used for sequence identification; the complete list is available in Supplementary File S1. Numbers indicate the percentage of bootstrap support for the respective branches. For simplification, bootstrap values are indicated only for the most external branches, critical for interpreting the trees. In yellow, chondrichthyan sequences. When possible, the osteichthyan sequences have been collapsed into groups. (a) Optimal tree obtained from an alignment of the conserved N-terminal region of synucleins (alignment algorithm: MAFFT FF-TNS-2; model of sequence evolution: Q.plant+G4; pers parameter = 0.2; LnL = −2287.782). Four groups of gnathostome sequences are shown. (b) Tree generated using the entire sequence (alignment algorithm: MAFFT FF-TNS-2; Model of sequence evolution: Q.plant+G4; pers = 0.5; LnL = −4914.911). Gnathostome α-syn and β-syn sequences have been grouped together. The topology of the tree precludes doing the same for their γ-syn sequences.

Figure 3

Synteny analyses comparing the regions around the three synuclein-coding genes in humans and the shark Rhincodon typus. In each region, the arrows indicate either the human genes (top line of arrows) or the most similar human genes when the R. typus genes are compared (BLASTP searches) against the complete set of human proteins (bottom line). The names of the R. typus genes are indicated below both lines of arrows. The orientations of the arrows reflect those of the genes on their respective chromosomes. Colors have been added to highlight potential orthologies.

Figure 4

Maximum-likelihood tree obtained by comparing the N-terminal regions of synucleins from the 22 model species. Agnathan, chondrichthyan and osteichthyan proteins are indicated in magenta, orange and blue, respectively. Bootstrap values are indicated as in Figure 2; again, only the values for the critical branches are detailed. An arrow indicates an alternative root of the tree (see main text). Alignment algorithm used in the tree: MAFFT FF-TNS-2; model of sequence evolution: Q.plant+G4; pers parameter = 0.2; LnL = −2845.072.

Figure 5

Maximum-likelihood tree obtained for the synucleins of the twenty-two model species, when the full sequences are analyzed. Conventions are the same as in Figure 4. The arrow once again indicates the alternative root of the tree discussed in the main text. Alignment algorithm: MAFFT FF-TNS-2; model of sequence evolution: Q.plant+G4; pers parameter = 0.5; LnL = −5816.227.

Figure 6

Genes surrounding the synuclein-coding ones in Petromyzon marinus. The names within the arrows indicate the names of the lamprey genes. Below the arrows, both the names of the most likely human orthologs and their chromosomal positions are provided. These are the ones with the highest scores when the lamprey genes were compared (BLASTP) with the complete set of human proteins. Lamprey ASYN1-ASYN3 genes are indicated in red. Lamprey genes with potential orthologs on human chromosomes 4, 5 or 10 are highlighted in green.

Figure 7

Models explaining the events leading to a single human gene showing significant similarity when compared with a given agnathan gene. Red and blue boxes are used to identify the chromosomes; only one of the two homologous chromosomes is depicted. (a) A gene loss occurred after WGD1 but before the agnathan/gnathostome split, followed by another loss after WGD2 in gnathostomes. In both cases, the lamprey and human genes are necessarily orthologs. (b) Two losses occurred in the gnathostome lineage, one before and the second after WGD2. In this scenario, there is a 50% probability of the human gene being an ortholog and 50% of being a paralog of a given agnathan gene. (c) Three losses occurred in gnathostomes after WGD2. Again, there is a 50% probability of the remaining human gene being either an ortholog or a paralog of a given lamprey gene. This evolutionary history is expected to be less frequent than those in (a) or (b), which require just two gene losses.

Figure 8

Alignment of selected sequences and schematic representation of the eight repeats hypothesized in synuclein sequences. Repeats 4 and 5 would have been truncated by a 7-amino-acid-long deletion that occurred very early in synuclein evolution. The arrowhead indicates the end of the highly conserved region present in all synucleins. The trees shown in Figure 2a and Figure 4 were obtained with sequences truncated at that point.