Deep conservation of prion-like composition in the eukaryotic prion-former Pub1/Tia1 family and its relatives

PeerJ. 2020 Apr 17;8:e9023. doi: 10.7717/peerj.9023. eCollection 2020.


Pub1 protein is an important RNA-binding protein functional in stress granule assembly in budding yeast Saccharomyces cerevisiae and, as its co-ortholog Tia1, in humans. It is unique among proteins in evidencing prion-like aggregation in both its yeast and human forms. Previously, we noted that Pub1/Tia1 was the only protein linked to human disease that has prion-like character and and has demonstrated such aggregation in both species. Thus, we were motivated to probe further into the evolution of the Pub1/Tia1 family (and its close relative Nam8 and its orthologs) to gain a picture of how such a protein has evolved over deep evolutionary time since the last common ancestor of eukaryotes. Here, we discover that the prion-like composition of this protein family is deeply conserved across eukaryotes, as is the prion-like composition of its close relative Nam8/Ngr1. A sizeable minority of protein orthologs have multiple prion-like domains within their sequences (6-20% depending on criteria). The number of RNA-binding RRM domains is conserved at three copies over >86% of the Pub1 family (>71% of the Nam8 family), but proteins with just one or two RRM domains occur frequently in some clades, indicating that these are not due to annotation errors. Overall, our results indicate that a basic scaffold comprising three RNA-binding domains and at least one prion-like region has been largely conserved since the last common ancestor of eukaryotes, providing further evidence that prion-like aggregation may be a very ancient and conserved phenomenon for certain specific proteins.

Keywords: Compositional bias; Evolution; Human; Intrinsic disorder; Nam8; Prion; Pub1; RNA-binding; Tia1; Yeast.

Grant support

This research was funded completely by the Natural Sciences Engineering Research Council of Canada, except that some of the calculations were performed on a desktop computer funded by both the Canada Foundation for Innovation and the Natural Sciences Engineering Research Council of Canada. Computing resources supplied by Compute Canada were also used. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.