Low-complexity regions in fungi display functional groups and are depleted in positively charged amino acids

Abstract

Reports on the diversity and occurrence of low-complexity regions (LCR) in Eukaryota are limited. Some studies have provided a more extensive characterization of LCR proteins in prokaryotes. There is a growing body of knowledge about a plethora of biological functions attributable to LCRs. However, it is hard to determine to what extent observed phenomena apply to fungi since most studies of fungal LCRs were limited to model yeasts. To fill this gap, we performed a survey of LCRs in proteins across all fungal tree of life branches. We show that the abundance of LCRs and the abundance of proteins with LCRs are positively correlated with proteome size. We observed that most LCRs are present in proteins with protein domains but do not overlap with the domain regions. LCRs are associated with many duplicated protein domains. The quantity of particular amino acids in LCRs deviates from the background frequency with a clear over-representation of amino acids with functional groups and a negative charge. Moreover, we discovered that each lineage of fungi favors distinct LCRs expansions. Early diverging fungal lineages differ in LCR abundance and composition pointing at a different evolutionary trajectory of each fungal group.

Figure 1.

LCRs are present in fungal proteins with and without Pfam domain. (A) The counts of proteins containing LCRs together with either protein domains, signal peptides, or transmembrane regions. Numbers provided denote protein counts in each category. (B) Distribution of the number of LCR per protein in proteins having at least one LCR.

Figure 2.

Total abundance of proteins with LCR varies among fungal phyla. (A) The abundance of LCR-containing proteins plotted against proteome size, colored by phylum with. (B) Overall abundance of proteins with LCRs per phylum. The taxonomic relationships between fungal lineages are derived from Spatafora et al. [40].

Figure 3.

LCR localization within proteins is non uniform. Normalized distribution of LCR localization within the protein length treated as 10 bins starting from N- to C-terminus within all proteins (A), and specifically for Neocallimastigomycota (B), Microsporidia (C), and Olpidium bornovanus (D).

Figure 4.

Simple LCRs composed predominantly of a single amino acid type are usually shorter comparing to the more complex LCRs (A) Distribution of lengths of homopolymers (dark gray) and complex LCR (light gray) as the number of amino acid residues. (B) For the relative localization of homopolymers within proteins, the length of individual protein is divided into 10 bins ranging from N- to C-terminus.

Figure 5.

The number of proteins containing homopolymeric LCRs correlates with the total number of LCRs, and varies among fungal phyla. (A) Overall abundance of homopolymer-containing proteins per phylum. The taxonomic relationships between fungal lineages are derived from Spatafora et al. [40]. (B) Number of proteins with complex LCRs compared to the number of proteins with homopolymers per phylum.

Figure 6.

LCRs usually do not overlap with the domain regions in fungal proteins. Percentage of LCR localized “within” and “outside” of protein domains, as well as in nondomain proteins, for both complex LCR and homorepeats in multiple fungal phyla.

Figure 7.

LCRs display a deviated amino acids composition compared to the UniProt. Average amino acid frequencies (as percentages) in LCRs, compared to the UniProt reference background.