Polyploidy is widespread in Microsporidia

Abstract

Microsporidia are obligate intracellular eukaryotic parasites with an extremely broad host range. They have both economic and public health importance. Ploidy in microsporidia is variable, with a few species formally identified as diploid and one as polyploid. Given the increase in the number of studies sequencing microsporidian genomes, it is now possible to assess ploidy levels across all currently explored microsporidian diversity. We estimate ploidy for all microsporidian data sets available on the Sequence Read Archive using k-mer-based analyses, indicating that polyploidy is widespread in Microsporidia and that ploidy change is dynamic in the group. Using genome-wide heterozygosity estimates, we also show that polyploid microsporidian genomes are relatively homozygous, and we discuss the implications of these findings on the timing of polyploidization events and their origin.IMPORTANCEMicrosporidia are single-celled intracellular parasites, distantly related to fungi, that can infect a broad range of hosts, from humans all the way to protozoans. Exploiting the wealth of microsporidian genomic data available, we use k-mer-based analyses to assess ploidy status across the group. Understanding a genome's ploidy is crucial in order to assemble it effectively and may also be relevant for better understanding a parasite's behavior and life cycle. We show that tetraploidy is present in at least six species in Microsporidia and that these polyploidization events are likely to have occurred independently. We discuss why these findings may be paradoxical, given that Microsporidia, like other intracellular parasites, have extremely small, reduced genomes.

Keywords: Microsporidia; k-mer; k-mer spectra; polyploidy; sequencing data; tetraploidy.

Fig 1

Examples of SRA data sets that were not analyzed further. (A) GenomeScope2-transformed linear plot of Vairimorpha ceranae (SRR7178080) as an example of a contaminated sample where the pattern coverage peaks are distorted by host contamination, and a good model fit is not achieved. (B) GenomeScope2-transformed linear plot of Encephalitozoon cuniculi (SRR122315) as an example of a low coverage sample. (C) GenomeScope2-transformed linear plot of Ordospora colligata (SRR18286429) as an example of a sample that is dominated by host data, showing the predicted genome size to be ~150 Mb.

Fig 2

Samples with Smudgeplot patterns that were difficult to interpret. (A) Vairimorpha ceranae (SRR18590836) Smudgeplot with a smeared tetraploid pattern between 4n and 3n coverage. (B) Vairimorpha ceranae (SRR18590837) Smudgeplot with a smeared octoploid pattern between 6n and 8n coverage. These patterns remain difficult to interpret using current data.

Fig 3

Ploidy estimates for Microsporidia species. (A and B) GenomeScope2 and Smudgeplot results for Astathelohania contejeani (SRR8476226) estimating tetraploidy. (C and D) GenomeScope2 and Smudgeplot results for Nematocida ausubeli (SRR350188) estimating diploidy. (E and F) GenomeScope2 and Smudgeplot results for Encephalitozoon intestinalis (SRR24007516) estimating diploidy with exceptionally high homozygosity.

Fig 4

A cladogram of microsporidian species with confident ploidy estimates, rooted on Rozella allomycis [diploid, (66)]. The cladogram was inferred from ribosomal small subunit sequences using IQ-TREE (version 2.2.2.3), with a GTR + F + I + R6 nucleotide substitution model.

Fig 5

Phylogeny of Microsporidia, rooted on Rozella allomycis [diploid, (66)]. The phylogeny was inferred from ribosomal small subunit sequences using IQ-TREE (version 2.2.2.3), with a GTR + F + I + R6 nucleotide substitution model. Nodes with bootstrap values <95% are indicated with gray circles.

Fig 6

Most diploid and tetraploid genomes are highly homozygous, with the frequency of AAAB k-mers greater than or equal to the frequency of AABB in nearly all cases. (A) Boxplot of mean heterozygosity across the genome per species, calculated using GenomeScope2 heterozygosity estimates (1% of AA for diploid species and 1% of AAAA for tetraploid species). (B) Line graph showing mean AAAB and AABB heterozygous k-mer patterns across the genome for each of the six tetraploid species. The shaded region represents the value range observed in the independent samples of each species.