Near the phylogenetic root of the Kingdom Fungi lies a branch that unites Microsporidia and a group variably known as Rozellomycota or Cryptomycota. This clade of intracellular parasites are considered energy thieves, because they import ATP from their hosts. The clade encompasses a great diversity of morphologies from the fungus-like Rozella spp. to the highly derived Microsporidia which are best known for the reductive nature of their cells (Hirt et al., 1997; Williams et al., 2002; Burri et al., 2006). The biology of Rozellomycota is unclear, being almost exclusively documented from diverse DNA sequences commonly found in environmental sequencing datasets from terrestrial and aquatic ecosystems (Lara et al., 2010; Jones et al., 2011; Livermore and Mattes, 2013; Lazarus and James, 2015). They appear to be related to but distinct from another lineage, Aphelidea, a group comprised largely of parasites of algae for which genome-scale data is lacking (Karpov et al., 2014). Formal descriptions of new species in Rozellomycota are very rare, and the hosts are diverse, ranging from Fungi, slime molds, amoebae, crustaceans, to algae (Karling, 1942; Kagami et al., 2007; Corsaro et al., 2014, 2016; Ishida et al., 2015). These newly described species possess some characteristics similar to those of Rozella and others that more closely resemble Microsporidia.

All Rozellomycota described to date are intracellular parasites that can produce a chitinous cell wall, yet grow as naked protoplasts within their hosts. However, they range greatly in morphology from species that are fungal-like (Rozella) and infect using a zoosporic (flagellated) stage to those without flagella where infections occur from spores through a polar filament (Corsaro et al., 2014). This latter structure and phylogenomic analyses unite Rozellomycota with the Microsporidia as the earliest diverging lineage of Fungi (James et al., 2013; Haag et al., 2014). The consideration of Rozellomycota and Microsporidia as members of the Fungi has been contentious due to the fact that they possess some, but not all, of the features characteristic of most Fungi (Richards et al., 2017). Some authors consider Rozellomycota to be outside of the Fungi (Cavalier-Smith, 2013; Karpov et al., 2014), because the genus Rozella and Aphelidea perform a form of phagocytosis of host cytoplasm making them appear to lack a defining feature of Fungi, osmoheterotrophy (Powell et al., 2017; Richards et al., 2017). Others have considered the clade as part of the Fungi, apparently accepting that the features of known species are sufficiently fungus-like (Hibbett et al., 2007; Berbee et al., 2017; Spatafora et al., 2017). Here, we show that all members of this clade are characterized by rampant gene loss, making it likely they are derived from a more fungus-like ancestor.

Microsporidians are represented by approximately 1700 species and are known to infect many animals, including humans (Keeling, 2009). They possess a mitochondrial-derived, genome-less organelle called a mitosome (Hirt et al., 1997; Williams et al., 2002; Burri et al., 2006), which has degenerated presumably because they steal ATP directly from their hosts (Tsaousis et al., 2008). Their nuclear genomes have also attracted considerable attention for being some of the smallest known for eukaryotes, reaching at the extreme only 2.3 Mb (Corradi et al., 2010).

To date, only one genome from the Rozellomycota lineage has been sequenced, that of Rozella allomycis, an obligate intracellular parasite of the water mold Allomyces (James et al., 2013). Phylogenetic analysis of the genome of R. allomycis strongly supported the placement of Rozella with Microsporidia and revealed previously unsuspected similarities with representatives of the phylum, as it harbored genes previously thought to be Microsporidia-specific. These genes included a nucleotide transporter essential for the acquisition of energy via ATP theft from the host and other genes with obvious signatures of horizontal gene transfers (James et al., 2013; Alexander et al., 2016). Other biochemical similarities included the presence of a small mitochondrial genome lacking electron transport Complex I and a seemingly degenerate sequence that is extremely AT–rich. The presence of the nucleotide transporter, the degeneracy or loss of the mitochondrion, and the fact that both Microsporidia and Rozella associate in the cell with the host’s mitochondria (Hacker et al., 2014; Powell et al., 2017), supported a hypothesis of direct uptake of ATP for nucleotide metabolism and energy from the host to parasite.

The Daphnia pathogen, Mitosporidium daphniae, is classified as an early diverging Microsporidia; however, it does not reside on the long branch with the rest of Microsporidia and genome drafts revealed that unlike those highly reduced parasites, it possesses a mitochondrial genome that lacks all genes involved in Complex I of oxidative phosphorylation (Haag et al., 2014), similar to R. allomycis. M. daphniae also lacks the ATP transporters that were suggested to facilitate mitochondrial degeneration in Rozella and Microsporidia. This pattern suggests that diverse strategies have evolved for parasitism that may be dependent on either host or the environment, as opposed to being lineage-specific. One common theme, however, is that the clade has evolved under repeated loss, going from a species with more genes than the free-living yeasts, for example Saccharomyces, which have approximately 5900 genes, to those with the smallest known eukaryotic genomes through the loss of function of genes involved in primary metabolism, presumably because the organism can acquire them from the host. The existence of a degenerate mitochondrial genome in R. allomycis and M. daphniae supported a model of step-wise degeneration of the respiratory chain in Rozellomycota (possibly following the acquisition of ATP transporters) that could have ultimately resulted in the emergence of genome-less mitosomes in Microsporidia (Corradi, 2015).

Paramicrosporidium saccamoebae (Corsaro et al., 2014) was described as an intranuclear parasite of Saccamoeba [Amoebozoa]. Morphologically, it somewhat resembles M. daphniae and Microsporidia, as it produces small-sized spores (approximately 1 µm) with a polar filament that lack observable mitochondria. However, ribosomal DNA phylogenies place P. saccamoebae within Rozellomycota, offering an opportunity to examine another member of Rozellomycota and ask questions about the evolution of this clade. Given the relationship between Rozellomycota and Microsporidia, and the fact that both lineages are known only from isolates which are obligate intracellular parasites, we sought to understand whether they are two distinct lineages with unifying changes such as gene loss and horizontal acquisition, or if the clade forms a gradient from fungus-like ancestors to highly reduced endoparasites through a series of nested reductions. Have reductions in gene content occurred many times independently throughout the evolution of these enigmatic, yet ubiquitous clades? Are there common strategies that Rozellomycota and Microsporidia share in drawing energy from their host? To address these questions, we sequenced the genome and transcriptome of P. saccamoebae. Our phylogenetic analyses and gene content comparisons revealed that this species is sister to Microsporidia (and not R. allomycis), and suggests that Rozellomycota is paraphyletic. Our genome investigations also uncovered a number of P. saccamoebae gene losses related to life inside the host nucleus and revealed that it shares more genes with distant lineages of Fungi and presumably therefore to the ancestor of all Fungi than with its closest sequenced relatives. Our study also highlights independent losses and reductions of mitochondrial functions in the evolution of Rozellomycota.

To acquire the nuclear and mitochondrial genome of P. saccamoebae, the microbial community of the Saccamoeba host was sequenced and assembled via metagenomic techniques (Figure 1). In addition, we sequenced RNA transcripts of P. saccamoebae in order to improve genome annotation and to verify expression of genes of interest. Our approaches produced a target nuclear genome assembly of 7.3 Mb in size on 221 scaffolds and possesses 3750 predicted genes (Table 1). This size is intermediate between R. allomycis (11.8 Mb) and M. daphniae (5.6 Mb), while the number of genes is similar to that of M. daphniae (3331 genes). The GC content is 47%, which is within the upper range of sequenced Rozellomycota and Microsporidia [34–47%] (Katinka et al., 2001; James et al., 2013). Analysis of the core eukaryotic genes (CEGs) indicates that the assembly is fairly complete with identification of 89% of complete CEGs and 91% complete or partial CEGs. As an additional measure of assembly continuity, we found that 50% of the genome resides on only 29 scaffolds. Over 76% of the genome is covered by gene space, while repeated content covers only 0.53% of the genome, most of which are simple or low complexity repeats.

Figure 1

Download asset Open asset

Phylogenetic analysis of a concatenated 53 protein data set, with 26,020 amino acid positions in the alignment, reconstructed a monophyletic clade that includes both Rozellomycota and Microsporidia (maximum likelihood bootstrap support [MLBP]=100), as sister to all the rest of Fungi (Figure 2). Within the clade, P. saccamoebae diverges within Rozellomycota after R. allomycis with strong support (MLBP = 100) but diverges before M. daphniae, which is still supported as the earliest diverging microsporidian species. With the current nomenclature, Rozellomycota is therefore reconstructed as paraphyletic. Hypothesis testing of alternative placements of P. saccamoebae and Microsporidia were all rejected, supporting a paraphyletic Rozellomycota (Figure 2—figure supplement 1), and reanalysis of the data following removal of Microsporidia species (excluding M. daphniae) also gave support for a monophyletic Kingdom Fungi (Figure 2—figure supplement 2), and for the branching order within Rozellomycota observed in analyses with Microsporidia included.

Figure 2 with 2 supplements see all

Download asset Open asset

We compared the predicted proteome of P. saccamoebae to R. allomycis, M. daphniae, 10 other Microsporidia, 19 other Fungi spanning the diversity of sequenced fungal lineages, and the outgroup, Fonticula alba (Figure 3A). We found P. saccamoebae shares the most of its 2470 orthogroups, 1975 (80%), with Spizellomyces punctatus (Figure 3B), a chytrid fungus which has been shown to share a few specific orthogroups with Microsporidia (Cuomo et al., 2012). Although a substantial number of P. saccamoebae orthogroups are also shared with R. allomycis (1,722; 70%) and M. daphniae (1,357;55%), it shares more with many other distantly related Fungi, including species in the derived fungal phyla of Ascomycota and Basidiomycota, than with these two closer relatives. The fewest P. saccamoebae orthogroups are shared with all other microsporidians sampled, 589 to 664 (24 to 27%). This pattern of orthology is similar to what is seen in both R. allomycis (Figure 3—figure supplement 1) and M. daphniae (Figure 3—figure supplement 2), whereby they share a higher percentage of their proteome with distant relatives than with one another. No orthogroups are universally present for Rozellomycota and microsporidians (Figure 3A), and very few clusters were specific to P. saccamoebae and either R. allomycis (12) or M. daphniae (13). There are 377 orthogroups that P. saccamoebae shares with other Fungi that are not found in R. allomycis, M. daphniae, or Microsporidia (Figure 3A). Of these, many have no hypothetical annotation, however, eight are related to oxidative phosphorylation functions in the mitochondrion, three are nuclear-pore-associated proteins, and two are proteins involved in converting ethanol to acetate (via alcohol and aldehyde dehydrogenases). Analysis of overall gene loss and gain revealed moderate loss leading to R. allomycis, followed by larger amounts of gene loss and lineage-specific gains in R. allomycis, P. saccamoebae, and M. daphniae (Figure 3C).

Figure 3 with 2 supplements see all

Download asset Open asset

Figure 3—source data 1

Orthologous cluster raw data for graphs (Figure 3B, Figure 3—figure supplements 1 and 2) and ancestral reconstruction (Figure 3C).

https://doi.org/10.7554/eLife.29594.010

Download elife-29594-fig3-data1-v1.xlsx

A circular mitochondrial genome that contains a standard complement of fungal mitochondrial genes was assembled onto a single contig of 25,401 bp with 41 predicted genes (Figure 4). All genes typically found in fungal mitochondrial genomes that are involved in oxidative phosphorylation are present including those of Complex I (nad1, nad2, nad3, nad4, nad4L, nad5, and nad6), Complex III (cob), Complex IV (cox1, cox2, and cox3), Complex V (atp6, atp8, and atp9), 23 tRNAs, one ribosomal protein (rps12), and the large and small rRNA subunits (rnL and rnS). This is in contrast to R. allomycis, M. daphniae, and the derived microsporidians, which have, respectively, lost all Complex I genes or the entire mitochondrial genome altogether (Figure 4—figure supplement 1). Like R. allomycis and M. daphniae, P. saccamoebae possesses rps12, not rps3 like most fungal mitochondrial genomes (Aguileta et al., 2014; Yang et al., 2017), indicating these proteins have been differentially retained between these two lineages (Figure 4—figure supplement 1). Due to the metagenomic nature of this sample, phylogenetic analysis was used to determine if the host (amoebozoan) mitochondrial genome had been inadvertently sequenced. Using the conserved mitochondrial genes (cob, cox1, cox2, cox3, and atp6), P. saccamoebae mitochondrion is placed as sister to R. allomycis and M. daphniae, and not with the available amoebozoan mitochondrial sequences (Figure 4—figure supplement 2). Moreover, supplementary genes involved in the electron transport chain (including ferrodoxin, ubiquinone, succinate dehydrogenase, etc.) are present in the nuclear genome of P. saccamoebae, in addition to being expressed, suggesting it possesses full oxidative phosphorylation functionality.

Figure 4 with 2 supplements see all

Download asset Open asset

To analyze host dependency and energy acquisition, we analyzed genes involved in nucleotide and amino acid biosynthesis and import. The genes necessary for de novo nucleotide (both purine and pyrimidine) biosynthesis are all absent in P. saccamoebae, suggesting it is not capable of producing its own nucleotides, and it also lacks genes for converting nucleosides to nucleotides and vice versa (Figure 5). One of the most highly expressed transporter genes is a nucleoside transporter (PSACC_00918), which has homologs in many other Fungi including M. daphniae, but which is not found in R. allomycis. Another nucleoside transporter (PSACC_02618) was also identified and is expressed. Other highly expressed transporters include an ABC transporter (PSACC_00512), an Na+ dicarboxylate and phosphate transporter (PSACC_03652) and a general substrate transporter (PSACC_01530). There is no evidence for the presence of the horizontally acquired thymidine kinases found by Alexander et al. (2016) in both R. allomycis and several microsporidians.

Figure 5

Download asset Open asset

Many amino acid biosynthesis pathways are reduced or absent in P. saccamoebae (Figure 6A), and as a result it is likely that it does not produce histidine or tryptophan, and may require intermediates from the host to complete synthesis of the following: lysine, methionine, arginine, phenylalanine, isoleucine, leucine, valine, and proline. For synthesis of branched chain amino acids (valine, leucine, and isoleucine), a single protein, a homolog of branched chain amino acid aminotransferase (PSACC_01141 in P. saccamoebae), is present and expressed based on our mRNA-seq data. Our investigations also revealed similar yet unique profiles of reductions in amino acid biosynthesis enzymes in R. allomycis and M. daphniae as well (Figure 6A), while the rest of Microsporidia cannot produce any amino acids. We identified two expressed amino acid permeases (PSACC_02864 and PSACC_03255) that may offset these gene losses in the P. saccamoebae genome, which are homologous to a single permease in both R. allomycis and M. daphniae and most closely related to bacterial, not fungal, amino acid permeases (Figure 6B). There are three fungal chitin synthase domain containing proteins in P. saccamoebae (PSACC_00558, PSACC_00865, PSACC_01336). These are all Division I chitin synthases with homologs in R. allomycis. Unlike R. allomycis, no Division II chitin synthases were identified, a characteristic P. saccamoebae shares with Microsporidia.

Figure 6

Download asset Open asset

P. saccamoebae has almost all of the conventional meiosis-related genes, including all known meiosis-specific genes (Rec8, Spo11-1, Dmc1, Hop2, Mnd1, Msh4, and Msh5; Figure 7A, Supplementary file 1). The only meiosis-related gene absent from the P. saccamoebae genome is a homolog of Mlh3 which is involved in DNA mismatch repair and appears to be absent in all other members of this earliest diverging fungal clade. The number of meiosis genes found in P. saccamoebae is greater than any other sequenced Rozellomycota or Microsporidia with available genomes. This indicates that this obligate intracellular parasite is theoretically able to undergo conventional meiotic recombination, which is bolstered by analysis of SNP frequencies that revealed that the vast majority of alleles possess a 50:50 ratio, indicating that P. saccamoebae is likely diploid, like most microsporidian species examined to date (Figure 7B) (Cuomo et al., 2012; Haag et al., 2013a, 2013b; Selman et al., 2013; Ndikumana et al., 2017; Pelin et al., 2016). Other P. saccamoebae genes potentially involved in mating include two putative homeodomains (HD1 and HD2 homologs) (PSACC_01945 and PSACC_01946) whose genome organization is strikingly similar to that found in loci that govern sexual identity in more derived Fungi, particularly Basidiomycota. However, heterozygosity and extensive allelic divergence typical of fungal MAT-loci is absent along those genes. Furthermore, their genome organization is not found to be conserved in other Rozellomycota and Microsporidia. Intriguingly, divergent copies of both HD1 and HD2 exist at other locations in the genome of the microsporidians Pseudoloma neurophilia and Nosema ceranae (Figure 7—figure supplement 1), but none of these show the organization or divergence expected for fungal-like MAT loci.

Figure 7 with 1 supplement see all

Download asset Open asset

To examine the evolution and history of the flagellum and its losses within Rozellomycota, a search was conducted for the dozens of genes involved in flagellum components and functioning. This revealed that unlike Rozella spp. and environmentally acquired Rozellomycota targeted using FISH (Jones et al., 2011), P. saccamoebae does not possess the ability to produce a flagellum (Supplementary file 2). Using hidden Markov models (HMMs) developed using alignments of known polar filament proteins (PFPs) in Microsporidia (Han and Weiss, 2017), we searched the predicted proteome of P. saccamoebae for sequence-based evidence of PFPs and found no significant hits (all conditional p>0.001). This likely reflects the rapid rate of evolution of these proteins, which are only known from species with a well-developed polar filament. Our HMMs similarly did not detect PFPs from M. daphniae, but detected all PFPs in the microsporidian Encephalitozoon romaleae.

The existence of a typical fungal mitochondrial genome, with a complete electron transport chain in P. saccamoebae is in stark contrast with the hypothesis of gradual reduction of oxidative respiration function within the Rozellomycota-Microsporidia clade eventually leading to the origin of mitosomes in Microsporidia (Tsaousis et al., 2008; James et al., 2013). Similar to what was found in M. daphniae, a mitochondrion was not identified using electron microscopy in the original species description of P. saccamoebae (Corsaro et al., 2014), indicating that organelles can be difficult to discern in micrographs of these very small (~1 µm) endoparasites. In fact, in both M. daphniae (Haag et al., 2014), and P. saccamoebae, the only evidence for a functional mitochondrion lies in the genomic data, yet interpretation of this data should be considered indisputable. However, the presence of all genes in Complex I of oxidative phosphorylation in P. saccamoebae, taken in concert with the phylogenetic results, indicates that the reductions in R. allomycis and M. daphniae mitochondrial genomes represent independent and convergent losses. Further evidence that P. saccamoebae possesses a functional mitochondrion derives from the mitochondrial genome phylogeny (Figure 4—figure supplement 2) and the presence and expression of genes for mitochondrial function (e.g. succinate dehydrogenase, NADH-ubiquinone oxidoreductase) in the nuclear genome.

It is unclear why the lifecycle and environment of P. saccamoebae has led to maintenance of a mitochondrion capable of all links of the electron transport chain, whereas this ability has been reduced (via loss of the same complex) in the related endoparasites R. allomycis and M. daphniae. The independent reductions in Complex I suggests that there may have been multiple transitions to intracellular parasitism and that putative ‘epiphytic’ Rozellomycota are or were once occupying non-parasitic ecological niches (Jones et al., 2011; Ishida et al., 2015). One explanation for the maintenance of a fully functioning mitochondrion in P. saccamoebae could be related to its subcellular location. Studies using light microscopy of R. allomycis (Powell et al., 2017) and Microsporidia (Hacker et al., 2014) have found that these parasites are surrounded by host mitochondria, presumably to facilitate energy theft from the host. Therefore, it could be that P. saccamoebae, which is sequestered to the nucleus, is prevented from co-locating with the host mitochondria from which it could steal energy. There are, however, Microsporidia that are intranuclear parasites of fish and crustaceans, for example Nucleospora (Freeman et al., 2013) and Enterospora (Stentiford et al., 2007), and they have no mitochondrial genome. R. allomycis and many Microsporidia possess horizontally acquired ATP transporters by which they steal ATP from their host, and these were not found in P. saccamoebae. Potentially, a lack of these (and the ability to steal energy) has led to the maintenance of Complex I genes in its mitochondrion, although M. daphniae (which has lost Complex I) does not possess these ATP transporters as well. Alternatively, the mitochondrion in P. saccamoebae could be maintained for use in a perhaps unobserved stage in its lifecycle (although our results have ruled out the possibility of a flagellated stage). This hypothesis is bolstered by the fact that a mitochondrion was not observed in the spores of P. saccamoebae via microscopy. In any case, additional genome sequences from this phylum will hopefully clarify how often mitochondrial genomes have been affected by reductive processes (including its complete loss in Microsporidia), and if these reductions have coincided with transitions to parasitism of particular hosts, acquisition of particular transporters, or in specific environmental conditions.

Our results suggest that Rozellomycota is paraphyletic. Microsporidia, which do form a monophyletic clade including the earliest diverging species, M. daphniae, are clearly nested within Rozellomycota (Figure 2, Figure 2—figure supplement 1). Based on the phylogenies published in this and previous (Haag et al., 2014) studies, however, M. daphniae does not reside on the long branch with the rest of Microsporidia, and there are several cellular (e.g. presence of mitochondrion) and genomic (e.g. ability to produce some amino acids) characteristics, which suggest this species is biologically more similar to Rozellomycota species than with the extremely reduced Microsporidia. Genomic analysis of additional Rozellomycota, such as Nucleophaga spp. (Corsaro et al., 2014; 2016) will be vital for understanding whether M. daphniae is in a monophyletic group with the Microsporidia or if, as rDNA trees suggest, it is in the Rozellomycota.

Results from the orthologous clustering analysis indicate that P. saccamoebae is more similar in gene content to distantly related Fungi, and presumably therefore to the ancestor of all Fungi, than it is to either R. allomycis or M. daphniae (Figure 3B). Shared gene content is clearly not correlated with evolutionary relationships within the Rozellomycota, and there are not obvious genomic synapomorphies that can be used to link Rozellomycota and Microsporidia together. This, in combination with the reconstruction of gains and losses within Rozellomycota (Figure 3C) and differential retention and loss of amino acid and nucleotide biosynthesis enzymes (Figures 5 and 6) provide evidence that this group of Fungi have been shaped by a series of independent and repeated losses, rather than nested losses toward more derived branches.

The loss of biosynthesis of nucleotides and multiple amino acids in P. saccamoebae indicates that like R. allomycis and Microsporidia (Dean et al., 2016), this species is highly dependent on its host for aspects of both protein synthesis and DNA replication. The highly expressed nucleoside transporter we identified likely facilitates import of nucleosides from the host amoeba, although the absence of any kinases to convert nucleosides to nucleotides is conspicuous. P. saccamoebae also lacks a homolog of the nucleotide transporters (identified in R. allomycis and Microsporidia) which were horizontally acquired from Chlamydiae (Tsaousis et al., 2008), but M. daphniae lacks these as well (Haag et al., 2014) suggesting these have either been missed in the genome assemblies, that this acquisition happened at least two independent times, or that the gene has been independently lost in both P. saccamoebae and M. daphniae after horizontal transfer into the ancestor of Rozellomycota. Within this context, the cellular location during parasitism (inside the nucleus) of P. saccamoebae may facilitate access to nucleotides from the host, and it may be that the mRNA of the host is phagocytized or translocated into the cell and then broken into essential components by the RNA degradation pathway. The same scenario is perhaps true of amino acids, that is, proteolysis of proteins following phagocytosis or translocation. Such a scenario would indicate an intermediate biology of these organisms which more closely resemble the eukaryotic ancestors from which they evolved, than the rest of the Fungi with chitinous-walled vegetative stages. The amino acid permeases in P. saccamoebae, R. allomycis, and M. daphniae, which are both absent in other Fungi and related to bacterial permeases (Figure 6B), suggest that these were present in the common ancestor of Rozellomycota. The permease gene has been duplicated in P. saccamoebae, possibly as a result of an increased necessity for amino acids from its amoeba host. In the gene-reduced budding yeast, Saccharomyces cerevisiae, amino acid biosynthesis pathways are all complete (Kanehisa et al., 2014), which taken in combination with the reduction in these in Rozellomycota and Microsporidia (which are even further reduced) suggests that intracellular parasitism may be driving this trend.

The lack of multiple genes involved in flagellum production in the P. saccamoebae genome was not unexpected given the microscopy-based observations of the life cycle, including spores possessing a polar filament. Thus, when or how many times the flagellum has been lost in this group of Fungi remains an open question in Rozellomycota evolution. Clearly, based on in situ hybridization studies and environmental DNA studies, disparate members of the phylum have zoospores with flagella (Jones et al., 2011), while others are more Microsporidia-like in their morphology without obvious flagellated stages and Microsporidia-sized spores (Corsaro et al., 2014; 2016). However, this study provides further support for the notion that where a polar filament is present, a flagellum is absent. In Rozella spp., the zoospores are the infective agents (Held, 1973), and therefore the advent of the infective polar filament may have rendered the flagellum unnecessary. Whether the polar filament has a single origin is uncertain, but likely given its complexity. That searches for PFPs failed likely reflects the rapid rate of sequence evolution (Slamovits et al., 2004) and lack of conserved gene synteny rather than an independent origin of the polar filament.

Knowledge of the mode of reproduction of Rozellomycota is essential to fully understand their transmission and life cycle. Sexual reproduction has not been observed in Rozellomycota, while rare microscopy-based evidence for meiosis was reported for two unrelated Microsporidia (Corradi, 2015). However, overwhelming genome data points toward the existence of cryptic sexual meiotic recombination in most microsporidians (Lee et al., 2014), and we uncovered identical signatures in the genome of P. saccamoebae. Although this species harbors many genomic signatures of sexual reproduction (i.e. a complete set of meiosis genes and significant evidence of diploidy, a prerequisite for meiosis), deciphering the reproductive strategies of this obligate intracellular parasite and allied species will likely require population-based genetic approaches. In particular, these seem necessary to expose strain-specific divergent loci potentially involved in compatibility, or reveal evidence of intraspecific gene exchange and recombination.

This high-quality assembly (as evidenced by the CEGMA results, presence of meiosis genes, mitochondrial genome assembly, etc.), especially compared to other sequenced species in the clade, demonstrates the power of sequencing metagenomic samples when pure samples are not readily available. Overall, the genome sequence of P. saccamoebae has provided crucial insights into the evolution and biology of this highly diverse but severely undersampled clade. In particular, this study revealed this intranuclear parasite has a mitochondrial genome possessing genes for all complexes involved in oxidative phosphorylation, and has a reduced capacity to produce nucleotides and amino acids, suggesting a dependence on the host for genetic building blocks and protein production but not energy. We also found evidence for independent reductions within the nuclear genomes of Rozellomycota, indicating there may have been multiple transitions to parasitism within the clade and that Microsporidia are a particularly highly derived and reduced lineage. Finally, these results provide further whole genome evidence for the paraphyly of Rozellomycota.

1. 1. Quandt CA

   (2015) KSL3_genome_submission

   Publicly available at the NCBI GenBank (accession no. MTSL0000000).

   https://www.ncbi.nlm.nih.gov/protein/MTSL0000000

1. 1. Aguileta G
   2. de Vienne DM
   3. Ross ON
   4. Hood ME
   5. Giraud T
   6. Petit E
   7. Gabaldón T

   (2014) High variability of mitochondrial gene order among fungi

   Genome Biology and Evolution 6:451–465.

   https://doi.org/10.1093/gbe/evu028

   • PubMed
   • Google Scholar
2. 1. Alexander WG
   2. Wisecaver JH
   3. Rokas A
   4. Hittinger CT

   (2016) Horizontally acquired genes in early-diverging pathogenic fungi enable the use of host nucleosides and nucleotides

   PNAS 113:4116–4121.

   https://doi.org/10.1073/pnas.1517242113

   • Google Scholar
3. Software

   1. Anderson A
   2. Itai S
   3. Jain S

   (2010) Scripts required to calculate tetramer frequencies and create input files for ESOM, version 22e0571

   Github.

   https://github.com/tetramerFreqs/Binning
4. 1. Beck N
   2. Lang BF

   (2010) MFannot, Organelle Genome Annotation Webserver

   MFannot, Organelle Genome Annotation Webserver, http://megasun.bch.umontreal.ca/cgi-bin/mfannot/mfannotInterface.pl.

   http://megasun.bch.umontreal.ca/cgi-bin/mfannot/mfannotInterface.pl

   • Google Scholar
5. 1. Berbee ML
   2. James TY
   3. Strullu-Derrien C

   (2017) Early diverging fungi: diversity and impact at the dawn of terrestrial life

   Annual Review of Microbiology 71:41–60.

   https://doi.org/10.1146/annurev-micro-030117-020324

   • PubMed
   • Google Scholar
6. 1. Boeckmann B
   2. Bairoch A
   3. Apweiler R
   4. Blatter MC
   5. Estreicher A
   6. Gasteiger E
   7. Martin MJ
   8. Michoud K
   9. O'Donovan C
   10. Phan I
   11. Pilbout S
   12. Schneider M

   (2003) The SWISS-PROT protein knowledgebase and its supplement TrEMBL in 2003

   Nucleic Acids Research 31:365–370.

   https://doi.org/10.1093/nar/gkg095

   • PubMed
   • Google Scholar
7. 1. Burri L
   2. Williams BA
   3. Bursac D
   4. Lithgow T
   5. Keeling PJ

   (2006) Microsporidian mitosomes retain elements of the general mitochondrial targeting system

   PNAS 103:15916–15920.

   https://doi.org/10.1073/pnas.0604109103

   • PubMed
   • Google Scholar
8. 1. Cantarel BL
   2. Korf I
   3. Robb SM
   4. Parra G
   5. Ross E
   6. Moore B
   7. Holt C
   8. Sánchez Alvarado A
   9. Yandell M

   (2008) MAKER: an easy-to-use annotation pipeline designed for emerging model organism genomes

   Genome Research 18:188–196.

   https://doi.org/10.1101/gr.6743907

   • PubMed
   • Google Scholar
9. 1. Capella-Gutiérrez S
   2. Marcet-Houben M
   3. Gabaldón T

   (2012) Phylogenomics supports microsporidia as the earliest diverging clade of sequenced fungi

   BMC Biology 10:47.

   https://doi.org/10.1186/1741-7007-10-47

   • PubMed
   • Google Scholar
10. 1. Capella-Gutiérrez S
    2. Silla-Martínez JM
    3. Gabaldón T

    (2009) trimAl: a tool for automated alignment trimming in large-scale phylogenetic analyses

    Bioinformatics 25:1972–1973.

    https://doi.org/10.1093/bioinformatics/btp348

    • PubMed
    • Google Scholar
11. 1. Cavalier-Smith T

    (2013) Early evolution of eukaryote feeding modes, cell structural diversity, and classification of the protozoan phyla loukozoa, sulcozoa, and choanozoa

    European Journal of Protistology 49:115–178.

    https://doi.org/10.1016/j.ejop.2012.06.001

    • PubMed
    • Google Scholar
12. 1. Corradi N
    2. Pombert JF
    3. Farinelli L
    4. Didier ES
    5. Keeling PJ

    (2010) The complete sequence of the smallest known nuclear genome from the microsporidian Encephalitozoon intestinalis

    Nature Communications 1:1–7.

    https://doi.org/10.1038/ncomms1082

    • PubMed
    • Google Scholar
13. 1. Corradi N

    (2015) Microsporidia: eukaryotic intracellular parasites shaped by gene loss and horizontal gene transfers

    Annual Review of Microbiology 69:167–183.

    https://doi.org/10.1146/annurev-micro-091014-104136

    • PubMed
    • Google Scholar
14. 1. Corsaro D
    2. Michel R
    3. Walochnik J
    4. Venditti D
    5. Müller KD
    6. Hauröder B
    7. Wylezich C

    (2016) Molecular identification of Nucleophaga terricolae sp. nov. (Rozellomycota), and new insights on the origin of the Microsporidia

    Parasitology Research 115:3003–3011.

    https://doi.org/10.1007/s00436-016-5055-9

    • PubMed
    • Google Scholar
15. 1. Corsaro D
    2. Walochnik J
    3. Venditti D
    4. Steinmann J
    5. Müller KD
    6. Michel R

    (2014) Microsporidia-like parasites of amoebae belong to the early fungal lineage rozellomycota

    Parasitology Research 113:1909–1918.

    https://doi.org/10.1007/s00436-014-3838-4

    • PubMed
    • Google Scholar
16. 1. Cuomo CA
    2. Desjardins CA
    3. Bakowski MA
    4. Goldberg J
    5. Ma AT
    6. Becnel JJ
    7. Didier ES
    8. Fan L
    9. Heiman DI
    10. Levin JZ
    11. Young S
    12. Zeng Q
    13. Troemel ER

    (2012) Microsporidian genome analysis reveals evolutionary strategies for obligate intracellular growth

    Genome Research 22:2478–2488.

    https://doi.org/10.1101/gr.142802.112

    • PubMed
    • Google Scholar
17. 1. Dean P
    2. Hirt RP
    3. Embley TM

    (2016) Microsporidia: why make nucleotides if you can steal them?

    PLOS Pathogens 12:e1005870.

    https://doi.org/10.1371/journal.ppat.1005870

    • PubMed
    • Google Scholar
18. 1. Dick GJ
    2. Andersson AF
    3. Baker BJ
    4. Simmons SL
    5. Thomas BC
    6. Yelton AP
    7. Banfield JF

    (2009) Community-wide analysis of microbial genome sequence signatures

    Genome Biology 10:R85.

    https://doi.org/10.1186/gb-2009-10-8-r85

    • PubMed
    • Google Scholar
19. 1. Eddy SR

    (2011) Accelerated Profile HMM Searches

    PLoS Computational Biology 7:e1002195.

    https://doi.org/10.1371/journal.pcbi.1002195

    • PubMed
    • Google Scholar
20. 1. Edgar RC

    (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput

    Nucleic Acids Research 32:1792–1797.

    https://doi.org/10.1093/nar/gkh340

    • PubMed
    • Google Scholar
21. 1. Freeman MA
    2. Kasper JM
    3. Kristmundsson Á

    (2013) Nucleospora cyclopteri n. sp., an intranuclear microsporidian infecting wild lumpfish, cyclopterus lumpus l., in icelandic waters

    Parasites & Vectors 6:49.

    https://doi.org/10.1186/1756-3305-6-49

    • PubMed
    • Google Scholar
22. Preprint

    1. Garrison E
    2. Marth G

    (2012) Haplotype-based variant detection from short-read sequencing

    arXiv.

    https://arxiv.org/abs/1207.3907v2

    • Google Scholar
23. 1. Garrison E

    (2012) vcflib: variant call file processing and manipulation utilities

    vcflib: variant call file processing and manipulation utilities, https://github.com/vcflib/vcflib#vcflib.

    https://github.com/vcflib/vcflib#vcflib

    • Google Scholar
24. 1. Gordon A

    (2011) FASTQ/A short-reads pre-processing tools

    FASTQ/A short-reads pre-processing tools, http://hannonlab.cshl.edu/fastx_toolkit/.

    http://hannonlab.cshl.edu/fastx_toolkit/

    • Google Scholar
25. 1. Haag KL
    2. James TY
    3. Pombert JF
    4. Larsson R
    5. Schaer TM
    6. Refardt D
    7. Ebert D

    (2014) Evolution of a morphological novelty occurred before genome compaction in a lineage of extreme parasites

    PNAS 111:15480–15485.

    https://doi.org/10.1073/pnas.1410442111

    • PubMed
    • Google Scholar
26. 1. Haag KL
    2. Sheikh-Jabbari E
    3. Ben-Ami F
    4. Ebert D

    (2013a) Microsatellite and single-nucleotide polymorphisms indicate recurrent transitions to asexuality in a microsporidian parasite

    Journal of Evolutionary Biology 26:1117–1128.

    https://doi.org/10.1111/jeb.12125

    • PubMed
    • Google Scholar
27. 1. Haag KL
    2. Traunecker E
    3. Ebert D

    (2013b) Single-nucleotide polymorphisms of two closely related microsporidian parasites suggest a clonal population expansion after the last glaciation

    Molecular Ecology 22:314–326.

    https://doi.org/10.1111/mec.12126

    • PubMed
    • Google Scholar
28. 1. Haas BJ
    2. Papanicolaou A
    3. Yassour M
    4. Grabherr M
    5. Blood PD
    6. Bowden J
    7. Couger MB
    8. Eccles D
    9. Li B
    10. Lieber M
    11. MacManes MD
    12. Ott M
    13. Orvis J
    14. Pochet N
    15. Strozzi F
    16. Weeks N
    17. Westerman R
    18. William T
    19. Dewey CN
    20. Henschel R
    21. LeDuc RD
    22. Friedman N
    23. Regev A

    (2013) De novo transcript sequence reconstruction from RNA-seq using the Trinity platform for reference generation and analysis

    Nature Protocols 8:1494–1512.

    https://doi.org/10.1038/nprot.2013.084

    • PubMed
    • Google Scholar
29. 1. Hacker C
    2. Howell M
    3. Bhella D
    4. Lucocq J

    (2014) Strategies for maximizing ATP supply in the microsporidian Encephalitozoon cuniculi: direct binding of mitochondria to the parasitophorous vacuole and clustering of the mitochondrial porin VDAC

    Cellular Microbiology 16:565–579.

    https://doi.org/10.1111/cmi.12240

    • PubMed
    • Google Scholar
30. 1. Halary S
    2. Malik SB
    3. Lildhar L
    4. Slamovits CH
    5. Hijri M
    6. Corradi N

    (2011) Conserved meiotic machinery in Glomus spp., a putatively ancient asexual fungal lineage

    Genome Biology and Evolution 3:950–958.

    https://doi.org/10.1093/gbe/evr089

    • PubMed
    • Google Scholar
31. 1. Han B
    2. Weiss LM

    (2017)

    Microsporidia: obligate intracellular pathogens within the fungal kingdom

    Microbiology Spectrum 5:1–17.

    • Google Scholar
32. 1. Held AA

    (1973) Encystment and germination of the parasitic chytrid Rozella allomycis on host hyphae

    Canadian Journal of Botany 51:1825–1835.

    https://doi.org/10.1139/b73-234

    • Google Scholar
33. 1. Hibbett DS
    2. Binder M
    3. Bischoff JF
    4. Blackwell M
    5. Cannon PF
    6. Eriksson OE
    7. Huhndorf S
    8. James T
    9. Kirk PM
    10. Lücking R
    11. Thorsten Lumbsch H
    12. Lutzoni F
    13. Matheny PB
    14. McLaughlin DJ
    15. Powell MJ
    16. Redhead S
    17. Schoch CL
    18. Spatafora JW
    19. Stalpers JA
    20. Vilgalys R
    21. Aime MC
    22. Aptroot A
    23. Bauer R
    24. Begerow D
    25. Benny GL
    26. Castlebury LA
    27. Crous PW
    28. Dai YC
    29. Gams W
    30. Geiser DM
    31. Griffith GW
    32. Gueidan C
    33. Hawksworth DL
    34. Hestmark G
    35. Hosaka K
    36. Humber RA
    37. Hyde KD
    38. Ironside JE
    39. Kõljalg U
    40. Kurtzman CP
    41. Larsson KH
    42. Lichtwardt R
    43. Longcore J
    44. Miadlikowska J
    45. Miller A
    46. Moncalvo JM
    47. Mozley-Standridge S
    48. Oberwinkler F
    49. Parmasto E
    50. Reeb V
    51. Rogers JD
    52. Roux C
    53. Ryvarden L
    54. Sampaio JP
    55. Schüssler A
    56. Sugiyama J
    57. Thorn RG
    58. Tibell L
    59. Untereiner WA
    60. Walker C
    61. Wang Z
    62. Weir A
    63. Weiss M
    64. White MM
    65. Winka K
    66. Yao YJ
    67. Zhang N

    (2007) A higher-level phylogenetic classification of the Fungi

    Mycological Research 111:509–547.

    https://doi.org/10.1016/j.mycres.2007.03.004

    • PubMed
    • Google Scholar
34. 1. Hirt RP
    2. Healy B
    3. Vossbrinck CR
    4. Canning EU
    5. Embley TM

    (1997) A mitochondrial Hsp70 orthologue in vairimorpha necatrix: molecular evidence that microsporidia once contained mitochondria

    Current Biology 7:995–998.

    https://doi.org/10.1016/S0960-9822(06)00420-9

    • PubMed
    • Google Scholar
35. 1. Ishida S
    2. Nozaki D
    3. Grossart HP
    4. Kagami M

    (2015) Novel basal, fungal lineages from freshwater phytoplankton and lake samples

    Environmental Microbiology Reports 7:435–441.

    https://doi.org/10.1111/1758-2229.12268

    • PubMed
    • Google Scholar
36. 1. James TY
    2. Pelin A
    3. Bonen L
    4. Ahrendt S
    5. Sain D
    6. Corradi N
    7. Stajich JE

    (2013) Shared signatures of parasitism and phylogenomics unite cryptomycota and microsporidia

    Current Biology 23:1548–1553.

    https://doi.org/10.1016/j.cub.2013.06.057

    • PubMed
    • Google Scholar
37. 1. Jones MD
    2. Forn I
    3. Gadelha C
    4. Egan MJ
    5. Bass D
    6. Massana R
    7. Richards TA
    8. a RT

    (2011) Discovery of novel intermediate forms redefines the fungal tree of life

    Nature 474:200–203.

    https://doi.org/10.1038/nature09984

    • PubMed
    • Google Scholar
38. 1. Kagami M
    2. de Bruin A
    3. Ibelings BW
    4. Van Donk E
    5. De BA
    6. Van DE

    (2007) Parasitic chytrids: their effects on phytoplankton communities and food-web dynamics

    Hydrobiologia 578:113–129.

    https://doi.org/10.1007/s10750-006-0438-z

    • Google Scholar
39. 1. Kanehisa M
    2. Goto S
    3. Sato Y
    4. Kawashima M
    5. Furumichi M
    6. Tanabe M

    (2014) Data, information, knowledge and principle: back to metabolism in KEGG

    Nucleic Acids Research 42:D199–D205.

    https://doi.org/10.1093/nar/gkt1076

    • PubMed
    • Google Scholar
40. 1. Karling JS

    (1942) Parasitism among the chytrids

    American Journal of Botany 29:24–35.

    https://doi.org/10.2307/2436540

    • Google Scholar
41. 1. Karpov SA
    2. Mamkaeva MA
    3. Aleoshin VV
    4. Nassonova E
    5. Lilje O
    6. Gleason FH

    (2014) Morphology, phylogeny, and ecology of the aphelids (aphelidea, opisthokonta) and proposal for the new superphylum Opisthosporidia

    Frontiers in Microbiology 5:1–11.

    https://doi.org/10.3389/fmicb.2014.00112

    • PubMed
    • Google Scholar
42. 1. Katinka MD
    2. Duprat S
    3. Cornillot E
    4. Méténier G
    5. Thomarat F
    6. Prensier G
    7. Barbe V
    8. Peyretaillade E
    9. Brottier P
    10. Wincker P
    11. Delbac F
    12. El Alaoui H
    13. Peyret P
    14. Saurin W
    15. Gouy M
    16. Weissenbach J
    17. Vivarès CP

    (2001) Genome sequence and gene compaction of the eukaryote parasite Encephalitozoon cuniculi

    Nature 414:450–453.

    https://doi.org/10.1038/35106579

    • PubMed
    • Google Scholar
43. 1. Keeling P

    (2009) Five questions about microsporidia

    PLoS Pathogens 5:e1000489.

    https://doi.org/10.1371/journal.ppat.1000489

    • PubMed
    • Google Scholar
44. 1. Krueger F

    (2015) A wrapper tool around Cutadapt and FastQC to consistently apply quality and adapter trimming to FastQ files

    A wrapper tool around Cutadapt and FastQC to consistently apply quality and adapter trimming to FastQ files, https://github.com/FelixKrueger/TrimGalore.

    https://github.com/FelixKrueger/TrimGalore

    • Google Scholar
45. 1. Langmead B
    2. Salzberg SL

    (2012) Fast gapped-read alignment with Bowtie 2

    Nature Methods 9:357–359.

    https://doi.org/10.1038/nmeth.1923

    • PubMed
    • Google Scholar
46. 1. Lara E
    2. Moreira D
    3. López-García P

    (2010) The environmental clade LKM11 and rozella form the deepest branching clade of fungi

    Protist 161:116–121.

    https://doi.org/10.1016/j.protis.2009.06.005

    • PubMed
    • Google Scholar
47. 1. Lazarus KL
    2. James TY

    (2015) Surveying the biodiversity of the cryptomycota using a targeted PCR approach

    Fungal Ecology 14:62–70.

    https://doi.org/10.1016/j.funeco.2014.11.004

    • Google Scholar
48. 1. Lee SC
    2. Heitman J
    3. Ironside JE

    (2014)

    Microsporidia Pathog Oppor

    231–243, Sex and the Microsporidia, Microsporidia Pathog Oppor.

    • Google Scholar
49. 1. Li H
    2. Durbin R

    (2009) Fast and accurate short read alignment with Burrows-Wheeler transform

    Bioinformatics 25:1754–1760.

    https://doi.org/10.1093/bioinformatics/btp324

    • PubMed
    • Google Scholar
50. 1. Li H
    2. Handsaker B
    3. Wysoker A
    4. Fennell T
    5. Ruan J
    6. Homer N
    7. Marth G
    8. Abecasis G
    9. Durbin R
    10. 1000 Genome Project Data Processing Subgroup

    (2009) The sequence alignment/map format and SAMtools

    Bioinformatics 25:2078–2079.

    https://doi.org/10.1093/bioinformatics/btp352

    • PubMed
    • Google Scholar
51. 1. Li L
    2. Stoeckert CJ
    3. Roos DS

    (2003) OrthoMCL: identification of ortholog groups for eukaryotic genomes

    Genome Research 13:2178–2189.

    https://doi.org/10.1101/gr.1224503

    • PubMed
    • Google Scholar
52. 1. Livermore JA
    2. Mattes TE

    (2013) Phylogenetic detection of novel cryptomycota in an Iowa (United States) aquifer and from previously collected marine and freshwater targeted high-throughput sequencing sets

    Environmental Microbiology 15:2333–2341.

    https://doi.org/10.1111/1462-2920.12106

    • PubMed
    • Google Scholar
53. 1. Lohse M
    2. Drechsel O
    3. Bock R

    (2007) OrganellarGenomeDRAW (OGDRAW): a tool for the easy generation of high-quality custom graphical maps of plastid and mitochondrial genomes

    Current Genetics 52:267–274.

    https://doi.org/10.1007/s00294-007-0161-y

    • PubMed
    • Google Scholar
54. 1. Moriya Y
    2. Itoh M
    3. Okuda S
    4. Yoshizawa AC
    5. Kanehisa M

    (2007) KAAS: an automatic genome annotation and pathway reconstruction server

    Nucleic Acids Research 35:W182–W185.

    https://doi.org/10.1093/nar/gkm321

    • PubMed
    • Google Scholar
55. 1. Namiki T
    2. Hachiya T
    3. Tanaka H
    4. Sakakibara Y

    (2012) MetaVelvet: an extension of Velvet assembler to de novo metagenome assembly from short sequence reads

    Nucleic Acids Research 40:e155.

    https://doi.org/10.1093/nar/gks678

    • PubMed
    • Google Scholar
56. 1. Ndikumana S
    2. Pelin A
    3. Williot A
    4. Sanders JL
    5. Kent M
    6. Corradi N

    (2017) Genome analysis of pseudoloma neurophilia: a microsporidian parasite of zebrafish (danio rerio)

    The Journal of Eukaryotic Microbiology 64:18–30.

    https://doi.org/10.1111/jeu.12331

    • PubMed
    • Google Scholar
57. 1. Paradis E
    2. Claude J
    3. Strimmer K

    (2004) APE: analyses of phylogenetics and evolution in r language

    Bioinformatics 20:289–290.

    https://doi.org/10.1093/bioinformatics/btg412

    • PubMed
    • Google Scholar
58. 1. Parra G
    2. Bradnam K
    3. Korf I

    (2007) CEGMA: a pipeline to accurately annotate core genes in eukaryotic genomes

    Bioinformatics 23:1061–1067.

    https://doi.org/10.1093/bioinformatics/btm071

    • PubMed
    • Google Scholar
59. 1. Pelin A
    2. Moteshareie H
    3. Sak B
    4. Selman M
    5. Naor A
    6. Eyahpaise MÈ
    7. Farinelli L
    8. Golshani A
    9. Kvac M
    10. Corradi N

    (2016) The genome of an Encephalitozoon cuniculi type III strain reveals insights into the genetic diversity and mode of reproduction of a ubiquitous vertebrate pathogen

    Heredity 116:458–465.

    https://doi.org/10.1038/hdy.2016.4

    • PubMed
    • Google Scholar
60. 1. Powell MJ
    2. Letcher PM
    3. James TY
    4. Beakes GW

    (2017) Ultrastructural characterization of the host-parasite interface between Allomyces anomalus (Blastocladiomycota) and Rozella allomycis (Cryptomycota)

    Fungal Biology 121:561–572.

    https://doi.org/10.1016/j.funbio.2017.03.002

    • PubMed
    • Google Scholar
61. 1. Price AL
    2. Jones NC
    3. Pevzner PA

    (2005) De novo identification of repeat families in large genomes

    Bioinformatics 21:i351–i358.

    https://doi.org/10.1093/bioinformatics/bti1018

    • PubMed
    • Google Scholar
62. 1. Quinlan AR

    (2014)

    BEDTools: The Swiss-Army Tool for Genome Feature Analysis

    11.12.1–11.1211, BEDTools: The Swiss-Army Tool for Genome Feature Analysis.

    • Google Scholar
63. 1. Richards TA
    2. Leonard G
    3. Wideman JG

    (2017) What Defines the "Kingdom" Fungi?

    Microbiology Spectrum 5:.

    https://doi.org/10.1128/microbiolspec.FUNK-0044-2017

    • PubMed
    • Google Scholar
64. 1. Selman M
    2. Sak B
    3. Kváč M
    4. Farinelli L
    5. Weiss LM
    6. Corradi N

    (2013) Extremely reduced levels of heterozygosity in the vertebrate pathogen Encephalitozoon cuniculi

    Eukaryotic Cell 12:496–502.

    https://doi.org/10.1128/EC.00307-12

    • PubMed
    • Google Scholar
65. 1. Shimodaira H
    2. Hasegawa M

    (2001) CONSEL: for assessing the confidence of phylogenetic tree selection

    Bioinformatics 17:1246–1247.

    https://doi.org/10.1093/bioinformatics/17.12.1246

    • PubMed
    • Google Scholar
66. 1. Slamovits CH
    2. Fast NM
    3. Law JS
    4. Keeling PJ

    (2004) Genome compaction and stability in microsporidian intracellular parasites

    Current Biology 14:891–896.

    https://doi.org/10.1016/j.cub.2004.04.041

    • PubMed
    • Google Scholar
67. 1. Smit AFA
    2. Hubley R
    3. Green P

    (2014) RepeatMasker

    RepeatMasker, Open 4.0, http://www.repeatmasker.org.

    http://www.repeatmasker.org

    • Google Scholar
68. 1. Spatafora JW
    2. Aime MC
    3. Grigoriev IV
    4. Martin F
    5. Stajich JE
    6. Blackwell M

    (2017) The fungal tree of life: From molecular systematics to genome-scale phylogenies

    Microbiology Spectrum 5:.

    https://doi.org/10.1128/microbiolspec.FUNK-0053-2016

    • PubMed
    • Google Scholar
69. 1. Stamatakis A

    (2014) RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies

    Bioinformatics 30:1312–1313.

    https://doi.org/10.1093/bioinformatics/btu033

    • PubMed
    • Google Scholar
70. 1. Stanke M
    2. Morgenstern B

    (2005) AUGUSTUS: a web server for gene prediction in eukaryotes that allows user-defined constraints

    Nucleic Acids Research 33:W465–W467.

    https://doi.org/10.1093/nar/gki458

    • PubMed
    • Google Scholar
71. 1. Stentiford GD
    2. Bateman KS
    3. Longshaw M
    4. Feist SW

    (2007) Enterospora canceri n. gen., n. sp., intranuclear within the hepatopancreatocytes of the european edible crab Cancer pagurus

    Diseases of Aquatic Organisms 75:61–72.

    https://doi.org/10.3354/dao075061

    • PubMed
    • Google Scholar
72. Software

    1. R Core Development Team

    (2013)

    R: A Language and Environment for Statistical ComputingR Foundation for Statistical Computing

    R: A Language and Environment for Statistical ComputingR Foundation for Statistical Computing, Vienna, Austria.
73. 1. Tsaousis AD
    2. Kunji ER
    3. Goldberg AV
    4. Lucocq JM
    5. Hirt RP
    6. Embley TM

    (2008) A novel route for ATP acquisition by the remnant mitochondria of Encephalitozoon cuniculi

    Nature 453:553–556.

    https://doi.org/10.1038/nature06903

    • PubMed
    • Google Scholar
74. Book

    1. Ultsch A
    2. Mörchen F

    (2005)

    ESOM-Maps: Tools for Clustering, Visualization, and Classification with Emergent SOM

    Tech Rep Dept Math Comput Sci Univ Marburg.

    • Google Scholar
75. 1. Wattam AR
    2. Abraham D
    3. Dalay O
    4. Disz TL
    5. Driscoll T
    6. Gabbard JL
    7. Gillespie JJ
    8. Gough R
    9. Hix D
    10. Kenyon R
    11. Machi D
    12. Mao C
    13. Nordberg EK
    14. Olson R
    15. Overbeek R
    16. Pusch GD
    17. Shukla M
    18. Schulman J
    19. Stevens RL
    20. Sullivan DE
    21. Vonstein V
    22. Warren A
    23. Will R
    24. Wilson MJ
    25. Yoo HS
    26. Zhang C
    27. Zhang Y
    28. Sobral BW

    (2014) PATRIC, the bacterial bioinformatics database and analysis resource

    Nucleic Acids Research 42:D581–D591.

    https://doi.org/10.1093/nar/gkt1099

    • PubMed
    • Google Scholar
76. 1. Williams BA
    2. Hirt RP
    3. Lucocq JM
    4. Embley TM

    (2002) A mitochondrial remnant in the microsporidian trachipleistophora hominis

    Nature 418:865–869.

    https://doi.org/10.1038/nature00949

    • PubMed
    • Google Scholar
77. 1. Yang J
    2. Harding T
    3. Kamikawa R
    4. Simpson AGB
    5. Roger AJ

    (2017) Mitochondrial genome evolution and a novel rna editing system in deep-branching heteroloboseids

    Genome Biology and Evolution 9:1161–1174.

    https://doi.org/10.1093/gbe/evx086

    • PubMed
    • Google Scholar
78. 1. Zerbino DR
    2. Birney E

    (2008) Velvet: algorithms for de novo short read assembly using de Bruijn graphs

    Genome Research 18:821–829.

    https://doi.org/10.1101/gr.074492.107

    • PubMed
    • Google Scholar

Thank you for submitting your article "The genome of an intranuclear parasite unveils alternative adaptations to obligate intracellular parasitism" for consideration by eLife. Your article has been favorably evaluated by Diethard Tautz (Senior Editor) and three reviewers, one of whom is a member of our Board of Reviewing Editors. The following individual involved in review of your submission has agreed to reveal his identity: Jeremy Wideman (Reviewer #2).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

The study by Quandt and collaborators describes the genome of the intracellular parasite Paramicrosporidium saccamoebae in the early fungal lineage Rozellomycota. The authors show that the P. saccamoebae genome is not as reduced (in terms of gene content) as the genomes of previously sequenced Rozellomycota or of previously sequenced Microsporidia and also contains a full mitochondrial genome (again in contrast to previously sequenced genomes from this part of the tree of life). Combined with the placement of P. saccamoebae on the fungal phylogeny (it appears to have diverged after the divergence of Rozella, another Rozellomycota genome), analysis of the gene content of this species suggests that the path toward adaptation to a intracellular lifestyle was not linear, but likely involved multiple rounds of gene losses and perhaps multiple transitions to parasitism.

Essential revisions:

1) Many microsporidia localize to the nucleus (e.g. Nucleospora and Enterospora see: PMC3606367 and pubmed id 17523545). This means that mitochondrial function does not correlate with nuclear localization. Therefore one of the major hypotheses that the authors put forth, namely that cellular localization correlates with retention of mitochondrial function is refuted. Furthermore, some misconceptions about mtDNAs seem to be held by the authors. Firstly, a "complete and fully functional mitochondrial genome" is a meaningless statement as many mitochondrial genomes contain very few genes but are fully functional (for example, dinoflagellates, and many apicomplexa encode very few mitochondrial proteins but still have aerobic fully functional mitochondria). Also, many mitochondrial genomes across the eukaryotes lack Complex I (including S. cerevisiae) and numerous other aerobic/facultative species (e.g. some algae like Chlamydomonas see Kamikawa et al. (PMID 26833505). Thus, convergent simplification of the mitochondrial genome is extremely common, and is in fact the major mode of mitochondrial evolution. However, this does not mean that there is not an interesting mitochondrial story in these data. Specifically, even though the reduction of genes may not be that interesting, it may point towards a different hypothesis, namely that "the cryptomycotan ancestor was a parasite with alternating lifecycle stages between an aerobic flagellate and an anaerobic intracellular non-flagellate". If mitochondria (with intact cristae) are not easily visible in the lifecycle stages that are being viewed, perhaps the mitochondrion is not functioning aerobically at this time. This would support a hypothesis that the intracellular phase is anaerobic. Perhaps the motile stage for Cryptomycota like Rozella, or infection/germination stage for Cryptomycota like Paramicrosporidium, requires oxidative phosphorylation for some reason. This would be a very attractive model for how ETC components (or the flagellum) could be readily (and convergently) lost. The authors are advised to rewrite the mitochondrial section(s) in a way that provides more information on comparisons between other holomycotan taxa and a reformulation (or omission) of the correlative hypothesis.

2) The authors consider the Rozellomycota belongs to the Fungi. This is not correct as 1) they are phagotrophic (not osmotrophic), 2) they do not have the chitinous cell wall at the vegetative stage. Otherwise, if they are included in the Fungi the diagnosis of the latter has to be cardinally changed, and in this case the other holomycots (amoebae Nuclearia, Parvularia, Fonticula) must be included in the Fungi. The authors are advised to discuss this point in the Introduction.

3) The overall presentation of the manuscript was below par; many of the figures were unclear, several terms were incorrectly used, and the manuscript contained several typos –. The authors are advised to carefully examine and address all of them and to carefully proof the revised manuscript for both science and style before resubmission.

[Editors' note: further revisions were requested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "The genome of an intranuclear parasite, Paramicrosporidium saccamoebae, reveals alternative adaptations to obligate intracellular parasitism" for further consideration at eLife. Your article has been favorably evaluated by Diethard Tautz (Senior Editor) and three reviewers, one of whom is a member of our Board of Reviewing Editors.

The manuscript has been improved but there are a few remaining issues raised by reviewer #2 that i would like you to consider before acceptance:

Reviewer #1:

The authors have satisfactorily addressed my comments.

Reviewer #2:

In general I am happy with the changes in the revision. This was a pleasurable review process.

The only thing that I think must change is the claim that fungi evolved from an amoebozoan ancestor.

e.g. the fifth paragraph of the Discussion is troublingly inaccurate. Fungi did not evolve from an amoebozoan ancestor… Amoebozoans form a very different group. Fonticula and nuclearia are not amoebozoans (they are technically amoeba, but this is an extremely derived state – almost as derived as fungi or microsporidia). Fonticula and Nuclearia are sister to fungi which means they shared a common ancestor, but the common ancestor of these two groups is likely to be a simple phagotrophic flagellate (just like the ancestor to every major eukaryotic group) perhaps with an amoeboid stage, but perhaps not.

Reviewer #3:

The manuscript has been seriously rewritten and looks now more complete. Concerning my worries about the incorrect inclusion of opisthosporidians in the Fungi: the reply of the authors doesn't suit me well. At the same time, the addition in the manuscript Introduction is correct. I understand that this question is still discussable and certainly does not affect the matter of the paper – I'd accept this part of manuscript as ready for publication.

Essential revisions:

1) Many microsporidia localize to the nucleus (e.g. Nucleospora and Enterospora see: PMC3606367 and pubmed id 17523545). This means that mitochondrial function does not correlate with nuclear localization. Therefore one of the major hypotheses that the authors put forth, namely that cellular localization correlates with retention of mitochondrial function is refuted. Furthermore, some misconceptions about mtDNAs seem to be held by the authors. Firstly, a "complete and fully functional mitochondrial genome" is a meaningless statement as many mitochondrial genomes contain very few genes but are fully functional (for example, dinoflagellates, and many apicomplexa encode very few mitochondrial proteins but still have aerobic fully functional mitochondria). Also, many mitochondrial genomes across the eukaryotes lack Complex I (including S. cerevisiae) and numerous other aerobic/facultative species (e.g. some algae like Chlamydomonas see Kamikawa et al. (PMID 26833505). Thus, convergent simplification of the mitochondrial genome is extremely common, and is in fact the major mode of mitochondrial evolution. However, this does not mean that there is not an interesting mitochondrial story in these data. Specifically, even though the reduction of genes may not be that interesting, it may point towards a different hypothesis, namely that "the cryptomycotan ancestor was a parasite with alternating lifecycle stages between an aerobic flagellate and an anaerobic intracellular non-flagellate". If mitochondria (with intact cristae) are not easily visible in the lifecycle stages that are being viewed, perhaps the mitochondrion is not functioning aerobically at this time. This would support a hypothesis that the intracellular phase is anaerobic. Perhaps the motile stage for Cryptomycota like Rozella, or infection/germination stage for Cryptomycota like Paramicrosporidium, requires oxidative phosphorylation for some reason. This would be a very attractive model for how ETC components (or the flagellum) could be readily (and convergently) lost. The authors are advised to rewrite the mitochondrial section(s) in a way that provides more information on comparisons between other holomycotan taxa and a reformulation (or omission) of the correlative hypothesis.

We thank the reviewers for raising these interesting points. We agree with a number of them, while holding slightly different opinions about others. Firstly, our statement of "complete and fully functional mitochondrial genome" is indeed an inaccurate characterization, and the manuscript has now been revised to address this point.

The loss of genes involved in Complex I leads a reduction in the primary mitochondrial function of generating energy in the form of ATP through oxidative phosphorylation. The point that this loss is common throughout eukaryotes and clearly not a barrier to aerobic growth in these organisms is well taken, and a point we observed when sequencing the first Rozellomycota genome (James et al. 2013). However, we developed the hypothesis, or rather expanded on the existing hypothesis (Tsaousis et al. 2008), that mitochondrial loss of function was related to acquisition of ATP transporters from Chlamydia by horizontal transfer. We have recently demonstrated that Rozella and Microsporidia both share the traits of genomes encoding ATP transporters and intimate association of host mitochondrial with the parasite membrane (Powell et al. 2017). This led to the hypothesis which we presented in this paper that the separation of Paramicrosporidium from the mitochondria by the host nuclear membrane would lower access to host ATP generated via oxidative phosphorylation and lead to differences in numerous aspects of energy metabolism. Indeed, the primary result of the mitochondrial data we obtained for Paramicrosporidium indeed does not show this loss of OXPHOS function, and that is the point that we are trying to make about the mitochondrion. Prior to our study, both Rozella and Mitosporidium were found to have mitochondrial genomes that were both reduced (compared to typical fungal mito genome and with Nuclearia) in that they lack all of Complex I of oxidative phosphorylation, and our hypothesis was that this loss corresponded to an ancient loss in the early evolution of this clade and potentially an ancient switch to intracellular parasitism. However, that is not what we found. We found all genes typically found in fungal oxidative phosphorylation, which suggests independent losses in both Rozella and Mitosporidium, and potentially suggests there have multiple transitions to a parasitic lifestyle within the clade. To further explain this gene composition to the reader, we have added Supplementary file 2 where we show genes present in target taxa, other exemplar fungi, and Nuclearia. Flagella use a lot of ATP, and yet Rozella spp. (the only Rozellomycota species possessing flagella) lack Complex I. In the paper, we clearly note that we searched for genes typically present in species with a flagellum, and Paramicrosporidium is missing these, providing further support that we have not “missed” a flagellated stage in its lifecycle. Several co-authors (including the species original descriptors) have spent a considerable amount of time looking at these organisms. We are aware that some Microsporidia are intranuclear parasites. However, we disagree that our hypothesis for the biology and evolution of Paramicrosporidium is necessarily invalidated by this. Specifically, the ancestor to “core Microsporidia” (those on the extremely long branch) lost their mitochondrial genomes in what appears to be a lineage-wide event. This would suggest that the ancestors to Paramicrosporidium (which had a mitochondrion capable of oxid. Phos.) and Nucleospora and Enterospora (which did not have a mitochondrion capable of oxid. Phos, but may have had ATP transporters by which to steal ATP) have evolved to exist in similar habitats with very different original molecular “toolkits”. Thus, there is no reason to say it is impossible that these parasites in the same niche cannot have evolved different mechanisms by which to tolerate the inter-nuclear environment.

We have removed the language in the Abstract hypothesizing that sub-cellular location of the parasite may have increased the selection pressure for a functioning mitochondrion. We have left this hypothesis in the Discussion, but have also added a caveat about intranuclear microsporidians Enterospora and Nucleospora along with a note that these species (and R. allomycis) likely have transporters by which to steal ATP from their hosts, while P. saccamoebae does not. We have also added an alternative hypothesis as suggested by the reviewer that the mitochondrion could function in some other unseen part of the P. saccamoebae lifecycle – although not a flagellated stage, which our results refute.

2) The authors consider the Rozellomycota belongs to the Fungi. This is not correct as 1) they are phagotrophic (not osmotrophic), 2) they do not have the chitinous cell wall at the vegetative stage. Otherwise, if they are included in the Fungi the diagnosis of the latter has to be cardinally changed, and in this case the other holomycots (amoebae Nuclearia, Parvularia, Fonticula) must be included in the Fungi. The authors are advised to discuss this point in the Introduction.

We thank the reviewers for raising these interesting points. This is an active area of current debate, namely whether or not to include the Rozellomycota/Microsporidia in the Fungi or to consider them as sister groups. Presently, there is no simple resolution as this is a debate fueled by a taxonomic passion of certain authors, including many on the author list (James and Berbee 2012; Karpov et al. 2014). On the whole, the mycological community has accepted that both Microsporidia and Rozella/Cryptomycota may be included in the Fungi (NCBI, Stajich et al.; Berbee et al. 2017) and this can also be witnessed in the myriad of fungal ecology studies which are uncovering Rozellomycota at a fast pace. Ironically, the biological community has readily accepted the Microsporidia as Fungi (Adl et al. 2012), but the finding of phagotrophy-like nutrition in Rozella has been an impediment to their full acceptance as Fungi. The discovery of Cryptomycota/Rozellomycota created a controversy through the discovery of taxa that seemed somewhat like Fungi and somewhat like protistan ancestors because they possessed symplesiomorphic traits not observed in other fungi (flagellation, possible phagotrophic nutrition, and lack of cell wall) (Jones et al. 2011). Further analysis shows that ultimately Rozella has a number of traits that align them with fungi, such as chitinous cell wall in both spore and cyst phase, polarized cell growth using myosin linked chitin synthases (which likely undergo growth through osmoheterotrophy), and development of chytrid like spores in a zoosporangium. The results we report here show that loss was widespread and radically changed the genomes of all extant Rozellomycota. Very likely the common ancestor to Fungi+Rozellomycota+Microsporidia possessed a set of traits that were lost. Ultimately if the definition of Fungi needs to be slightly amended so be it; there is no synapomorphy for the Fungi (Richards et al. 2017). If we can accept that corals with zooxanthellae are still animals, then we can accommodate Rozellomycota into Fungi. After all, the majority of members in the lineage do not undergo any form of phagocytosis, and even Rozella phagocytosis is very distinct from that of nucleariids phagocytosing whole cells with cell walls and digesting them.

To address these points, we have added a short discussion of this to the Introduction. Our intention here, though, is not one of where to draw nomenclatural lines outside of the Rozellomycota+Microsporidia clade. Our main systematics point is that more and more evidence (including ours presented) is mounting that Microsporidia are nested within Rozellomycota (Cryptomycota), rendering Rozellomycota a paraphyletic group. We contend that most of the mycological community (to which most of the authors belong) refer to both Rozella, Rozellomycota, and Microsporidia as fungi (James and Berbee 2012; Han and Weiss 2017; Spatafora et al. 2017; Stajich 2017), and therefore we refer to them as such in the paper. Finally, we have recently suggested that the simplest definition of the Fungi is the sister group to the Holozoa (Berbee et al. 2017). A definition that requires fungi to possess osmotrophic nutrition could exclude some early-diverging taxa in the fungal clade while failing to exclude osmotrophic members of the Holozoa. Ultimately the higher classification arguments are interesting fodder for discussion, but that lies beyond the scope of this manuscript, and more data on Aphelids are needed before we know how they are related to the other groups (Karpov et al. 2014).

3) The overall presentation of the manuscript was below par; many of the figures were unclear, several terms were incorrectly used, and the manuscript contained several typos. The authors are advised to carefully examine and address all of them and to carefully proof the revised manuscript for both science and style before resubmission.

We have made changes to figures, legends, and the text where appropriate to improve readability.

[Editors' note: further revisions were requested prior to acceptance, as described below.]

Reviewer #2:

In general I am happy with the changes in the revision. This was a pleasurable review process.

The only thing that I think must change is the claim that fungi evolved from an amoebozoan ancestor.

e.g. the fifth paragraph of the Discussion is troublingly inaccurate. Fungi did not evolve from an amoebozoan ancestor… Amoebozoans form a very different group. Fonticula and nuclearia are not amoebozoans (they are technically amoeba, but this is an extremely derived state – almost as derived as fungi or microsporidia). Fonticula and Nuclearia are sister to fungi which means they shared a common ancestor, but the common ancestor of these two groups is likely to be a simple phagotrophic flagellate (just like the ancestor to every major eukaryotic group) perhaps with an amoeboid stage, but perhaps not.

Reviewer #2 made the following comment: “The only thing that I think must change is the claim that fungi evolved from an amoebozoan ancestor.” We have made changes in the text to reflect this, and sincerely regret the choice of wording here. The text now reads, “… the eukaryotic ancestors from which they evolved[…]”.

Author details

1. C Alisha Quandt

   Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-0260-8995

2. Denis Beaudet

   Department of Biology, University of Ottawa, Ottawa, Canada

   Contribution

   Data curation, Formal analysis, Validation, Investigation, Writing—original draft, Writing—review and editing

   Competing interests

   No competing interests declared

3. Daniele Corsaro

   CHLAREAS Chlamydia Research Association, Nancy, France

   Contribution

   Conceptualization, Investigation, Visualization, Methodology, Writing—review and editing

   Competing interests

   No competing interests declared

4. Julia Walochnik

   Molecular Parasitology, Institute for Specific Prophylaxis and Tropical Medicine, Medical University of Vienna, Koblenz, Germany

   Contribution

   Visualization, Methodology, Writing—review and editing

   Competing interests

   No competing interests declared

5. Rolf Michel

   Laboratory of Electron Microscopy, Central Institute of the Federal Armed Forces Medical Services, Koblenz, Germany

   Contribution

   Resources, Methodology, Writing—review and editing

   Competing interests

   No competing interests declared

6. Nicolas Corradi

   Department of Biology, University of Ottawa, Ottawa, Canada

   Contribution

   Conceptualization, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Writing—original draft, Writing—review and editing

   For correspondence

   ncorradi@uottawa.ca

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-7932-7932

7. Timothy Y James

   Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, United States

   Contribution

   Conceptualization, Resources, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Methodology, Writing—original draft, Writing—review and editing

   For correspondence

   tyjames@umich.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-1123-5986

• 2,451

  Page views

• 327

  Downloads

• 50

  Citations

Article citation count generated by polling the highest count across the following sources: Scopus, Crossref, PubMed Central.