Australia: The Land Where Time Began

A biography of the Australian continent 

Early Fungi Evolved Digesting Cell Walls of Algal Ancestors of Land Plants Indicated by Phylogenomic Analyses

Fungi, which are decomposers, are key organisms in the recycling of plant material in global carbon cycles. It was hypothesised by Chang et al. that genomes of early diverging fungi may have inherited from an ancestral species pectinases that had been able to extract nutrients from land pants that contained pectin and their algal allies (Streptophytes). This study was aimed at inferring the geological timing of the plant-fungus association, based on pectinase gene expansions and on the organismal phylogeny. In this study 40 fungal genomes were studied, 3 of which, including Gonapodya prolifera, were sequenced. Rozella diverged first from all other fungi, according to the organismal phylogeny from 136 housekeeping loci. Gonapodya was included among the flagellated fungal species that were predominantly aquatic in the Chytridiomycota. A sister group to the Chytridiomycota were the zygomycota I and zygomycota II, that were predominantly terrestrial fungi, as well as the ascomycetes and basidiomycetes that comprise Dikarya. There are 27 genes in the genome of Gonapodya, which represented 5 of the 7 classes of pectin-specific enzymes that are known from fungi. A common ancestry is shared by most of these with pectinases from Dikarya. Like many Dikarya Gonapodya can use pectin as a carbon source for growth in a pure culture, which indicates functional and sequence similarity. Evidence that even ancient aquatic fungi had adapted to extract nutrients from the plants in the green lineage is provided by shared pectinases of Dikarya and Gonapodya. It is implied by this that 750 Ma (million years ago) the estimated maximum age of origin of the Streptophytes, which contained pectin, represents a maximum age for the divergence Chytridiomycota from the lineage that included the Dikarya.

The majority of modern fungi are characterised by close associations with plants or plant products. Fungi in Dikarya (Ascomycota and Basidiomycota) have key roles as decomposers (Floudas et al., 2012). Contrasting with the Dikarya, the lineages that diverged early in the evolution of the fungi, such as the zoosporic chytrids and some of the zygomycetes, use diverse sources of nutrition (Stajich et al., 2009). Many are associated with animals or animal products, though there are some that are associated with plants. Among early lineages the diversity of nutritional modes raises the questions of what the ancestral food of fungi might have been, and when the origin of the close and ecologically vital association between plants and fungi may have occurred (James et al., 2006; Sprockett et al., 2011).

The questions about the ancestral nutrition of fungi were approached by Chang et al. through analysis of the fungal digestive enzyme genes, which break down plant material. Studies of fungal degradation of plant material have traditionally been focused on the breakdown of cellulose and lignin. However, cellulose is distributed too widely across organisms that include green, red and brown algae (Popper et al., 2011) for cellulases to be used as good markers for the association with the lineage of land plants. The earliest land plants had not lignin in their structure (Popper et al., 2011); therefore enzymes for degrading lignin may not have been present in the ancestral fungi to break down plant material (Floudas et al., 2012). On the other hand pectins are polysaccharides that are present only in the cell walls of the streptophyte algae (Sørensen et al., 2011; Wickett et al., 2014). Pectin-specific polysaccharides and the genes for their synthesis, more specifically, have only been identified in the land plants and in “advanced” streptophyte algae though not in the streptophyte algal species that diverged early (Sorensen et al., 2011; Mikkelsen et al., 2014). Enzymes that are present in fungi that degrade multiple pectic molecules are potentially good indicators of the association between fungi and the land plant lineage.

In contrast with lignin and cellulose, pectins are water-soluble and form a jelly-like matrix in the primary cell walls of plants. Pectins are complex polymers often with poly- or oligosaccharide (e.g., aribogalactan, b-1, 4 galactan) side chains decorating backbones of homogalactuonan, xylogalacturonan, rhamnogalacturonan, rhamnogalacturonan I, and rhamnogalacturonan II. Multiple enzymes from different families are required of glycoside hydrolases (GHs), polysaccharide lyases (PL), and carbohydrate esterases (CE) (van den Brink & de Vries, 2011; Benoit et al., 2012). There are at least 9 families among the pectinases that are pectin specific, including 3 GH families (GH28, GH53 & GH93), 4 polysaccharide lyase families (PL1, PL3, PL4 and PL11), and 2 carbohydrate esterase families CE8 and CE13). An important role is played by enzymes from GH28 in the degradation of pectin backbones by fungi (van der Brink & de Vries, 2011). The other nonpectin-specific families, which may also breakdown other molecules include (GH2, GH35, GH43, GH51, GH54, GH78, GH88, GH105, PL9, CE12, and CE1. In this study, Chang et al. first surveyed the distribution of families of pectinase genes across fungi and outgroups. They then reconstructed pectinase phylogenies for a detailed picture of the diversification of fungi during through the evolution of major fungal lineages.

The goal of Chang et al. was to reconcile periods when there was expansion of pectinase genes with the fungal organismal phylogeny in order to identify times when ancestral fungi were adapting to using plants for acquiring nutrients. Chang et al. predicted that if the ancestral fungal species were associated with land plants and the charophytes that contained pectin the ancestral fungal species would have contained at least some pectinases. The pectinases may not have diversified until more recent cladogenesis and the rise of the Dikarya, if the early fungi fed on animal or animal-like material, and only later became associated with plants.

Chang et al. needed a robust fungal phylogeny in order to infer patterns of pectinase evolution among fungi. The branching order for the earliest splits among fungi has remained controversial, in spite of recent efforts at resolution (James et al., 2006; Liu et al., 2009; Ebersberger et al., 2012).

A large number of fungal genomes have been sequenced and annotated in recent years, which led to phylogenies that had increasingly rich sampling of loci and taxa (Ma et al., 2009; James et al., 2013; Suga et al., 2013; Tisserant et al., 2013). In this study, Chang et al., initiated a whole genome sequencing of Gonapodya prolifera, Coemansia reversa and Conidiobolus coronatus, in collaborating with Joint Genome Institute. They sampled whole genome data from 127 fungi, as well as 13 species from other eukaryotic groups, with an emphasis on the taxa that represented early splits in fungi. In order to explore factors that contributed to phylogenetic conflicts and to phylogenetic resolution, they applied multiple analytic methods to individual and concatenated genes.

With improved genomic and taxonomic sampling of fungi and related taxa, and with the study of     fungal pectinases, the aim of this study was to:

1)   Improve understanding of higher level fungal phylogeny,

2)   Infer the origin and diversification pattern of pectinases among fungi, and

3)   Infer the phylogenetic and geological age of plant/fungus associations, based on patterns of pectinase gene expansion.

Resolving the early splits in fungal evolution

Chang et al. have contributed to a picture that is well resolved of the divergence of early fungi with new genome sequences of 3 fungi that diverged early. The results are congruent with other studies that used whole genome sequences for taxa they have in common. The early evolution of fungi took place among zoosporic lineages that were predominantly aquatic, that are represented by the extant genera Cryptomycota, Chytridiomycota and Blastocladiomycota (Ebersberger et al., 2012; Torruella et al., 2012; James et al., 2013). The arbuscular mycorrhizal Rhizophagus, among the terrestrial fungi, which represents Glomeromycota, is a member of the zygomycota I clade. This is consistent with the Tisserant et al. (2013) analysis of complete genome sequences, though it contradicts earlier studies that were based partly or largely on ribosomal gene sequences showing Glomeromycota to be a sister group of Dikarya.

According to Chang et al. phylogenomics has proven to be useful in resolving some controversial phylogenetic problems. Potential pitfalls remain, however, even with genome-wide data (Fernández et al., 2014). In order to test for one of these pitfalls conflict among individual genes was looked for using the FcLM tests and AU tests. Low information content in individual genes was revealed by FcLM tests and AU analyses, instead of finding strong conflicting signals.

Equivocal support for competing topologies at the nodes of interest was shown by almost all individual genes. The conflict among earlier studies may be explained by the lack of information content per gene (James et al., 2006: Sekimoto et al., 2011; Ebersberger et al., 2012). It is suggested by this that further resolution of the deeper divergence will result from additional sequencing of genomes of taxa that were diverging early. The nature of these early splits may be maybe the reason for the lack of information content per gene. In fungal evolution the earliest splits occurred much further back in time than 500 Ma (Taylor & Barbee, 2006; Stajich et al., 2009; Parfree et al., 2011). Any evolutionary signal may have eroded over time, and therefore became difficult to recover. During early fungal evolution radiation events may have been too rapid to allow enough substitutions to accumulate for phylogenetic resolution (Sekimoto et al., 2011; Ebersberger et al., 2012). It was suggested by the results of the Bayesian analysis of Chang et al. that a short internode, if not a real polytomy, is present between the first 2 splits in the fungal tree though not at subsequent nodes. The lack of data for resolving these relationships is also demonstrated by high fixation points or lack of fixation for the earliest splits in RADICAL analysis. Difficulty in resolving whether Dikarya is the sisters group to zygomycota I, as well as the relationship between Rhizophagus and Mortierella is similarly reflected in failure to reach fixation in RADICAL analyses. It was shown by the experiments of Chang et al. with alternative data matrices that the branching order of species that were evolutionarily isolated of Rozella, Allomyces, and Catenaria was influenced by gene sequences that were missing from 1 or more of these taxa. It is not clear whether the loci that are missing from such taxa, nevertheless, improve overall inferences about their relationships (Lemmon et al., 2009; Wiens & Tiu, 2012). Relatively few sites that are reliably aligned were available even for genes that were present, which made it difficult to associated substitutions with the short branches of those lineages that were diverging early, as a result of the broad taxonomic sampling in this study.

Evidence of pectinase gene expansion for geological age divergence of fungi

An ancestral species of fungi that was taking advantage of nutrients from streptophyte plants that contained pectin, was found by the analysis of Chang et al. to be pointing to the common ancestor of Chytridiomycota and the terrestrial fungi that underwent at least 9 duplications of pectinase genes. There are 4 or more duplications that mapped to the ancestor of Pezizomycotina (Ascomycota), that are the only other concentration of pectinase duplications that are inferred on internal fungal branches. There are 3 fungus-specific clades in the GH28 gene tree that contain members of both Chytridiomycota and Dikarya that received 89% or better bootstrap support, is supported by their interpretation as clades of orthologues. There are many extant chytrids that are associated with streptophyte plants (Sparrow, 1960). Chang et al. provide evidence of shared pectinase function, beyond homology and the genes and proteins, by showing that the chytrids G. prolifera can use the purified pectins rhamnogalacturonan I and polygalacturonic acid as sources of carbon. The fossil record for the plant lineage is rich compared with that of the fungi. Chang et al. provide the first estimate of a maximum age for the fungi by linking the evolution of the fungi to Streptophytes. It is estimated that the streptophytes that contain pectin are no older than 750 Ma (Douzery et al., 2004; Zimmer et al., 2007; Parfrey et al., 2011), and therefore the common ancestor of Chytridiomycota and Dikarya that degraded pectin is no older than 750 Ma. The minimum age of origin of fungi and the major terrestrial fungal groups has been placed           at more than 1 Ga (billion years ago) by past studies that had no maximum age restraint (Heckman et al., 2001; Hedges et al., 2004; Padovan et al., 2005; Parfree et al., 2011). The very old ages for fungi are inconsistent with the notion that the diversification of plants and terrestrial fungi were interdependent on each other, as well a being inconsistent with the age of plants that contain pectin (Lücking et al., 2009).

Plants that are homologous to pectinases predate the origin of pectin, as is clear from the gene trees. It is suggested by reconstructions that at least 5 copies of the GH28 family were already present in the common ancestor of plants, Chromalveolata, and fungi. There are copies from 6 families in the oomycete Phytophthora, and the high copy numbers in each family trace back to recent gene duplications following its divergence from Saprolegnia. The distribution of pectinases across disparate eukaryotes could have resulted from horizontal transfer from plants or fungi to other organisms, as has been suggested by other genes for enzymes that were secreted (Richards et al., 2011). In restructuring the gene tree it could also reflect phylogenetic error (though the methods used by Chang et al. were conservative), or it could be a reflection of diversity of ancestral genes with other specificities that were co-opted for the breakdown of pectin.

Pectinases that evolve rapidly and have been repeatedly lost from fungi that switched from nutrition based on plants. All or nearly all of the pectinases were lost by yeasts (Saccharomyces and Shizosaccharomyces) as they adopted nutrition form simple sugars. Pectinases were lost from several fungal lineages when they associated with animals; examples include species of the chytrid Bathtrachochytrium, which is parasitic on frogs, Malassezia, the cause of dandruff, and of inhabitants of mucous membranes, Candida.

Similarly, it appears that pectinases have duplicated rapidly in organisms that adopt nutrition that is plant-based. It is indicated by the analysis of Chang et al. that zygomycetes ancestors lost nearly all of the pectinase homologues. Though there are no pectin-specific pectinases in Rhizophagus irregularis the high numbers of pectinase copies and low gene diversity of the major pectinase family GH28 in Phycomyces, Rhizopus, and Mucor represent recent gene duplications, which is consistent with previous findings (Mertens et al., 2008).

Microbial slime and origin of terrestrial fungi

Fungi, which are decomposers, are key organisms in the recycling of plant material in global carbon cycles. It was hypothesised by Chang et al. that genomes of early diverging fungi may have inherited from an ancestral species pectinases that had been able to extract nutrients from land pants that contained pectin and their algal allies (Streptophytes). This study was aimed at inferring the geological timing of the plant-fungus association, based on pectinase gene expansions and on the organismal phylogeny. In this study 40 fungal genomes were studied, 3 of which, including Gonapodya prolifera, were sequenced. Rozella diverged first from all other fungi, according to the organismal phylogeny from 136 housekeeping loci. Gonapodya was included among the flagellated fungal species that were predominantly aquatic in the Chytridiomycota. A sister group to the Chytridiomycota were the zygomycota I and zygomycota II, that were predominantly terrestrial fungi, as well as the ascomycetes and basidiomycetes that comprise Dikarya. There are 27 genes in the genome of Gonapodya, which represented 5 of the 7 classes of pectin-specific enzymes that are known from fungi. A common ancestry is shared by most of these with pectinases from Dikarya. Like many Dikarya, Gonapodya can use pectin as a carbon source for growth in a pure culture, which indicates functional and sequence similarity. Evidence that even ancient aquatic fungi had adapted to extract nutrients from the plants in the green lineage is provided by shared pectinases of Dikarya and Gonapodya. It is implied by this that 750 Ma (million years ago) the estimated maximum age of origin of the Streptophytes, which contained pectin, represents a maximum age for the divergence Chytridiomycota from the lineage that included the Dikarya.

The majority of modern fungi are characterised by close associations with plants or plant products. Fungi in Dikarya (Ascomycota and Basidiomycota) have key roles as decomposers (Floudas et al., 2012). Contrasting with the Dikarya, the lineages that diverged early in the evolution of the fungi, such as the zoosporic chytrids and some of the zygomycetes, use diverse sources of nutrition (Stajich et al., 2009). Many are associated with animals or animal products, though there are some that are associated with plants. Among early lineages the diversity of nutritional modes raises the questions of what the ancestral food of fungi might have been, and when the origin of the close and ecologically vital association between plants and fungi may have occurred (James et al., 2006; Sprockett et al., 2011).

The questions about the ancestral nutrition of fungi were approached by Chang et al. through analysis of the fungal digestive enzyme genes, which break down plant material. Studies of fungal degradation of plant material have traditionally been focused on the breakdown of cellulose and lignin. However, cellulose is distributed too widely across organisms that include green, red and brown algae (Popper et al., 2011) for cellulases to be used as good markers for the association with the lineage of land plants. The earliest land plants had no lignin in their structure (Popper et al., 2011), therefore enzymes for degrading lignin may not have been present in the ancestral fungi to break down plant material (Floudas et al., 2012). On the other hand pectins are polysaccharides that are present only in the cell walls of the streptophyte algae (Sørensen et al., 2011; Wickett et al., 2014). Pectin-specific polysaccharides and the genes for their synthesis, more specifically, have only been identified in the land plants and in “advanced” streptophyte algae though not in the streptophyte algal species that diverged early (Sorensen et al., 2011; Mikkelsen et al., 2014). Enzymes that are present in fungi that degrade multiple pectic molecules are potentially good indicators of the association between fungi and the land plant lineage.

In contrast to Lignin and cellulose, pectins are water-soluble and form a jelly-like matrix in the primary cell walls of plants. Pectins are complex polymers often with poly- or oligosaccharide (e.g., aribogalactan, b-1, 4 galactan) side chains decorating backbones of homogalactuonan, xylogalacturonan, rhamnogalacturonan, rhamnogalacturonan I, and rhamnogalacturonan II. Multiple enzymes from different families of glycoside hydrolases (GHs), polysaccharide lyases (PL), and carbohydrate esterases (CE) are required for the complete degradation of the pectin complex (van den Brink & de Vries, 2011; Benoit et al., 2012). At least 9 families among the pectinases are pectin specific, including 3 GH families (GH28, GH53 & GH93), 4 polysaccharide lyase families (PL1, PL3, PL4 and PL11), and 2 carbohydrate esterase families CE8 and CE13). An important role is played by enzymes from GH28 in the degradation of pectin backbones by fungi (van der Brink & de Vries, 2011). The other nonpectin-specific families, which may also breakdown other molecules, include (GH2, GH35, GH43, GH51, GH54, GH78, GH88, GH105, PL9, CE12, and CE1. In this study, Chang et al. first surveyed the distribution of families of pectinase genes across fungi and outgroups. They then reconstructed pectinase phylogenies for a detailed picture of the diversification of fungi during the evolution of major fungal lineages.

The goal of Chang et al. was to reconcile periods when there was expansion of pectinase genes with the fungal organismal phylogeny in order to identify times when ancestral fungi were adapting to using plants for acquiring nutrients. Chang et al. predicted that if the ancestral fungal species were associated with land plants and the charophytes that contained pectin the ancestral fungal species would have contained at least some pectinases. The pectinases may not have diversified until more recent cladogenesis and the rise of the Dikarya, if the early fungi fed on animal or animal-like material, and only later became associated with plants.

Chang et al. needed a robust fungal phylogeny in order to infer patterns of pectinase evolution among fungi. The branching order for the earliest splits among fungi has remained controversial, in spite of recent efforts at resolution (James et al., 2006; Liu et al., 2009; Ebersberger et al., 2012).

A large number of fungal genomes have been sequenced and annotated in recent years, which led to phylogenies that had increasingly rich sampling of loci and taxa (Ma et al., 2009; James et al., 2013; Suga et al., 2013; Tisserant et al., 2013). In this study, Chang et al., initiated a whole genome sequencing of Gonapodya prolifera, Coemansia reversa and Conidiobolus coronatus, in collaboration with Joint Genome Institute. They sampled whole genome data from 127 fungi, as well as 13 species from other eukaryotic groups, with an emphasis on the taxa that represented early splits in fungi. In order to explore factors that contributed to phylogenetic conflicts and to phylogenetic resolution, they applied multiple analytic methods to individual and concatenated genes.

With improved genomic and taxonomic sampling of fungi and related taxa, and with the study of     fungal pectinases, the aim of this study was to:

1)   Improve understanding of higher level fungal phylogeny,

2)   Infer the origin and diversification pattern of pectinases among fungi, and

3)   Infer the phylogenetic and geological age of plant/fungus associations, based on patterns of pectinase gene expansion.

Conclusions

It was shown by Change et al. by using a robust phylogeny that pectinases, enzymes for degrading plant cell walls, duplicated in an ancestral fungus that probably still lived in freshwater, in association with streptophyte algae that contained pectin. Using the age estimate of the streptophytes as a constant, the age estimate of the common ancestor of terrestrial fungi predates the origin of land plants. Early terrestrial fungi    may have evolved in semiaquatic microbial slime, with the ancestors of zygomycetes tracking arthropods or other animals, while the ancestors of Dikarya followed plants onto land. The evidence from the results of this study for an association of early fungi with early plants implicates fungi in processes of decomposition and symbiosis in the earliest terrestrial ecosystems.

Chang, Y., et al. (2015). "Phylogenomic Analyses Indicate that Early Fungi Evolved Digesting Cell Walls of Algal Ancestors of Land Plants." Genome Biology and Evolution 7(6): 1590-1601.