Understanding the evolution of the mycorrhizal symbioses: a matter of phylogenetics or paleontology?

 


This is Part 4 of a blog series from the South American Mycorrhizal Reserach Network. Read Part 1, Part 2, and Part 3.

A mycorrhiza, from the greek mýkēs (“fungus”) and rhiza (“roots”), i.e. “fungus roots”, is the symbiotic association between a fungus and the roots of vascular plants that serve as hosts. In this remarkable symbioses, the fungus receives carbon from the plant in the form of carbohydrates as glucose and sucrose, and -as recently described- in the form of fatty acids (Jiang et al. 2017; Keymer et al. 2017; Luginbuehl et al. 2017). Through its hyphae (long branching filamentous vegetative structure of a fungus, “fungus roots”), which are way longer and finer than root hairs, the fungus explores a significantly bigger soil volume than the roots, transferring to the plant host, water and nutrients such as nitrogen, phosphorous, calcium, magnesium, potassium, iron, among others. Usually these nutrients are not chemically or spatially available for the plant hosts. Mycorrhizal fungus can form extensive networks which connect and communicate plants and trees from the same and different species. How has this crucial symbiosis evolved? Two recent New Phytologist papers by Brundrett and Tedersoo (2018) and by Strullu-Derrien et al. (2018) present complementing insights, from a phylogenetic and a paleontologic approach, respectively, on this issue.

First, to identify how the mycorrhizal symbioses has evolved, Brundrett and Tedersoo (2018) identified precisely how many types of mycorrhiza exist, which is not a trivial task. They delimit four mycorrhizal types: arbuscular mycorrhiza (AM), ectomycorrhizas (EcM), ericoid mycorrhizas (ErM), and orchid mycorrhizas (OrM). Briefly, EcM fungi are the ones forming mushrooms or fruiting bodies, while the other mycorrhizal types do not form fruiting bodies and are found just below the soil surface. While AM and EcM symbioses are mostly obligatory and the carbon energy and water benefits to the plant are clear, ErM and OrM symbioses are obligatory but the benefits to the fungus are not so clear. ErM are restricted to the Ericaceae and Diapensiaceae (Ericales) plant families; OrM are restricted to Orchidaceae family; AM is present in multiple vascular plants and bryophytes lineages, but many plants within these lineages have switched which type of mycorrhizal fungi they associate with over time; finally, EcM are present in two Gymnosperms and 28 Angiosperms lineages, and some plants within these lineages have shifted to AM or to non-mycorrhizal states.

 
Figure 2 of Strullu-Derrien et al. (2018). Fossil (c. 407 million years) of the Rhynie plant Horneophyton lignieri. Left: plant (5-10 cm height) habitat. Right: ‘Paramycorrhizas’ (mycorrhizal-like) structures such as hyphae, vesicles, arbuscules, an…

Figure 2 of Strullu-Derrien et al. (2018). Fossil (c. 407 million years) of the Rhynie plant Horneophyton lignieri. Left: plant (5-10 cm height) habitat. Right: ‘Paramycorrhizas’ (mycorrhizal-like) structures such as hyphae, vesicles, arbuscules, and spores from different fungal phyla (Glomeromycotina and Mucoromycotina). Figure reproduced with permission of the authors and New Phytologist.

 

At a global scale, most plant communities are dominated by mycorrhized plants, but some patterns are shown by Brundrett and Tedersoo (2018): AM plants dominated all ecosystem types in the world, EcM plants are very rare or absent in tropical areas, whereas non-mycorrhizal plants are more common and diverse in degraded, arid, arctic, and alpine zones. As mentioned before, plant lineages commonly include members that have switched from AM or EcM to other mycorrhizal types, so Brundrett and Tedersoo (2018) coupled a literature review spanning 135 years and phylogenetic analysis of recently published large-scale plant phylogenies, to show the evolutionary history and host-plant abundance of the different mycorrhizal types. Regarding the evolutionary history, they identified three waves of mycorrhizal colonization. The first wave started > 450 million years ago, with AM fungi colonizing early land plants (or allowing colonization of land by plants, as suggested by Pirozynski and Malloch, 1975). The second wave started with the Cretaceous some 145.5 million years ago with the appearance of the EcM-associated Pinaceae, which come to mainly constitute extensive forested areas in the world, in this second wave also appeared the ErM (Ericaceae and Diapensiaceae) and OrM (Orchidaceae) associated plant families, as well as several non-mycorrhized plant families, and although most plant families had a single mycorrhizal type, changes to another are starting to appear in plant families previously associated with AM fungi. Finally, a third wave of mycorrhizal colonization started at the Paleogene (c. 65 million years ago), with the diversification and appearance of more families with different mycorrhizal types within each family (both families previously only AM or EcM). Finally, and using the phylogenetic methods described above, Brundrett and Tedersoo (2018) conclude that 72.0% of vascular plants species are AM, 2.0% are EcM, 1.5% are ErM, 10% are OrM, just 8% are non-mycorrhizal, and 7% of vascular plant species can be either AM or non-mycorrhizal.

Figure 1 of Strullu-Derrien et al. (2018) showing the appearance of genomic traits related to mycorrhizal evolution (left)and the oldest known fossils (right). The asterisk represents the Rhynie chert. AM, arbuscular mycorrhizas; CAZymes, Carbohydra…

Figure 1 of Strullu-Derrien et al. (2018) showing the appearance of genomic traits related to mycorrhizal evolution (left)and the oldest known fossils (right). The asterisk represents the Rhynie chert. AM, arbuscular mycorrhizas; CAZymes, Carbohydrate-Active enZYmes; CMm, coil-forming mycorrhizas in Mucoromycotina; MiSSPs, mycorrhizainduced small secreted proteins; PCWDEs, plant cell wall-degrading enzymes. Figure reproduced with permission of the authors and New Phytologist.

Contrasting with Brundrett and Tedersoo (2018), Strullu-Derrien et al. (2018) used a paleontological approach, based on fossil evidence, to produce a model of mycorrhizal evolution. In this model, they included genomic traits related to mycorrhizal evolution based on molecular clocks estimates, and reported in published literature. This model is shown in the image to the right.

What is the difference between this model and the one of Brundrett and Tedersoo (2018)? Simply put, Brundrett and Tedersoo (2018) propose the starting point of mycorrhizal colonization waves very close in time to the origin of the plant hosts: first land plants spores appeared c. 460 million years ago, and the first wave of arbuscular mycorrhizal (AM) fungi colonization started <450 million years ago; Pinaceae plants appeared c. 150 million years ago, and the appearance of ectomycorrhizal (EcM) fungi is assumed shortly after, c. 145.5 million years ago. In contrast, Strullu-Derrien et al. (2018) put the origin of AM colonization with the appearance of their fossils, c. 407 million years ago, the oldest known fossils of EcM symbiosis c. 52 million years ago. Those are differences of approximately 43 and 93.5 million years, respectively!

Can we assume that because no fossil have been founded, plants survived the Earth some 53 million years without mycorrhizas? Or do we assume – as Brundrett and Tedersoo, (2018) – that it was not such a long time (just 10 million years)? In the other hand, currently the plants of Pinaceae are highly interdependent and host a great diversity of EcM fungi, do we assume that these plants survived without their fungal symbionts for 98 million years because we do not have earlier fossil evidence? Strullu-Derrien et al. (2018) highlight that due to exceptional geological requirments, it is difficult to find earlier EcM well-conserved fossils, and state that the first EcM may have evolved way earlier, somewhere between the origin and diversification of Pinaceae. These questions are not trivial and easy to answer. Although the use of ‘omics’ data to construct mycorrhizal evolution models has been very useful, there are still some caveats with this techniques, as a strongly biased sampling towards some plant groups and the northern hemisphere (Bueno et al. 2017), incorrect assignment of the mycorrhizal type in published publications (as reported by Brundrett and Tedersoo, 2018), a non-clear evolutionary history of some fungal groups, and scarce fossil sources to calibrate phylogenetic trees. Precisely, and from the paleontological point of view, this scarcity is also an issue discussed by Strullu-Derrien et al. (2018), which indicate that mycorrhizas need exceptional geological conditions to be fossilized, although some hints are given by the authors. Besides, they indicate that once a fossil fungus is found inside a fossil root, this does not necessarily mean its forming a mycorrhiza. An additional problem with the mycorrhizal record, also seems to be a geographical bias.

A possible solution to this contrasting -yet complementing- clash of approaches may be to relate mycorrhizal evolution to the Earth geochemical history. As Strullu-Derrien et al. (2018) putted “Mycorrhizas are not just passive responders to the environment; they are also active agents of environmental change”. An example of this is given in the image below.

Figure 3 of Strullu-Derrien et al. (2018) showing the evolution of the endomycorrhizal symbioses during the Palaeozoic, and its relation with CO2 and O2 atmospheric concentrations. AM, arbuscular mycorrhizas; CMm, coil-forming mycorrhizas in Mucorom…

Figure 3 of Strullu-Derrien et al. (2018) showing the evolution of the endomycorrhizal symbioses during the Palaeozoic, and its relation with CO2 and O2 atmospheric concentrations. AM, arbuscular mycorrhizas; CMm, coil-forming mycorrhizas in Mucoromycotina. Figure reproduced with permission of the authors and New Phytologist.

 
GSBI