How Mycorrhizal Symbiosis Shaped Angiosperm Evolution

Hal xxx

How mycorrhizal symbiosis shaped Angiosperm evolution

Tom Hattermann1*

Abstract The contribution of mycorrhiza to plant performance, distribution and diversity is well documented. This contribution un- derlines the intricate relationship between Plants and Fungi functionality. Yet, how such a symbiosis shaped Angiosperm evolution remains poorly understood due to the difficulties of obtaining retrospective evidence. Biotic interactions are consistently evolving through time and space, but mycorrhizal symbiosis is remarkably stable over the macroevolutionary time-scale. Based on a biogeographical and functional approach, we propose here an overview of the potential effect of a fungal-partner on the evolution of plant strategies. We first review some generalities about mycorrhiza. Then, we discuss phylogenetic and climatic patterns of root traits to justify bidirectional control. By an integrative approach to plant functionality, we propose a debate on the potential influence of mycorrhiza and other fungi on foliar trait plasticity and evolution. Next, we claim the importance of considering ontogeny in ecosystem dynamics and functionality studies. Following this, we address the evolutionary profile of two emblematic Angiosperm groups: mycoheterotrophic plants from a trait point of view and Fabaceae from an ecosystem point of view. Plants have an extensive cortège of associated microorganisms making it work as a whole. Plasticity of the association and multidirectional control could be the main catalysts for Angiosperm success through evolutionary time. This coordination interferes with the traceability of mycorrhizal contribution to Angiosperm evolution, which could have shaped it in many ways. By further extending our knowledge on the communication between mutualistic microorganisms and plants, new research will increase our under- standing of how mycorrhizal symbiosis has shaped Angiosperms genetically, functionally and morphologically through time.

Keywords Mycorrhizal fungi, plant microbiota, Angiosperm, multipartite interactions, functional ecology, biogeographic pattern

1Master 2 EcoTrop ( AgroParisTech) & BioGET (Université de Montpellier), UMR EcoFoG, Kourou, French Guiana. * Corresponding author: [email protected]

Contents 1 ; Martin et al. 2017). While origin of the major groups of terrestrial fungi (Ascomycota, Basidiomycota, and Glom- eromycota) started around 600 million years ago, various Introduction 1 endophyte fungal forms are more than 450 million-year-old 1 Roots morphology, functionality and and have adopted a plethora of colonization strategies and compatibility 3 relationship profiles ranging from parasitic to mutualistic 1.1 Biogeographic patterns 3 (Krings et al. 2007). 1.2 Roots as microbiota niche 3 2 A fungal-root-leaf continuum? 3 The first known mycorrhizal structure was observed 2.1 Is there a mycorrhizal contribution? 3 between a Bryophyte-like-plant and a Glomales-like-fungi 2.2 Foliar endophyte partners 4 and was very similar to extant arbuscular mycorrhizal types 3 Ontogenetic modulation 4 (Redecker et al. 2000). Shared genes related to symbiosis 3.1 Inter-individual level 4 establishment were found throughout terrestrial Embryo- 3.2 Intra-individual level 4 phyta and support the idea of a long and ancient coevolu- 4 Mycoheterotrophic strategies 4 tionary process between plants and fungi (Bonfante & 4.1 Implications for strict mycoheterotrophic plants 5 Genre 2010). 4.2 Implications for myxotrophic plants 5 Today, most of plants and fungi follow the usual lati- 5 Tripartite interaction in Fabaceae 5 tudinal richness patterns and express a higher functional 5.1 Functional and ecological implications 6 and taxonomic diversity at low latitudes (Tedersoo et al. Conclusion 6 2014). Mycorrhizal symbiosis accounts for 0.5 to 10% of References 6 global fungal diversity (Blackwell 2011; Taylor et al. 2014) and affects 65% of all land plants represented by various growth forms in very different biomes (Wang & Qiu 2006; Brundrett 2009). Currently, almost every Angiosperm plant Introduction is able to form mycorrhizas (92%) leading to a high diversity of associations. Morphologically, four main mycorrhizal types have been identified: Arbuscular mycorrhiza The phytosphere is one of the world's largest habitats (AM), Ectomycorrhiza (EM), Orchids mycorrhiza (OcM) for microorganisms such as bacteria or fungi. The fossil and Ericoids mycorrhiza (ErM). record reports that plants and fungi have had an intricate relationship since the beginning of land life evolution (fig. How mycorrhizal symbiosis shaped Angiosperm evolution —2/10

The success of the association could be explained by The purpose here is to give arguments about different the mutual investment between host plant and fungal sym- roles played by mycorrhiza in plant diversification and evo- bionts. In exchange of carbon resources produced by pho- lution. This assessment will focus on the Angiosperm tosynthesis, fungi optimize hydraulic and mineral nutrition group given its great importance in the terrestrial ecosys- of host plants by expanding root exploitation systems and tem. This subject couples very diverse research fields with by making sophisticated compounds from the soil multiple context-dependant specificities. Because of the (Marschner & Dell 1994). Host-plants can benefit from difficulty of summarizing such a vast and variable subject, multifunctional services such as resistance to drought we propose here a non-exhaustive view to address some po- (Mariotte et al. 2017), a better tolerance to heavy metals tential and innovative research directions. This overview and pathogens (Hildebrandt et al. 2007; Wehner et al. will discuss to what extent mycorrhiza shaped flowering 2010) and can have an impact on seed germination (Jacque- plant strategies through its influence on root and leaf traits. myn et al. 2015) and seedling establishment (van der Then, it will approach the current need to consider ontog- Heijden & Horton 2009). In this sense, having mycorrhizal eny in studying ecosystem functionality through mycorrhi- structure generates high functional diversity in three non- zal symbiosis. Next, we will see how a strong partner-de- exclusive ways : i) habitat or environment specialization, pendency can uniformly shape plant morphology. Then, we ii) internalizing of one partner by the other for protection or address the evolutionary profile of Fabaceae, the third most iii) by coordinating plants with multiple partners (Buscot diverse Angiosperm family. Finally, we address some con- 2015). At a larger scale, these interactions play a key role sequences of multipartite interaction on ecosystems evolu- in the multi-functionality of terrestrial ecosystems by regu- tion. lating biogeochemical cycles, influencing soil properties and biodiversity (Rillig & Mummey 2006; Schöll et al. 2008; Wagg et al. 2014). Consequently, mycorrhiza have a direct effect on plant productivity, resistance and distribution (Leifheit et al. 2014; Moënne-Loccoz et al. 2015). At macroevolutionary time-scales, mutualisms sel- dom lead to parasitism. Some authors argued that mycorrhiza is a case of reciprocal parasitism which is very dynamic and reversible (Egger & Hibbett 2004), while others argued it is a good mutualistic example of an ecological and evolutionary stable relationship (Frederickson 2017). In view of the wide diversity of associations, every case can exist and could have arisen during evolution. This assessment can be supported by various processes of symbiont transmission, from independent reproduction as well as horizontal and pseudo-vertical transmission, to synchro- nized and coordinated vertical transmission, as in the case of vegetative reproduction or obligate partners. How interaction is transmitted could play an important role in symbiosis stability (Wilkinson 1997; Herre et al. 1999; Wil- kinson & Sherratt 2001). Moreover, natural selection oper- ates differently for each partner, due to difference in origins and biological functioning. In this sense, evolutional control is at least bidirectional. However, each organism at- tempts to reach the greater ratio between benefit and cost. So in many cases, the best rate of exchange is rewarded (Kiers & Heijden 2006; Kiers et al. 2011). How mycorrhiza contributes to plant performance and functioning is well-documented. This knowledge was mainly obtained through experimental studies due to the difficulty of isolating the main effects of mycorrhiza from potential sources of interference in the field (e.g. bacterial communities and climatic variation). There is a serious lack of information for certain fungal groups (e.g. Ectomycor- Figure 1: A summary of land colonization by plant and the mutualistic rhiza) at the global scale, due to biological specificities and symbiont(s) associated through time. One possible scenario is that An- relative difficulties of finding sporocarps in the field giosperm success could be explain by their abilities to integrate multi- (Brundrett 2017). It is also a serious challenge for investi- ple symbionts, making them work as a whole organism (Martin, Uroz gators to study these interactions at evolutionary scales. et Dijskra, 2017). How mycorrhizal symbiosis shaped Angiosperm evolution —3/10

1 Roots morphology, functionality determine the relationship between root traits, climatic factors and mycorrhizal associations at a global scale. From and compatibility another perspective, mycorrhizal colonization is also known to induce modifications in root system architecture During plant evolution, the switch in morphotype was through a metabolic pathway (Wu et al. 2012; Eissenstat et correlated with deep implications for root morphology and al. 2015), suggesting that plants have also shaped mycor- belowground-plant strategies (Brundrett 2002; Smith & rhizal fungi and that root morphological and functional Read 2008a). The study of Valverde-Barrantes et al. (2017) traits are under bidirectional biotic control. Thus, two be- showed that root variations at the global scale are mainly coming for the association can be identified: i) fungal mu- explained by phylogenetic structure, then by climatic tation process lead to coevolution through coordinate adap- factors, whereas the mycorrhizal morphotype had little ef- tations to root morphology, anatomy and functionality, ii) fect. These authors proposed that the morphological and initial fungal partner is succeed by another or, eventually anatomical alterations of root could have facilitated transi- the plant remains without symbiont. tion from the ancestral AM to alternative fungal associations. In this sense, growth form evolution and occupancy of contrasting biomes by plants over time could have led to a shift in fungal-symbionts and consequently induced 2 A fungal-root-leaf continuum? variation in mycorrhizal functionality. Root trait trade-offs have a broader set of possiblities 1.1 Biogeographic patterns due to less constraining forces than those exerted on leaves Today, most Rosids have few families with many (Valverde-Barrantes et al. 2017). Despite that, evidence non-mycorrhizal plants (e.g. Polygonaceae and Nyctagina- showed no universal resource economy across leaves and ceae) and many families with AM plants (e.g. Fabaceae, roots. Coordination between the phyllosphere and the Myrtaceae and Rosaceae). A simultaneous evolution was rhyzosphere varies between growth forms and plant clades reported between Ectomycorrhizals and Rosids during the (Weemstra et al. 2016). The relationship could determine Jurassic and Cretaceous leading to disparate distributions at plant functioning in certain interesting context. In fact, the global scale (Brundrett 2002; Tedersoo et al. 2012). EM some authors have shown that belowground traits related to are more widely distributed in northern temperate zones resource acquisition and conservation have a control effect and boreal forest biomes than in the tropics, where species on leaf functionality, particularly in terms of biomass allo- are sporadically distributed. The predominance of EM cation (Freschet et al. 2015). strategies in boreal ecosystems is correlated with an increase in branched root systems (Guo et al. 2008; Ku- 2.1 Do mycorrhiza contribute? bisch et al. 2015). In cold climates, the acute need for plants Leaves are generally the main headquarter for photo- to have efficient soil exploration strategies is met by my- synthesis and then, are responsible for carbohydrates prod- corrhizal associations and morphological modifications ucts required by fungal symbionts. Plant production and in- (Comas et al. 2012). Also, freezing temperatures have vestment can have a balancing effect on mutualistic rela- strong filtering effects on plant anatomy, such as the pro- tionships. In this way, leaf traits and composition could portional amount of cortical tissue permitted (Valverde- have partially evolved under nitrogen (N) and phosphorous Barrantes et al. 2016). It appears that the climatic resistance (P) fungal-dependency (Craine et al. 2009). Studies have is provided by both this anatomical property and by extra- already shown that highly specific responses in leaf metab- cellular fungal position (EM). olome is related to an increase in foliar P content after AM colonization (Müller et al. 2014). Through another meta- 1.2 Roots as a fungal niche bolic pathway, such colonization could also induce a Root characteristics such as tissue density or the ratio greater tolerance towards leaf pathogens (Gernns et al. between the lignified stele and parenchymatous cortical tis- 2001). Otherwise, experiences have shown on a tripartite sue seems also to play a determinant role in mycorrhizal interaction between a generalist herbivore, a leguminous habitat availability in plant tissue (Kong et al. 2017). An plant (e.g. Plantago lanceolata) and an AM symbiont have extended root niche could lead to a wider spectrum of sym- shown differential effects on plant and herbivore perfor- biont possibilities or in some cases, to a partitioning of root mance depending on plant and fungal age (Wurst et al. functionality by multiple symbionts. At least, woody spe- 2004). A recent study have shown that fungal-symbiont cies with large root diameters seem to compensate their ac- confer functional plasticity against foliar and climatic de- quisition abilities by hosting higher mycorrhizal fungal bi- mands by coupling root nutrient acquisition (Kong et al. omass per unit of root mass (Comas et al. 2014; Valverde- 2017). In absence of fungal partner, plant responses would Barrantes et al. 2016). As a critical indicator of root physi- be materialized by an adjustment of foliar tissue density or ological function, root diameter has been reported to be surface area. very variable between tropical families. Causes of such Such adaption could only be realized by phenotypic variation remain unidentified but could be partially plasticity or evolutionary processes which only occur at explained by mycorrhizal compatibility (Gu et al. 2014). A larger time scales. In this sense, mycorrhiza could play a serious lack of information about belowground traits and critical role in leaf responses to biotic interaction and cli- soil communities in tropical species makes it difficult to How mycorrhizal symbiosis shaped Angiosperm evolution —4/10 matic variation. By permitting wider range of metabolic re- 3.1 Inter-individual level sponses despite of morphological modification, the contri- A recent study has shown that collaborative interac- bution of mycorrhiza in leaf trait evolution should not be tions permits substantial exchange of carbon between well- neglected. established trees. Through a common EM network, photo- assimilates produced in the foliage could be transmitted to 2.2 Foliar endophyte partners other individuals. By applying an isotope labeling approach In another way, evolution lead fungi to colonize leaf at canopy level, bidirectional transfer between neighboring structures directly and are thus called foliar endophyte interspecific trees has been demonstrated (e.g. about 280 fungi. Recent studies have reported the major role of these kilograms per hectare per year tree-to-tree transfer ; Klein symbionts on plant productivity, diversity and functional- et al. 2016). Another study has demonstrated that mycor- ity. This highly diverse group of fungi have many variable rhizal networks allow collaborative processes between ma- impacts on plant communities depending on the status of ture trees and closely related seedlings. More mature their interaction. Mutualistic ones influence plant fitness by (taller) trees from the same genera seem to alter mycorrhi- conferring biotic stress tolerance, playing on stomatal aper- zal communities which induce abundant and diverse my- ture and altering resource allocation (Rodriguez et al. corrhizal colonization in healthy related seedlings. This fact 2009). These symbionts are quite different of mycorrhizal is positively correlated with an increase of N and P content, associations but have coevolved simultaneously during An- suggesting that tall trees enable better nutrient uptake to giosperm evolution (Krings et al. 2007). As mycorrhizal- near-seedlings. Consequently, the growth of seedlings fungi, there are two broad classes of potential mechanisms seems to depend directly on mutualistic symbiont commu- by which endophytes could contribute to host-plant protec- nities and indirectly on the distance from the related tall tree tion. It could induce the expression of intrinsic defense (Dickie et al. 2002). In this sense, tall trees represent a mas- mechanisms or provide additional sources of defense like sive storage of nutrients and a source of mycorrhizal-fungal endophyte-metabolites. Some evidence argued that endo- diversity. Consequently, the size and the ontogenetic stage phytic and mycorrhizal fungi could play a coordinate role of each tree could differentially influence resource parti- by augmenting host defensive responses against pathogens tioning via the mycorrhizal network. (Herre et al. 2007). Another study has shown that, injection of foliar endophyte fungi to mycorrhizal plant influences 3.2 Intra-individual level host-plant biomass by inducing N and P accumulation and The development of the root system could permit modulating their competitive ability. Authors also suggest plants to explore deeper in the ground and access new re- a feedback effect by foliar symbionts which could affect sources. Also, plants induce differential climatic require- mycorrhizal status and growth of co-infected and neighbor- ments (e.g. light availability, wind exposure, temperature, ing plants (Omacini et al. 2006). etc.) by vertically exploring the aboveground environment, By linking phyllosphere to belowground communi- extending their crown and increasing their foliar surfaces. ties, mycorrhiza and foliar endophyte fungal should be Finally, the expression of sexuality could create new needs. more considered in forthcoming studies about plant eco- All these processes are sources of variation in plant func- physiology, ecosystem dynamic and functioning. Their co- tioning over time (e.g. biomass allocation in plant structure ordinated implication in Angiosperm evolution has un- and tissue) and involve differential resource requirements doubtedly contributed to their success through metabolic and functioning for the mycorrhizal partner. In this sense, adaptations. the association can evolve in many ways during plant ontogeny: i) by influencing exchange rates between both partners (e.g. unidirectional at juvenile stage and bidirectional 3 Ontogenetic modulation at maturity), ii) being necessary only for a specific stage of plant growth, or even iii) by specific fungal partners for Mycorrhiza play a key role in plants intra- and inter- each ontogenetic stage. specific competitiveness by affecting belowground interac- Finally, it should be mentioned that fungal-partner age also tion networks. Nestedness, modularity and specificity of has an impact on association functionality. Depending on such interactions vary and depend on mycorrhizal type, plant ontogeny, fungal development induces differential community structure and environmental conditions (van gene expression and plant responses (Daguerre et al. 2017). der Heijden et al. 2015). Depending on plant identity, fun- However, this will not be further explored here due to brev- gal-symbionts modulate temporal stability of the commu- ity, and seeing as the focus of this paper is plants. nity and mediate compensatory effects among plant species and functional groups (Yang et al. 2014). The variable effect of mycorrhiza at the community level is well documented, but very few studies talk about variation during a 4 Mycoheterotrophic strategies single plant ontogeny. In fact, fungal-partner and functional responses could evolve through plant growth and expan- Some mycorrhizal symbiosis has led to an extreme sion. end in the mutualism–parasitism continuum. These plants are qualified as epiparasitic because they directly depend on a single fungal-species for carbon metabolites. Most of How mycorrhizal symbiosis shaped Angiosperm evolution —5/10 the time, such strategy has resulted in the loss of full pho- root system is developed from a lateral vegetative meristem tosynthetic abilities. This mode of life could have led to resulting on homogenous root axis (e.g. homorhizy). Such vestigial stomata on leaves and in the more extreme cases, conformation had led to a star-like root system, which in stomata are totally missing from aerial parts. The absence dicots can be achieved only by reductions of allorhizy. All of mycorrhizal specificity and the absence of specialized those traits constitute greater abilities to develop volumi- plant–fungal interface for resource exchange, do not permit nous and diverse subterranean organs (e.g. rhizomes, scale them to be considered as mycorrhizal plants in sensu leaves, roots), which promote the development of more stricto. In fact, their symbionts are generally saprotrophs complex symbiont colonization patterns. from the soil or exploit the surrounding vegetation through the mycorrhizal network (e.g. endophytic-EM associations 4.2 Implications for mixotrophic plants ; Smith & Read 2008b). Globally distributed, such plants Less extreme strategies also exist and are known as display strong convergent evolution manifested by similar mixotrophic plants. Carbon resources are then provided adaption to their peculiar mode of life. Mycoheterotrophy through photosynthesis and via hyphal links. Such strategy in angiosperms has multiple independent origins and are is more permissive in term of growth form evolution. In widely represented in Monocots (7 families, 100 genera) fact, mixotrophic lianas exist in tropical forest and some and Dicots (3 families, 42 genera ; Merckx & Freudenstein mixotrophic shrubs are frequent in temperate areas. Then, 2010); each one mainly composed of Orchidaceae and Eri- the net flow of carbon occurring in the belowground inter- caceae species which mostly manifest strong specificities action could be bidirectional and could led to a more partial and dependency between partners (Shefferson et al. 2007). relationship between both partners. How these plants invest in mycorrhizal networks depends on various factors such as 4.1 Implications for strict mycohetero- plant and symbiont identity, soil and plant communities and trophic plants also environmental conditions (van der Heijden & Horton At the individual plant level, such evolution has in- 2009; Schlaeppi et al. 2015). Consequently, their contribu- duced strong physiological, morphological and functional tion to biogeochemical cycles and to the dynamics and consequences. Vegetative parts are conspicuously reduced, composition of forest communities remains unclear (Bidar- permitting for a number of species to live entirely under- tondo et al. 2000; Jacquemyn et al. 2017). ground during almost every stage of plant ontogeny. This trophic mode of life has deep implications for organ functionality. In the case of the root system, an almost universal 5 Tripartite interaction in loss of root hairs, decrease in surface area and in extreme cases, the complete suppression of roots is often reported. Fabaceae In this extreme case, mycorrhizal colonization may be functionally transferred to belowground modified stems. Fabaceae is known as the third most diverse plant Mycoheterotrophy also induce strong implications for the family (Roskov et al. 2017) and this success could be par- stem system, most of the time drastically reduced in terms tially explained by simultaneous evolution of mycorrhiza of architecture (e.g. thin and thread-like) and anatomy (e.g. and rhizobia in leguminous plants during the early Ceno- rare vascular tissues and secondary thickening poorly de- zoic (Epihov et al. 2017). The interacting abilities of legu- veloped). Specific vegetative adaptations (e.g. root tubers, minous plants play a major role in land colonization by tubercles and rhizomes) have permitted them to spread plants (e.g. on poor soils ; Sprent et al. 2017). There are two through asexual reproduction. However every Angiosperm main types of mycorrhiza in legumes: the ubiquitous AM recorded presents the ability to reproduce sexually (Leake and the sporadically distributed EM (see 1.1 section). AM 1994). These plants generally produced very large number hyphae have been occasionally reported as colonizing non- of size reduced seed and embryo. By containing minimal functional nodules, while mycorrhizal and nitrogen-fixing reserve carbohydrates at an early stage, the germination de- symbioses have distantly evolved (Scheublin & Van Der pends severely on the appropriate symbiotic fungus coloni- Heijden 2006). Among the three subfamilies, nodulation is zation. The shape, structure and surface features of dia- known to be rare in Caesalpinioideae, common in Mimo- spores involved often non-active and wind dispersal strate- soideae, and very common in Papilionoideae (Sprent & gies. Such strategies have mostly resulted on understorey James 2007). Mycorrhiza and rhizobia both benefit from plants, underling the strong incompatibility with certain carbohydrates provided by the host plant, which uses these growth forms (e.g. epiphyte or hemi-epiphyte). However symbionts as a source of energy. No direct contact seems which factors induced the most such evolution remains a to occur between the two potential plant symbionts, sug- mystery. Light availability in addition to a weak potential gesting that the host plant could have a strong control effect to explore the environment vertically, resource partitioning on the equilibrium of the tripartite interaction. In fact, the and fungal implications are many potential reasons which hostplant’s genes are responsible for P and N metabolism, could lead to these reduced vegetative forms. Some evi- translocation and regulation. This tripartite evolution has dence argues in favor of a phylogenetic predisposition for probably induced diverse mechanisms of common symbi- Monocots for many reasons (Imhof 2010). Primarily, due otic pathways as gene coordination or mutual exchange of to absence of secondary growth this group is mainly com- signal molecules (Chang et al. 2017). posed by herbaceous members. It also presents very specific root system which conserve the primary cortex. The How mycorrhizal symbiosis shaped Angiosperm evolution —6/10

5.1 Ecological implications As we have seen, mycorrhizal associations involved Through an isotopic approach, Roggy (1998) has numerous organisms from different phyla (Fig. 2). Today delimited eco-unity in French Guyana’s forest and suggests most of the plants have an extensive cortège of associated a linear co-evolution between macro and micro symbionts microorganisms, from mutualistic fungal to bacterial sym- during legume evolution. Nowadays, evidence shows that bionts which have intricately coevolved into their host-or- this tripartite interaction manifests an efficient resource gans. Working as a whole, plant-microbiota associations partitioning which could have shaped ecosystem function- promote plant growth by facilitating nutrient acquisitions ality, plant growth strategies and host-plants traits (e.g. tis- and providing metabolic responses. By conferring plastic sues N-P enrichment). In fact, nodulation has a more sig- and diverse responses, they have shaped Angiosperm evo- nificant requirement for P elements than mycorrhizal sym- lution through time and space. The stability of mycorrhizal bionts near nodules could probably provide. In this way, associations may arise from the coordination of every sym- evidence suggests that legumes are very versatile in their bionts, then exerting multidirectional control on plant evo- symbioses at life- and on an evolutionary time-scale lution. However, a lot of work remains to better understand (Sprent & James 2007). how mutualistic microorganisms communicate and interact The increase in N foliar tissue could have led to heavy eco- with plants and how they shaped Angiosperms genetically, logical consequences. In fact, primitive Fabaceae plants functionally and morphologically through time. showed strong insect damage through the fossil record (Epihov et al. 2017). Today, studies about Fabaceae ecology report a large set of defense mechanism against her- bivory. It could have played an indirect but critical role in the evolution of numerous leaf traits such as the degree of division or the presence of nectarous glands permitting ant- mediated defenses (Kursar & Coley 2003). In this perspective, Fabaceae evolution and diversification seems to be a remarkable example of how mycorrhizal symbiosis could have acted with multiple partners to result in soil fertilization effects, phytophagous regulation and plant diversification. These multipartite interactions have shaped tropical forests and probably induced the tran- sition from palm dominant trees to the current diverse dicots forest that we know.

Figure 2: Summary of the interaction network made possible by my- Conclusion corrhizal fungi at the individual plant-scale. Full red lines represent the physical and direct link made by mycorrhizal fungi and dashed Mycorrhizal associations are ancient and have helped lines represent the indirect ones. plants colonize terrestrial ecosystems. This interaction has become almost universal for the current globally distributed plant group, known as Angiosperm. The direct consequences of mycorrhizal association on plant morphology References can’t be clear at a macroevolutionary time-scale. In fact, both partners have influenced each other and many other Bidartondo, M.I., Kretzer, A.M., Pine, E.M. & Bruns, T.D. factors have stronger effects on plants such as phylogenetic (2000). High root concentration and uneven ecto- origin and climate. We also debated the indirect relation- mycorrhizal diversity near Sarcodes sanguinea ship between mycorrhiza and leaves and realized that being (Ericaceae): a cheater that stimulates its victims? associated can provide a wide range of foliar responses to Am. J. Bot., 87, 1783–1788. abiotic constraints and biotic threats. Through evolution, it Blackwell, M. (2011). The Fungi: 1, 2, 3 … 5.1 million spe- has played an important role by influencing plant metabo- cies? Am. J. Bot., 98, 426–438. lomes but also extrinsic organisms such as phytophagous. Bonfante, P. & Genre, A. (2010). Mechanisms underlying Considering plant functionality through ontogeny, we also beneficial plant–fungus interactions in mycorrhi- claimed how variable mycorrhizal associations can be and zal symbiosis. Nat. Commun., 1, 1–11. how they can differentially contribute to plant performance Brundrett, M.C. (2002). Coevolution of roots and mycor- and ecosystem functionality. By focusing on Mycohetero- rhizas of land plants. New Phytol., 154, 275–304. trophic strategies, we saw how an obligate association can Brundrett, M.C. (2009). Mycorrhizal associations and other induce a convergent evolution by reducing plant size and means of nutrition of vascular plants: understand- organs. Finally, we examined the case of leguminous plants ing the global diversity of host plants by resolving and how mycorrhizal-fungi and nitrogen-fixing bacteria conflicting information and developing reliable could have played a catalytic role in plant diversification means of diagnosis. Plant Soil, 320, 37–77. leading to the current ecosystems. How mycorrhizal symbiosis shaped Angiosperm evolution —7/10

Brundrett, M.C. (2017). Global Diversity and Importance roles of specific leaf area and root mass fraction. of Mycorrhizal and Nonmycorrhizal Plants. In: New Phytol., 206, 1247–1260. Biogeography of Mycorrhizal Symbiosis, Ecolog- Gernns, H., Alten, H. & Poehling, H.-M. (2001). Arbuscu- ical Studies. Springer, Cham, pp. 533–556. lar mycorrhiza increased the activity of a bio- Buscot, F. (2015). Implication of evolution and diversity in trophic leaf pathogen – is a compensation possi- arbuscular and ectomycorrhizal symbioses. J. ble? Mycorrhiza, 11, 237–243. Plant Physiol., 172, 55–61. Gu, J., Xu, Y., Dong, X., Wang, H. & Wang, Z. (2014). Chang, C., Nasir, F., Ma, L. & Tian, C. (2017). Molecular Root diameter variations explained by anatomy Communication and Nutrient Transfer of Arbus- and phylogeny of 50 tropical and temperate tree cular Mycorrhizal Fungi, Symbiotic Nitrogen- species. Tree Physiol., 34, 415–425. Fixing Bacteria, and Host Plant in Tripartite Sym- Guo, D., Xia, M., Wei, X., Chang, W., Liu, Y. & Wang, Z. biosis. In: Legume Nitrogen Fixation in Soils with (2008). Anatomical traits associated with absorp- Low Phosphorus Availability. Springer, Cham, tion and mycorrhizal colonization are linked to pp. 169–183. root branch order in twenty-three Chinese temper- Comas, L.H., Callahan, H.S. & Midford, P.E. (2014). Pat- ate tree species. New Phytol., 180, 673–683. terns in root traits of woody species hosting arbus- van der Heijden, M.G.A. & Horton, T.R. (2009). Socialism cular and ectomycorrhizas: implications for the in soil? The importance of mycorrhizal fungal net- evolution of belowground strategies. Ecol. Evol., works for facilitation in natural ecosystems. J. 4, 2979–2990. Ecol., 97, 1139–1150. Comas, L.H., Mueller, K.E., Taylor, L.L., Midford, P.E., van der Heijden, M.G.A., Martin, F.M., Selosse, M.-A. & Callahan, H.S. & Beerling, D.J. (2012). Evolu- Sanders, I.R. (2015). Mycorrhizal ecology and tionary Patterns and Biogeochemical Significance evolution: the past, the present, and the future. of Angiosperm Root Traits. Int. J. Plant Sci., 173, New Phytol., 205, 1406–1423. 584–595. Herre, E.A., Knowlton, N., Mueller, U.G. & Rehner, S.A. Craine, J.M., Elmore, A.J., Aidar, M.P.M., Bustamante, (1999). The evolution of mutualisms: exploring M., Dawson, T.E., Hobbie, E.A., et al. (2009). the paths between conflict and cooperation. Global patterns of foliar nitrogen isotopes and Trends Ecol. Evol., 14, 49–53. their relationships with climate, mycorrhizal Herre, E.A., Mejía, L.C., Kyllo, D.A., Rojas, E., Maynard, fungi, foliar nutrient concentrations, and nitrogen Z., Butler, A., et al. (2007). Ecological Implica- availability. New Phytol., 183, 980–992. tions of Anti-Pathogen Effects of Tropical Fungal Daguerre, Y., Levati, E., Ruytinx, J., Tisserant, E., Morin, Endophytes and Mycorrhizae. Ecology, 88, 550– E., Kohler, A., et al. (2017). Regulatory networks 558. underlying mycorrhizal development delineated Hildebrandt, U., Regvar, M. & Bothe, H. (2007). Arbuscu- by genome-wide expression profiling and func- lar mycorrhiza and heavy metal tolerance. Phyto- tional analysis of the transcription factor reper- chemistry, Molecular Basics of Mycorrhizal Sym- toire of the plant symbiotic fungus Laccaria bi- biosis, 68, 139–146. color. BMC Genomics, 18, 737. Imhof, S. (2010). Are Monocots particularly suited to de- Dickie, I.A., Koide, R.T. & Steiner, K.C. (2002). Influ- velop mycoheterotrophy ? Divers. Phylogeny ences of Established Trees on Mycorrhizas, Nutri- Evol. Monocotyledons. tion, and Growth of Quercus Rubra Seedlings. Jacquemyn, H., Waud, M., Brys, R., Lallemand, F., Courty, Ecol. Monogr., 72, 505–521. P.-E., Robionek, A., et al. (2017). Mycorrhizal Egger, K.N. & Hibbett, D.S. (2004). The evolutionary im- Associations and Trophic Modes in Coexisting plications of exploitation in mycorrhizas. Can. J. Orchids: An Ecological Continuum between Bot., 82, 1110–1121. Auto- and Mixotrophy. Front. Plant Sci., 8. Eissenstat, D.M., Kucharski, J.M., Zadworny, M., Adams, Jacquemyn, H., Waud, M., Merckx, V.S.F.T., Lievens, B. T.S. & Koide, R.T. (2015). Linking root traits to & Brys, R. (2015). Mycorrhizal diversity, seed nutrient foraging in arbuscular mycorrhizal trees germination and long-term changes in population in a temperate forest. New Phytol., 208, 114–124. size across nine populations of the terrestrial or- Epihov, D.Z., Batterman, S.A., Hedin, L.O., Leake, J.R., chid Neottia ovata. Mol. Ecol., 24, 3269–3280. Smith, L.M. & Beerling, D.J. (2017). N2-fixing Kiers, E.T., Duhamel, M., Beesetty, Y., Mensah, J.A., tropical legume evolution: a contributor to en- Franken, O., Verbruggen, E., et al. (2011). Recip- hanced weathering through the Cenozoic? Proc R rocal Rewards Stabilize Cooperation in the My- Soc B, 284, 20170370. corrhizal Symbiosis. Science, 333, 880–882. Frederickson, M.E. (2017). Mutualisms Are Not on the Kiers, E.T. & Heijden, M.G.A. van der. (2006). Mutualistic Verge of Breakdown. Trends Ecol. Evol., 32, 727– Stability in the Arbuscular Mycorrhizal Symbio- 734. sis: Exploring Hypotheses of Evolutionary Coop- Freschet, G.T., Swart, E.M. & Cornelissen, J.H.C. (2015). eration. Ecology, 87, 1627–1636. Integrated plant phenotypic responses to con- Klein, T., Siegwolf, R.T.W. & Korner, C. (2016). Below- trasting above- and below-ground resources: key ground carbon trade among tall trees in a temperate forest. Science, 352, 342–344. How mycorrhizal symbiosis shaped Angiosperm evolution —8/10

Kong, D., Wang, J., Zeng, H., Liu, M., Miao, Y., Wu, H., Rodriguez, R.J., White Jr, J.F., Arnold, A.E. & Redman, et al. (2017). The nutrient absorption–transporta- R.S. (2009). Fungal endophytes: diversity and tion hypothesis: optimizing structural traits in ab- functional roles. New Phytol., 182, 314–330. sorptive roots. New Phytol., 213, 1569–1572. Roggy, J.-C. (1998). Contribution des symbioses fixatrices Krings, M., Taylor, T.N., Hass, H., Kerp, H., Dotzler, N. & d’azote à la stabilité de l’écosystème forestier tro- Hermsen, E.J. (2007). Fungal endophytes in a pical guyanais. Lyon 1. 400-million-yr-old land plant: infection pathways, Roskov, Y., Abucay, L., Orrell, T., Nicolson, D., Bailly, N., spatial distribution, and host responses. New Phy- Kirk, P.M., et al. (2017). Species 2000 & ITIS Cat- tol., 174, 648–657. alogue of Life. Species 2000 Nat. Available at: Kubisch, P., Hertel, D. & Leuschner, C. (2015). Do ecto- www.catalogueoflife.org/col. Last accessed . mycorrhizal and arbuscular mycorrhizal temper- Scheublin, T.R. & Van Der Heijden, M.G.A. (2006). Ar- ate tree species systematically differ in root order- buscular mycorrhizal fungi colonize nonfixing related fine root morphology and biomass? Front. root nodules of several legume species. New Phy- Plant Sci., 6. tol., 172, 732–738. Kursar, T.A. & Coley, P.D. (2003). Convergence in de- Schlaeppi, K., Luckerhoff, L., Heijden, M.G. van der, Log- fense syndromes of young leaves in tropical rain- testijn, R.S. van & Bruin, S. de. (2015). A wide- forests. Biochem. Syst. Ecol., Proceedings of the spread plant-fungal-bacterial symbiosis promotes Phytochemistry and Legume/Animal Interaction plant biodiversity, plant nutrition and seedling re- Symposia held at the 4th International Legume cruitment. ISME J., 10, 389. Conference in Canberra, Australia, 2-6 July 2001, Schöll, L. van, Kuyper, T.W., Smits, M.M., Landeweert, 31, 929–949. R., Hoffland, E. & Breemen, N. van. (2008). Leake, J.R. (1994). The biology of myco-heterotrophic Rock-eating mycorrhizas: their role in plant nutri- (‘saprophytic’) plants. New Phytol., 127, 171– tion and biogeochemical cycles. Plant Soil, 303, 216. 35–47. Leifheit, E.F., Veresoglou, S.D., Lehmann, A., Morris, Shefferson, R.P., Taylor, D.L., Weiß, M., Garnica, S., E.K. & Rillig, M.C. (2014). Multiple factors influ- McCormick, M.K., Adams, S., et al. (2007). The ence the role of arbuscular mycorrhizal fungi in evolutionary history of mycorrhizal specificity soil aggregation—a meta-analysis. Plant Soil, among lady’s slipper orchids. Evolution, 61, 374, 523–537. 1380–1390. Mariotte, P., Canarini, A. & Dijkstra, F.A. (2017). Stoichi- Smith, S.E. & Read, D. (2008a). 2 - Colonization of roots ometric N:P flexibility and mycorrhizal symbiosis and anatomy of arbuscular mycorrhizas. In: My- favour plant resistance against drought. J. Ecol., corrhizal Symbiosis (Third Edition). Academic 105, 958–967. Press, London, pp. 42–90. Marschner, H. & Dell, B. (1994). Nutrient uptake in my- Smith, S.E. & Read, D. (2008b). 13 - Mycorrhizas in achlo- corrhizal symbiosis. Plant Soil, 159, 89–102. rophyllous plants (mycoheterotrophs). In: Mycor- Martin, F.M., Uroz, S. & Barker, D.G. (2017). Ancestral rhizal Symbiosis (Third Edition). Academic Press, alliances: Plant mutualistic symbioses with fungi London, p. 458–XV. and bacteria. Science, 356, eaad4501. Sprent, J.I., Ardley, J. & James, E.K. (2017). Biogeography Merckx, V. & Freudenstein, J.V. (2010). Evolution of my- of nodulated legumes and their nitrogen-fixing coheterotrophy in plants: a phylogenetic perspec- symbionts. New Phytol., 215, 40–56. tive. New Phytol., 185, 605–609. Sprent, J.I. & James, E.K. (2007). Legume Evolution: Moënne-Loccoz, Y., Mavingui, P., Combes, C., Normand, Where Do Nodules and Mycorrhizas Fit In? Plant P. & Steinberg, C. (2015). Microorganisms and Physiol., 144, 575–581. Biotic Interactions. In: Environmental Microbiol- Taylor, D.L., Hollingsworth, T.N., McFarland, J.W., Len- ogy: Fundamentals and Applications. Springer, non, N.J., Nusbaum, C. & Ruess, R.W. (2014). A Dordrecht, pp. 395–444. first comprehensive census of fungi in soil reveals Müller, C., Persicke, M., Baier, M.C. & Schweiger, R. both hyperdiversity and fine-scale niche partition- (2014). High specificity in plant leaf metabolic re- ing. Ecol. Monogr., 84, 3–20. sponses to arbuscular mycorrhiza. Nat. Commun., Tedersoo, L., Bahram, M., Põlme, S., Kõljalg, U., Yorou, 5, 3886. N.S., Wijesundera, R., et al. (2014). Global diver- Omacini, M., Eggers, T., Bonkowski, M., Gange, A.C. & sity and geography of soil fungi. Science, 346, Jones, T.H. (2006). Leaf endophytes affect mycor- 1256688. rhizal status and growth of co-infected and neigh- Tedersoo, L., Bahram, M., Toots, M., Diédhiou, A.G., Hen- bouring plants. Funct. Ecol., 20, 226–232. kel, T.W., Kjøller, R., et al. (2012). Towards Redecker, D., Kodner, R. & Graham, L.E. (2000). Gloma- global patterns in the diversity and community lean Fungi from the Ordovician. Science, 289, structure of ectomycorrhizal fungi. Mol. Ecol., 21, 1920–1921. 4160–4170. Rillig, M.C. & Mummey, D.L. (2006). Mycorrhizas and Valverde-Barrantes, O.J., Freschet, G.T., Roumet, C. & soil structure. New Phytol., 171, 41–53. Blackwood, C.B. (2017). A worldview of root How mycorrhizal symbiosis shaped Angiosperm evolution —9/10

traits: the influence of ancestry, growth form, climate and mycorrhizal association on the functional trait variation of fine-root tissues in seed plants. New Phytol., 215, 1562–1573. Valverde-Barrantes, O.J., Horning, A.L., Smemo, K.A. & Blackwood, C.B. (2016). Phylogenetically struc- tured traits in root systems influence arbuscular mycorrhizal colonization in woody angiosperms. Plant Soil, 404, 1–12. Wagg, C., Bender, S.F., Widmer, F. & Heijden, M.G.A. van der. (2014). Soil biodiversity and soil community composition determine ecosystem multifunc- tionality. Proc. Natl. Acad. Sci., 111, 5266–5270. Wang, B. & Qiu, Y.-L. (2006). Phylogenetic distribution and evolution of mycorrhizas in land plants. My- corrhiza, 16, 299–363. Weemstra, M., Mommer, L., Visser, E.J.W., van Ruijven, J., Kuyper, T.W., Mohren, G.M.J., et al. (2016). Towards a multidimensional root trait framework: a tree root review. New Phytol., 211, 1159–1169. Wehner, J., Antunes, P.M., Powell, J.R., Mazukatow, J. & Rillig, M.C. (2010). Plant pathogen protection by arbuscular mycorrhizas: A role for fungal diversity? Pedobiologia, 53, 197–201. Wilkinson, D.M. (1997). The Role of Seed Dispersal in the Evolution of Mycorrhizae. Oikos, 78, 394–396. Wilkinson, D.M. & Sherratt, T.N. (2001). Horizontally ac- quired mutualisms, an unsolved problem in ecology? Oikos, 92, 377–384. Wu, Q.-S., He, X.-H., Zou, Y.-N., Liu, C.-Y., Xiao, J. & Li, Y. (2012). Arbuscular mycorrhizas alter root system architecture of Citrus tangerine through regu- lating metabolism of endogenous polyamines. Plant Growth Regul., 68, 27–35. Wurst, S., Dugassa-Gobena, D., Langel, R., Bonkowski, M. & Scheu, S. (2004). Combined effects of earth- worms and vesicular–arbuscular mycorrhizas on plant and aphid performance. New Phytol., 163, 169–176. Yang, G., Liu, N., Lu, W., Wang, S., Kan, H., Zhang, Y., et al. (2014). The interaction between arbuscular mycorrhizal fungi and soil phosphorus availability influences plant community productivity and ecosystem stability. J. Ecol., 102, 1072–1082. How mycorrhizal symbiosis shaped Angiosperm evolution —10/10