Mycorrhizal fungi as drivers and modulators of terrestrial ecosystem processes

This is the authors' abstract. We don't add key points for this paper.

‘… integrating phosphorus (P) into existing paradigms of mycorrhizal symbioses in both P-limited and P-sufficient systems may further illuminate mechanisms by which mycorrhizal fungi either determine or respond to ecosystem properties.’ Within the last 5 years, there have been a number of key studies expanding upon Read's ideas, deepening our conceptual understanding of the ecosystem-level consequences of mycorrhizal fungi. In a cross-biome analysis, Averill et al. (2014) reported that soil carbon (C) per unit of soil nitrogen (N) is greater in temperate and boreal forests dominated by ECM and ERM plants compared to grasslands and tropical forests dominated by AM plants. Orwin et al. (2011) incorporated mycorrhizal fungi into biogeochemical models, and found that when inorganic N is scarce, the capacity for ECM and ERM fungi to take up organic N may increase soil C pools because it facilitates greater plant productivity and imposes N limitation on decomposers. To further understand the mechanism behind the accumulation of C in boreal forests, Clemmensen et al. (2013) analyzed bomb 14C to demonstrate that a substantial fraction of soil C is derived from roots and mycorrhizal fungi and thus, that ECM and ERM fungi play an important role in the development of soil C stocks. Together, these findings support the idea that mycorrhizal associations contribute to the emergence of patterns across ecosystems and provide the grounds for new predictive frameworks for C and nutrient cycling (e.g. Phillips et al., 2013). These new perspectives motivated our organized oral session at the Ecological Society of America's 2016 meeting, but several critical questions remain. If mycorrhizal symbioses are responsible for distinct ecosystem properties, they offer a simple classification system that could help in making biogeochemical predictions. Is the AM vs ECM/ERM dichotomy really that simple? In other words, is the AM symbiosis functionally distinct from the ECM and ERM symbioses across their respective phylogenetic breadths? The AM habit likely arose at a single time during the early evolution of both plants and fungi (Selosse et al., 2015); however, the ECM habit has evolved multiple times in both plants (at least 12 times; Maherali et al., 2016) and fungi (up to 80 times; Tedersoo & Smith, 2013). Despite the morphological consistency of the ECM structures formed by varying plant and fungal partners, we must be cautious in assuming functional consistency across independent evolutionary events, as each ECM plant and fungal clade may retain idiosyncratic attributes derived from their non-ECM ancestors. Due to their unique evolutionary histories, the three major types of mycorrhizal symbioses display an interesting mismatch in the phylogenetic diversity of plants and fungi associated with each: AM symbioses occur between a highly diverse suite of plants but a single clade of fungi (although both likely represent a single, ancient evolutionary event; Selosse et al., 2015), ECM symbioses occur between a limited, but phylogenetically widespread, set of both plants and fungi (representing multiple evolutionary events in each partner), and ERM symbioses arose once in plants but involve a phylogenetically diverse set of fungi (likely representing multiple evolutionary transitions in fungi; Fig. 1). Can these differences in evolutionary history and phylogenetic breadth help explain the role of mycorrhizal types in ecosystems? When considering mycorrhizal function, does variability in plant traits matter more than fungal traits in the AM symbiosis, and vice versa in the ECM and ERM symbioses? Moving forward, a consideration of plant lineage with broader sampling efforts (e.g. sampling outside the Pinales and Fagales for ECM plants, and considering AM plants in the Pinales), combined with an understanding of trait diversity within fungal communities, may help fill this knowledge gap. Considering the evolutionary transitions among ECM plants also creates an intriguing wrinkle in Read's neat depiction of the mycorrhizal distribution across biomes. In particular, there are several plant orders in which ECM-associated species occur exclusively in the tropics, which is contrary to the expectation of ECM lineages dominating the higher latitudes. While AM species dominate the tropical biome, ECM tree species can be locally aggregated even in warm lowland forests, including the Dipertocarps of Southeast Asia (i.e. Malvales), the non-nodulating group of Caesalpinioids of south America and Africa (i.e. Fabales), and south American species in the Polygonaceae and Nyctaginaceae (i.e. Caryophyllales) (Bâ et al., 2014). Perhaps not surprisingly, aggregations of tropical ECM trees can be associated with biogeochemical patterns more typical of those in temperate ecosystems, including dampened rates of decomposition and the accumulation of organic matter (Bâ et al., 2014); however, our current understanding of tropical biogeochemical cycles does not account for the phylogenetic diversity of ECM species across the biome. Organic matter abundances are distinct across biomes, but what stabilizes and allows for the persistence of this pattern during ecosystem development? Mycorrhizal symbioses and their ecosystem consequences are classic examples of complex adaptive systems, in which macroscopic patterns emerge from localized interactions between component parts. Such localized interactions are characterized by feedbacks (Levin, 1998), but what specific mycorrhizal feedbacks give rise to the predictable patterns in the balance of ecosystem C and N? While the geographical distribution of AM, ECM and ERM-dominated biomes implies the long-term stability of such feedbacks, it remains unclear whether mycorrhizal fungi drive ecosystem processes directly or if the mycorrhizal composition of a system is instead the response of the plant community itself to climatic or soil gradients. Positive biogeochemical feedbacks are often invoked in studies of mycorrhizal fungi, but positive feedbacks are unsustainable without negative feedbacks to keep systems in balance (e.g. Caruso & Rillig, 2011). For example, assuming that net primary productivity is similar between AM and ECM-dominated ecosystems within climate zones, larger soil C pools in ECM ecosystems may be the result of slower decomposition, or the rates of other C loss processes, but soil C cannot accumulate, nor can organic N be recycled, indefinitely. What then are the negative feedbacks that stabilize mycorrhizal effects on C? Is long-term organic matter accumulation kept in check by losses from fire and losses of dissolved organic C? Since C and N cycles are tightly interrelated, the answers may also lie in how AM and ECM-dominated ecosystems differ in N inputs and losses, and how N limitation persists at steady state (e.g. Menge, 2011). Furthermore, integrating phosphorus (P) into existing paradigms of mycorrhizal symbioses in both P-limited and P-sufficient systems may further illuminate mechanisms by which mycorrhizal fungi either determine or respond to ecosystem properties. Read's explanation for global mycorrhizal patterns and our continued pursuit to understand them requires melding evolutionary and ecosystem perspectives. How might we expand upon this to predict how mycorrhizal fungi mediate or create feedbacks in response to global change? A recent meta-analysis by Terrer et al. (2016) suggests that when soil N is strongly limiting, ECM plants may be more capable of sustaining increased productivity with elevated atmospheric CO2. Such a result appears consistent with distinct growth responses to anthropogenic N deposition between AM and ECM plants (Thomas et al., 2010). But are these patterns robust across the diversity of plants and fungi within each mycorrhizal association, or can they be explained simply by plant functional types or phylogeny? A consideration of mycorrhizal identity may also improve predictions for how individual plant species may shift in geographic distribution under future climate scenarios (Lankau et al., 2015). The success or failure of plant species migrations may depend on their mycorrhizal type and that of the existing community and the species they are replacing. If migrating species replace those of the same mycorrhizal type, are the biogeochemical properties of the ecosystem maintained? How much environmental force is necessary to induce a state change in the mycorrhizal identity of the ecosystem? Our ability to predict the future of terrestrial ecosystems may depend on our understanding of the role of mycorrhizal fungi in the evolution of biogeochemical cycles and, in turn, how biogeochemistry shaped the evolution of mycorrhizal symbioses. The authors thank all contributors of the 2016 Ecological Society of America organized oral session on mycorrhizal fungi and C. Phillips and M-A. Selosse for constructive comments.