Context-dependency in plant response to mycorrhizal fungi: A meta-analysis of the mutualism-parasitism continuum

Background/Question/Methods Symbiotic associations between microbes and terrestrial plants have a long evolutionary history, are ubiquitous, and can have dramatic consequences for plant growth, local biodiversity, and ecosystem function.  Research on the symbiosis between plants and mycorrhizal fungi has grown dramatically in recent years, generating a complex body of empirical results demonstrating that the symbiosis spans the continuum from mutualism to parasitism of host plants by their symbionts.  Numerous hypotheses have been proposed to explain the wide variation in plant response to mycorrhizal fungi, but these hypotheses are usually tested either in isolation using single experiments, or in research syntheses using single-factor statistical models.  We compiled a database of more than 700 mycorrhizal inoculation experiments conducted in the laboratory (greenhouse or growth chamber) and field and used information-theoretic criteria to assess the relative importance of different explanatory variables in a series of multi-factor meta-analyses of plant response to inoculation with mycorrhizal fungi.  We assessed the relative importance of factors in four categories: host plant identity and characteristics, fungal identity, biotic factors in the soil, and experimental practices. 

Results/Conclusions The most striking result was that plant response to mycorrhizal fungi was substantially more positive when the soil community was more complex, specifically when non-mycorrhizal microbes were added to soils, when soils were not sterilized, and when multiple mycorrhizal fungal species were used in the inoculation.  In addition, plant functional group was consistently an important explanatory variable, with C₄ grasses, non-leguminous forbs, and non-leguminous woody plants exhibiting more positive response to mycorrhizal inoculation than other functional groups.  Fungal genera also differed in their effects on plants.  When accounting for other explanatory factors, plant response was not strongly affected by whether experiments took place in field or laboratory settings, did not differ between experiments utilizing AM vs. EM fungi, and were not influenced by the root:shoot ratios of plants.  These results emphasize the importance of the biotic context for ecological outcomes in mycorrhizal symbioses, including the host plant characteristics, the fungal symbionts, and the soil biotic milieu.