Investigating the Disparity in Host Specificity between AM and EM Fungi: Lessons from Theory and Better-Studied Systems

Abstract

Over 80% of vascular plant species are associated with some type of mycorrhizal fungus (Trappe 1987). These associations are thought to be mutualistic in many cases, with the host plant exchanging photosynthate with the fungus in return for mineral nutrients (Harley and Smith 1983, Smith and Read 1997). The two most abundant and studied mycorrhizal types are ectomycorrhizal (EM) fungi and arbuscular mycorrhizal (AM) fungi. A generally agreed-upon phenomenon is that AM fungi are broad generalists with respect to the number of host plant species with which they can associate (Smith and Read 1997). Because of the difficulty of detecting variation in the relative abundances of AM fungal species on plant roots, as well as the difficulty of understanding the causes (environmental factors, host genotype, interspecific interactions) of such variation, there exist gaps in our knowledge about the extent to which AM fungal species associate with different plant species, especially in complex natural fungal communities. However, based on evidence accumulated so far from field and laboratory observations, AM fungi exhibit broad host ranges, and the rare exceptions to this (e.g., Graw et al. 1979) must not be allowed to cloud the observation that VA mycorrhizal fungi and their host plants have generally non-specific (Smith and Read 1997). In contrast, specificity among EM fungi ranges from very broad to extremely specific, with many EM fungi associating with only a single host plant species (Janos 1 980a, Alexander 1989, Harley 1989, Borowicz and Juliano 1991, Molina et al. 1992, Smith and Read 1997). Explanations for the disparity in host specificity between the two types of fungi have been mentioned briefly in other contexts, but have not been developed, tested, or compared with other potential explanations (e.g., Janos 1980b, 1983, Malloch et al. 1980, Connell and Lowman 1989, Molina et al. 1992). There is clearly a need to explicitly develop and test a set of hypotheses explaining the higher incidence of host-specificity among EM fungi than among AM fungi. The general topic of specificity is one of broad interest to ecologists and evolutionary biologists, as the degree of specificity in interactions is predicted to strongly influence the nature of evolution of those interactions (e.g., whether two species co-evolve) (Janzen 1980, Thompson 1994). As a result, it has received a great deal of attention both theoretically and empirically. Much of this work has focused on specificity in interactions between plants and their herbivores (e.g., Futuyma and Moreno 1988, Berenbaum 1996), pollinators (e.g., Kiester et al. 1984, Waser et al. 1996), and fruit dispersers (e.g., Jordano 1987, Fleming et al. 1993). The voluminous literature generated from investigations of these three types of interactions provides valuable information and predictions about specificity phenomena. Many of these predictions are particular to specific systems, but some are very general and could aid in attempts to understand specificity in mycorrhizal mutualisms. However, these results have not been utilized to help understand the disparity in host specificity between AM and EM fungi. Here I discuss three principles regarding the evolution of specificity that emerge from an examination of the plant-herbivore, plant-pollinator, and plant-fruit-disperser literature and other theoretical literature. These three principles suggest three hypotheses (not mutually exclusive) for the disparity in host specificity between AM and EM fungi, which I discuss and evaluate.