Phytopathogenic fungi and oomycetes cause many highly destructive plant diseases, often with severe economic consequences for producers (Meng et al., 2009). While they are taxonomically distinct, mechanisms of pathogenesis are often similar between the two groups due to shared morphological and physiological traits (Meng et al., 2009). When microbial infection and insect herbivory occur at the same time, interspecific interactions that alter the individual effects of either organism on plant performance can occur. These interactions may take place directly or indirectly (Figures 1A–C) (Hatcher, 1995). Direct interactions occur when the ability of one organism to access plant resources is altered by another, without any influence from the plant itself (Figure 1B). Indirect interactions include those in which one organism induces a physical or physiological change in the plant that modifies its response to other organisms (Figure 1C). Indirect interactions are therefore plant-mediated. In either case, tripartite interactions between plants, pathogens and insects can complicate diagnoses and make infestation patterns and yield loss difficult to predict (Pieterse and Dicke, 2007). Despite growing recognition of the relationship between plant-associated insects and fungal or oomycete pathogens (see Hatcher, 1995; Hatcher and Ayres, 1997; Fournier et al., 2006; Stout et al., 2006), our understanding of the interface between entomological and pathological research remains limited. Moreover, existing research is heavily biased toward insects and pathogens that attack aerial parts of the plant. Far less is known about how these organisms interact belowground, as subterranean interactions are difficult to observe. As a result, there is a substantial gap in our understanding of how interspecific interactions between soilborne plant pathogens and root-feeding insects can impact pest population dynamics and plant development. The identification of belowground interactions is essential, both to our understanding of rhizosphere ecology and to inform effective management strategies in agricultural crops. The purpose of this review is to discuss currently available information regarding direct and indirect interactions between root-feeding insects, fungal and oomycete pathogens, and plants. In doing so, major gaps in our understanding of these interactions will be identified as potential areas for future research. Only interactions involving phytopathogenic microbes will be considered, as the association of root-feeding insects with plant mutualists and symbionts has been reviewed in detail elsewhere (Johnson and Rasmann, 2015).

Root-associated insects and phytopathogens often attack plants simultaneously, inevitably leading to direct insect–microbe interactions. Currently, there are two main avenues of research regarding direct interactions of rhizophagous insects with soilborne fungal and oomycete plant pathogens: (1) pathogen colonization of insect feeding injury, and (2) insects as vectors of root pathogens. As a result, there are several well-established examples of insects that facilitate the development of root diseases, and these will be discussed in more detail below. Less is known about how fungi and oomycetes directly impact the population dynamics of insects in soil, despite evidence that aboveground pathogens can influence insect development and mortality. For example, the European grapevine moth Lobesia botrana Den. & Schiff. (Lepidotera: Tortricidae) interacts mutualistically with the fungal pathogen Botrytis cinerea Pers. (Mondy et al., 1998). Larvae act as vectors of B. cinerea, and ingestion of the fungus increases the survival and development of larvae and the fecundity of adult insects (Fermaud and Le Menn, 1992; Mondy et al., 1998; Mondy and Corio-Costet, 2004). Evidence of similar interactions occurring below the soil is largely lacking. Clover root borers (Hylastinus obscurus Marsham; Coleoptera: Scolytidae) show a clear preference for red clover roots infected with fungal pathogens over healthy roots (Leath and Byers, 1973). Additionally, certain phytopathogenic fungi such as Fusarium and Alternaria spp. are known to produce mycotoxins with insecticidal properties (Gupta et al., 1991; Abbas and Mulrooney, 1994; Logrieco et al., 1998). However, the significance of these interactions on the performance of root-feeding insects has not been determined.

The role of plants in mediating interactions between insects and phytopathogens has become a subject of interest relatively recently, and it has rapidly become clear that plant responses to attack have important impacts on the population dynamics of both insects and phytopathogens. Our current understanding of plant-mediated insect–phytopathogen interactions is based mainly on aboveground interactions. Significant progress has also been made in defining interactions between above- and belowground herbivores and pathogens that are mediated by plant defense responses (see reviews by Hatcher, 1995; Van der Putten et al., 2001; Bezemer and van Dam, 2005; Stout et al., 2006). Plants similarly modify interactions between pathogens and insects in the rhizosphere, but the mechanisms behind these interactions are often poorly understood. Root exudates such as ions, oxygen, water, enzymes, mucilage, and primary and secondary metabolites accumulate in the rhizosphere and can have important roles in plant biological processes, including defense against pathogens and insects (Bais et al., 2006). Sufficient data to draw definitive conclusions is lacking, but evidence suggests that root exudates may also play an important role in mediating interactions between insects and phytopathogens in soil. Stutz et al. (1985) found that wounded alfalfa and red clover roots induced the production of distributive hyphae rather than chlamydospores in some species of Fusarium, accelerating fungal penetration and colonization near, but not in, injured tissue. Exudates from injured roots may have augmented the nutritional status of the rhizosphere, thus favoring pathogen growth (Stutz et al., 1985). This is consistent with reports that the weevil Diaprepes abbreviatus L. (Coleoptera: Curculionidae) promotes infection of citrus roots by Phytophthora spp. in tissues spatially separated from feeding wounds (Rogers et al., 1996; Graham et al., 2003). Increased decay in tissues distal to weevil damage was attributed to sugars and other nutritional compounds leaking into the rhizosphere from damaged cells (Graham et al., 2003). In contrast, herbivory of geranium roots by fungus gnat larvae increased seedling resistance to infection by Pythium aphanidermatum, possibly due to an induced defense response (Braun et al., 2009).

Further investigation of how root chemical signals influence the population dynamics of multiple organisms will be critical to our understanding of plant defense and rhizosphere ecology. Plant-mediated interactions between belowground insects and phytopathogens are undoubtedly complex, and our picture of how root exudates influence these interactions is largely incomplete. Chemical signals from roots attract and repel insects and pathogens (Bais et al., 2006), alter nutritional quality (Stutz et al., 1985; Graham et al., 2003), and attract natural enemies of herbivores in sophisticated tritrophic interactions similar to those occurring aboveground (Rasmann et al., 2005). Future research is required to reveal how root exudates influence the population dynamics of insects, phytopathogenic fungi and oomycetes. This will be vital in increasing our understanding of how plants defend themselves against diverse organisms in the rhizosphere.

Synergistic interactions between herbivores and phytopathogens can be problematic in agricultural crops, but have potential for use in biocontrol. Insects and fungi are the primary organisms used in weed biocontrol, but are rarely used in combination (Hatcher and Paul, 2001; Caesar et al., 2010). Synergisms between root-feeding insects and soilborne pathogens have been identified in several species of rangeland weeds. Spotted and diffuse knapweed, Centaurea maculosa Lam. and C. diffusa Lam., are pervasive across the western United States and Canada (Caesar et al., 2002). Within their native Eurasian range, populations are apparently kept in check by a combination of root herbivory and fungal infection. Several highly virulent strains of Fusarium spp. are associated with the feeding wounds of Cyphocleonus (Coleoptera: Curculionidae) and Agapeta spp. (Lepidoptera: Cochylidae) in their larval stage (Caesar et al., 2002). Likewise, populations of Fusarium spp. associated with leafy spurge (Euphorbia esula/virgata L.) increase in density around roots attacked by Aphthona spp. (Coleoptera: Chrysomelidae), Chamaesphecia spp. (Lepidoptera: Sesiidae), and Oberea erythrocephala (Schrank) (Coleoptera: Cerambycidae) (Caesar, 2003). Greenhouse experiments showed that E. esula/virgata exposed to combinations of Rhizoctonia solani Kühn, F. oxysporum (Schlect) Snyd. & Hans., and Aphthona spp. had higher disease levels than any single inoculation (Caesar, 2003). The invasive perennial weed Lepidium draba L. ssp. draba sp. [ = Cardaria draba (L.) Desv.] is associated with the gall-forming weevil Ceutorhynchus assimilis Paykull (Coleoptera: Curculionidae) (Fumanal et al., 2004; Caesar et al., 2010). Some populations within this species demonstrate a high level of host-specificity to L. draba as larvae (Fumanal et al., 2004). Ceutorhynchus assimilis galls are frequently colonized by pathogenic Rhizoctonia and Fusarium spp., suggesting that galling promotes fungal infection (Caesar et al., 2010). Many of the insect and pathogen species associated with rangeland weeds also have narrow host ranges, making them suitable for consideration as biocontrol agents (Caesar et al., 1999, 2002, 2010). Synergistic interactions between insects and plant pathogens can enhance weed management and provide an effective alternative to herbicides in multiple plant species. Despite this, utilization of these interactions in weed biocontrol remains rare, thus emphasizing the importance of identifying the impact of insect–pathogen interactions on plant populations.

Root-feeding insects are closely associated with the soilborne microorganisms that colonize their mutual plant host. Direct and indirect interactions between insects and pathogenic fungi or oomycetes can increase the incidence and severity of host injury, influencing plant performance and mortality. Tripartite interactions have been identified in several agroecosystems, but it is likely that many more have gone undetected. Belowground interactions are difficult to observe, and have largely been ignored despite the potential impacts of root damage on overall plant functioning. Soilborne fungi and oomycete pathogens may interact directly with root-feeding insects by colonizing injured plant tissue. Some insects vector soilborne pathogens via internal or external transport, which may be of particular importance in commercial greenhouses and other closed environments. Plant responses to attack appear to be important factors in defining belowground insect–fungi interactions, but our mechanistic understanding of these interactions remains limited. Identifying and understanding tripartite interactions between plants, insects and pathogens will therefore be an important step in furthering our understanding of plant defense responses to belowground threats. An emphasis on integrating entomological and pathological research will allow for more accurate predictions of pest-related crop damage, and will guide effective monitoring and crop protection strategies. Finally, interactions between insects and microorganisms have the potential for use in biocontrol. Synergism between herbivores and pathogens of rangeland weeds suggest that suppression of invasive plant species can be increased by coordinating the release of multiple antagonists. Whether they appear as beneficial organisms or damaging pests, root associated insects, fungi and oomycetes have important consequences for plant health, and the interactions between these organisms are worthy of further study.

TW collected literature and wrote the paper. SC and HC provided revisions, and all three authors approved this mini review for publication.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewer SS and handling Editor declared their shared affiliation.