Aphid Protected from Pathogen by Endosymbiont

See allHide authors and affiliations

Science  16 Dec 2005:
Vol. 310, Issue 5755, pp. 1781
DOI: 10.1126/science.1120180


Aphids are associated with several facultative bacterial endosymbionts that may influence their interactions with other organisms. We show here that one of the three most common facultative symbionts of pea aphid (Acyrthosiphon pisum), the bacterium Regiella insecticola, has a major effect on host resistance to a fungal pathogen. Experimental establishment of the bacterium in uninfected aphid clones led to higher survival after fungal attack. The bacteria also increased the aphid's inclusive fitness, because the presence of the symbiont reduced the probability of fungal sporulation on aphid cadavers, hence lowering the rate of transmission of the disease to nearby related aphids.

Many invertebrates have intimate associations with bacterial symbionts, and over the past few years, several have been shown to influence their hosts' ability to defend against natural enemies (1, 2). Aphids possess a range of facultative bacterial endosymbionts that may help in defense against parasitoids (2) and influence the aphids' ability to use different plant species (3). We show here that one of the common facultative symbionts of pea aphid (Acyrthosiphon pisum), the bacterium Regiella insecticola (synonyms: U-type, PAUS) (4), has a major effect on host resistance to a fungal pathogen and lowers the rate of transmission of the fungus. Regiella is vertically transmitted during both asexual and sexual reproduction.

A vertically transmitted symbiont (that does not manipulate sex ratio) or one with low levels of horizontal transmission will eventually be lost from a population if carrying it imposes a cost on the host. This loss has led to a search for beneficial effects of the symbiont. By experimentally introducing symbionts, it has been shown that both Hamiltonella defensa and Serratia symbiotica increased resistance to an aphid parasitoid (2), whereas Regiella improved the performance of one aphid clone on one of its host plants (3).

A major pathogen of aphids is the Entomophthorales fungus Pandora (Erynia) neoaphidis. Infectious spores of the fungus germinate on the aphid cuticle, and mycelia penetrate and fill the aphid's body cavity, leading to death within a few days. Spores are then produced on the cadaver's surface. We previously found a correlation across pea aphid clones between the possession of Regiella and resistance to Pandora (6). To test whether carriage of Regiella enhances resistance, we took five facultative symbiont-free pea aphid clones and established novel infections by injecting hemolymph from pea aphids that carried only Regiella. After several generations, the aphids were tested for resistance to Pandora (5).

Across all five aphid clones, the association with Regiella improved the aphids' ability to survive exposure to the fungal pathogen (Fig. 1A). Symbiont presence (treatment) increased the log odds of survival by 1.8 ± 0.2 (SE), which is highly significant (χ12 = 173.47, P < 0.001). There were significant differences among clones in their susceptibility to fungi (χ42 = 22.39, P = 0.02), although all responded similarly to Regiella (clone × treatment, χ42 = 3.46, P = 0.78).

Fig. 1.

The facultative endosymbionts R. insecticola increase resistance to the entomopathogenic fungus P. neoaphidis. Five aphid genotypes were artificially infected with Regiella. (A) All aphid genotypes had significantly increased survival. (B) Even when killed by P. neoaphidis, pea aphids harboring Regiella produced fewer sporulating cadavers. Means and SEs are shown.

The presence of the symbiont influenced whether the fungus would sporulate successfully on those aphids it had killed (Fig. 1B). Regiella reduced the log odds of successful sporulation by –2.7 ± 0.2 (SE) (χ12 = 423.21, P < 0.001). This effect differed significantly among aphid clones (clone, χ42 = 26.98, P < 0.001; clone × treatment, χ42 = 56.93, P < 0.001). The interaction between clone and treatment may represent differences among aphid genotypes in their response to Regiella or vice versa, because each aphid clone was injected with a different strain of Regiella. Failure to sporulate has no bearing on the fate of the individual insect; however, aphids reproduce clonally and are often wingless. Hence, nearby individuals are more likely to be genetically identical, and this will lead to an increase in the aphid's inclusive fitness.

Because Pandora is the most common aphid fungal pathogen, our results provide an explanation for Regiella's maintenance in pea aphid populations. But why do not all aphids carry Regiella? One possibility is that the advantage of disease resistance varies over time and space and, in Pandora's absence, carrying the symbiont is costly, although no costs to date have been identified. Regiella can influence performance on certain host plants (3), and, possibly, the net benefits of carrying the symbiont depend both on its effect on resource use and the risk of infection on different food plants. Such an effect might explain why pea aphid clones specialized on Trifolium both tend to harbor Regiella and show high resistance to Pandora (6, 7).

Supporting Online Material

Materials and Methods

SOM Text

References and Notes


View Abstract

Navigate This Article