Common mycorrhizal networks (CMNs) can mediate interactions among host plants, including interplant communication. CMN-mediated interplant communication has been demonstrated to induce defense responses in undamaged plants connected to a damaged plant. Undamaged plant hosts connected to a CMN can upregulate anti-herbivore defenses and reduce potential damage from herbivore attack. However, it is not known whether defense cues can be propagated through multiple plants connected to a CMN. Therefore, we tested the hypothesis that defense cues will propagate through multiple plants connected to a CMN. To test this hypothesis, we compared gene expression of proteinase inhibitor II (PI-II) between undamaged plants connected directly and indirectly (receivers) via a CMN to a plant experiencing herbivore damage (emitter).
The study was conducted using tomato (Solanum lycopersicum c.v. Castlemart) seedlings connected through a common mycorrhizal network by the mycorrhizal fungus Rhizophagus intraradices and grown in experimental mesocosms. Mesocosms were designed with a combination of mesh screens and partitions to impose a plant-mycorrhizal network architecture such that defense cues could propagate only from damaged emitter plants to primary and secondary plant receivers. To control for mycorrhizal signaling from damaged emitters, each mesocosm included parallel chambers in which the CMN was severed between the emitter and primary control receiver. To elicit defense cues in emitter plants, larval Manduca sexta was applied at a density of 12-15 larvae per plant. All plants were covered by transparent plastic bags to prevent interplant communication via airborne pathways (i.e. volatile organic copmounds).
To measure defense response in emitter and receiver seedlings and controls, gene expression of proteinase inhibitor II (PI-II) was quantified using RT-qPCR. Expression of PI-II among treatments was calculated using the ∆∆Cq method. Mean expression of PI-II was compared between treatment and control plants to identify receivers with upregulated defenses.
Results/Conclusions
Upregulation of PI-II was observed in all emitters but not in the mean expression of primary receivers nor secondary receivers. However, although the mean expression of PI-II was not significantly different in primary receivers, half of the replicate primary receivers were observed to have upregulation of PI-II. This observation suggests that the other half of primary receivers may not have had a CMN connection to emitters. Upregulation of PI-II was not observed in any control plants. Although our results suggest that defense cues did not propagate to secondary receivers, defense cues from primary receivers may not have been sufficient to elicit upregulation in secondary receivers.