Our study shows that pear plants respond to herbivore attack, specifically by psylla adults and leafroller larvae, by modifying their volatile profiles. Further, this change in volatile emissions shows plasticity in the chemistry of induced compounds, depending on the attacking herbivore. These induced volatiles most likely explain the behavioral response of predatory C. externa larvae, which are not only attracted to herbivore-damaged plants, but also show differential attraction to odors from psylla-damaged plants to which more predatory larvae respond and they do so more rapidly.
As shown by multivariate and ordination analyses of the volatile blends, undamaged pear plants emit odors that differ from those of plants damaged by either psylla adults or leafroller larvae. Delving into the specific volatile compounds that may explain these differences, the only compound that showed a significant increase in the blend after damage by both herbivores was the sesquiterpene caryophyllene. This compound has been reported as induced in several plants attacked by chewing herbivores, miners and sap-phloem feeders (Ayelo et al. 2021b; Conti et al. 2008; El-Sayed et al. 2018; Kelly et al. 2014; Peterson et al. 2022; Silva et al. 2017; Yang et al. 2021). In addition, caryophyllene has been shown to attract various natural enemies (Jayanthi et al. 2021; Riffel et al. 2021; Zhang et al. 2020). Although this compound is usually regarded as beneficial to the plant, it has been reported as an attractant to Cacopsylla picta toward plants affected by the phytoplasma Candidatus phytoplasma mali, thereby potentially contributing to the spread of the pathogen (Mayer 2008a, b and 2011). In our study system, caryophyllene represented a very small contribution to the volatile blends, not surpassing 1% of the volatiles in both undamaged and damaged plants.
Most damage-induced volatiles in pear plants were different in plants damaged by either herbivore. Plants damaged by C. bidens emitted blends enriched in aliphatic aldehydes, a result in line with previous studies (Scutareanu et al. 1997). Nonanal and decanal were the predominant induced compounds in psylla-damaged plants, becoming major components of the blend. As reported by several authors, these aldehydes are induced by herbivory in several plants and are attractive to parasitoids and predators (Badra et al. 2021; Birkett et al. 2003; Dicke 1999; Du et al. 2022; Rott et al. 2005; Yu et al. 2008). These induced aldehydes may also interact directly with herbivores; they can either be attractive in association with the search for oviposition sites (Borrero-Echeverry et al. 2015; Sarkar et al. 2016), or they may be repellent as observed for Spodoptera litura, for which food plant preferences correlate negatively with plant nonanal emissions (Du et al. 2022). An increase in aldehyde emissions in psylla-damaged plants may also influence plant-plant communication and prime the activation of pathogen defenses in neighboring plants (Brambilla et al. 2022; Yi et al. 2009; Yi et al. 2010). Taking into account that Cacopsylla species are important phytoplasma vectors, it is conceivable that psylla-induced volatiles such as aliphatic aldehydes may both attract natural enemies and elicit a defensive response in nearby plants (Görg et al. 2021; Gross et al. 2019; Rid et al. 2016).
In the case of pear plants damaged by the leafroller A. sphaleropa, the homoterpene (E)-4,8-dimethylnona-1,3,7-triene (DMNT) and the aromatic benzeneacetonitrile were the most relevant induced volatiles. Benzeneacetonitrile increased from negligible amounts to become the major volatile component emitted by leafroller-damaged plants. This compound has been reported in the odor blend of melons (Majithia et al. 2021) and as part of the induced response of hybrid aspen after exposure to odors from damaged neighboring branches (Li and Blande 2017), as well as of silver birch under direct herbivory (Koski et al. 2015). The role of benzeneacetonitrile as an attractant of natural enemies has not yet been determined, requiring more bioassays specifically directed to this end.
The homoterpene DMNT, in turn, has been reported in several systems as an herbivore-induced plant volatile (Arimura et al. 2000; Birkett et al. 2003; Degenhardt and Gershenzon 2000; Takabayashi et al. 1994), including both leaf chewers and sap feeders. Several studies have demonstrated that this compound serves as a cue for natural enemies, as an activator of plant defenses and as a deterrent for different herbivores (Arimura et al. 2000; Ayelo et al. 2021b; Borrero et al. 2015; Dicke 1994; James 2003; Takabayashi et al. 1994). Recent research has revealed its role in luring different lepidopteran pests, to the point that traps have been tested with the inclusion of this compound to enhance lepidopteran catches (Knight et al. 2011, 2019; Preti et al. 2021). This compound was not induced by C. bidens damage, a result that is in line with previous studies in pear plants (Scutareanu et al. 1997). Another compound that showed a significant increase in the emission from leafroller-damaged plants was cis-ꞵ-ocimene. This compound is well known to mediate the attraction of natural enemies of various herbivores, playing an important role in the indirect defense of plants (Cascone et al. 2015; Farré-Armengol et al. 2017; Mohammed et al. 2020). Several studies have shown that this compound plays a primary role in the communication between plants, leading to an increase in the synthesis of methyl jasmonate in plant tissues and providing greater resistance against pests and diseases (Arimura et al. 2012; Godard et al. 2008; Howe and Jander 2008; Muroi et al. 2011).
Green leaf volatiles showed a significant decrease in blend proportions in leafroller-damaged plants, and a similar trend in psylla-damaged plants. These compounds are expected to increase as a result of herbivore damage, and its activity in eliciting predator responses has been demonstrated in several systems (Dudareva et al. 2006; Gatehouse 2002; Moayeri et al. 2007; Takabayashi and Shiojiri 2019; Van Poecke et al. 2001; Weissbecker et al. 1999; Yu et al. 2008). This result may be explained by the time-window used in our experiments; GLVs are the first compounds to be released when a plant is attacked, while the other compounds are induced only after 24 hours (Dudareva et al. 2006; Heil 2008; Pichersky et al. 2006; Turlings et al. 1998). In our study, VOC collections were carried out between 24 and 48 hours post-damage, a period in which GLVs most likely decrease their prevalence in the volatile blend from damaged plants (Heil 2008; Turlings et al. 1998).
Our olfactometer bioassays showed that odors emitted by pear plants attacked by psylla adults or leafroller larvae are attractive to green lacewing C. externa larvae. These odors were contrasted with those from undamaged plants, so even though no bioassays were performed with pure compound mixtures, it seems reasonable to assume that the behavioral responses of C. externa larvae are due to induced volatiles emitted by the damaged plants. The chrysopid larvae used in our study were reared on lepidopteran eggs and artificial diet based on chicken liver, with no prior contact with plant volatiles, therefore discarding the possibility of associative learning by previous experience (de Oliveira et al. 2019; Drukker et al. 2000). Conversely, it is possible that C. externa, as a generalist predator, has innate preference for plant volatiles associated with various herbivores, such as caryophyllene in our study. The role of this compound in attracting chrysopids was reported several decades ago in trapping studies in which high captures of lacewings were observed in traps baited with this compound (Flint et al. 1979). Phenylacetaldehyde has also been shown to attract chrysopids (Tóth et al. 2006; Tóth et al. 2009), but in our study was only slightly and not significantly increased in leafroller-damaged plants.
Even though a clear attraction of C. externa was observed towards volatiles induced by A. sphaleropa against undamaged plants, when contrasted with psylla-damaged plants, C. externa larvae clearly preferred the latter. Furthermore, odors from psylla-damaged plants resulted in faster response times and a lower rate of non-responsive chrysopid larvae. Such reduction in response times to locate the prey’s habitat plays a primary role in the foraging efficiency of natural enemies, and consequently in their fitness, since the energy and predation risks associated with searching are reduced (Bell 1990; Mills and Heimpel 2018). Our previous studies have shown that C. externa readily preys on C. bidens. (Valle et al. 2022), and our current results add further evidence that C. bidens induces volatiles in pear plants that promote the predator’s activity. To our knowledge, no data are available on the response of chrysopid larvae to plant aliphatic aldehydes, so bioassays with the specific psylla-induced aldehyde volatiles are warranted to continue these studies.
All in all, our results support the notion that C. externa larvae exploit HIPV cues to locate pear plants on which potential prey are present, and that they do so with some degree of prey specificity. These results provide information towards the development of semiochemicals as tools to implement a biological control plan of C. bidens in pear orchards, either conservative or augmentative. In any scenario, attracting natural enemies and retaining them in the agro-ecosystem are key elements in which semiochemicals may be incorporated.