Discussion
Plants emit more volatiles during early stages of development than during mature stages because in order to grow and proliferate successfully, they must protect juvenile tissues, such as young leaves and buds, against herbivore attack (Meenaetalet al . 2017). This is consistent with our heat map of flower development (Figure S1), showing that various volatile compounds play significant roles in plant defence during early stages. Identification of gene function in the Japanese pepper (Zanthoxylum piperitum ) (Fujita et al. 2017) and inA. thaliana(Boachonet al. 2019) also showed that genes participating in defence are mainly active during early stages of development and in young tissues of plants. This is a time period when plants need to supply energy for growth and defensive strategies concurrently, which is difficult because each of these two pathways requires high energy expenditure (Meena et al et al . 2017). Collectively, these results indicate that genes and their products in the lavender bud stages likely function in plant defence.
Volatile compounds protect plants not only by repelling herbivores, but also by attracting the natural enemies of herbivores, forming a tritrophic interaction among plant-herbivores-carnivores. This provides means of natural biological control and may serve as a new approach to pest eradication, decreasing the use of insecticides, which are used excessively in modern agriculture and other industrial applications. Our results show that limonene and carveol, expressed in tobacco leaves, repel aphids and attract ladybugs. This has also been shown in soybeans (Glycine max ) and mint (ZHU et al . 2005; Togashi et al. 2019). Additionally, limonene attracts predatory mites,Phytoseiulus persimilis and Neoseiulus californicus(Togashi et al. 2019) and repels western flower thrips (Frankliniella occidentalis ) (Picardet al . 2012), indicating that limonene can defend a wide variety of plants (Smith et al . 2018). Combining carveol, pinene, and limonene is important in developing plant-derived insect repellents (Picard et al. 2012; Smithet al. 2018). However, other studies indicate that limonene promotes herbivore and pathogenic attack in the orange fruit (Rodrıguezet al . 2011). Moreover, Musca domestica L. catalyses limonene into carveol and carvone via endogenous P450, resulting in lower toxicity to flies (Rossiet al . 2013). This suggests that the same terpene may function differently among different species; this needs to be investigated in future studies.
Terpenes or terpenoids in plants are mainly catalysed by terpene synthases (TPSs). The sequences and functions of TPSs are redundant, causing difficulty in annotation and assignment even in Gossypiumgenomes (Huang et al . 2018). The similarity betweenLaTPS7and LaTPS8 is quite high, showing 77.61% identity; correspondingly, the products of these genes are highly redundant in vitro.However, both of LaTPS7 and LaTPS8 can catalyze GPP and NPP into various products, which shows the promiscuous and multi-substrate characteristics of TPSs (Major, 2012). Although most monoterpene synthases, including La TPS7 andLa TPS8, are located in the plastids, exceptions exist and can be located in the mitochondria and cytosol ( Yamaga et al . 1993; Aharoni et al . 2004; Lee & Chappell, 2008; Magnardet al. 2015). This suggests that there is a complex metabolic network of terpenes. Additionally, La TPS7 andLa TPS8 show different localization patterns in plastids. La TPS7 wrapped around chloroplasts, while La TPS8 distribution was punctate. Although reasons for these differences remain undetermined, these different patterns of localization may stem from different substrates in organelles. Moreover, several noncanonical metabolic pathways have been recently identified. For example, a novel enzyme called Rh NUDX1 (a nudix hydrolase) producing geraniol has been discovered in roses and constitutes a catalysis model that differs from known monoterpene synthases (Magnard et al . 2019). This indicates that our understanding of net-regulatory metabolism of terpenes, including substrate mediation and stimuli by environmental factors, are presently incomplete.
We used two promoters to examine regulation of terpene synthesis. Using the LaTPS7 promoter predicted by PlantCARE resulted in our discovery of MYC transcription factors (TFs), which helped explain the wound-associated pattern of expression observed via the GUS stain bioassay. An analogous pattern was discovered in Capsicum annuumL. (Wu et al. 2019; Wang et al. 2016). Studies on relationships between TFs and TPSs, such as those among NAC, ethylene-insensitive3-like TF, and R2R3-MYB, can be conducted in lavenders (Nieuwenhuizen et al.2015; Reddy et al. 2017). Although some of TFs have been demonstrated in L. × intermedia (Sarker et al . 2019), other regulatory factors modulating terpene release will likely be discovered in lavenders.
Conversely, CYPs participate in defence of a wide variety of plants. Plant CYPS include CYP51 in oats (Avena spp.), CYP99A2 and CYP99A3 in rice (Oryza sativa ), and CYP82G1 in Arabidopsis(Qi et al . 2006; Shimura et al . 2007; Lee et al . 2010). LaCYP71D582 , the first CYP gene cloned from lavenders, protects the plants by converting limonene into carveol to enrich terpenoid metabolites in early stages of flowering. Although some reports show that CYPs target to plastidial membranes (Kim et al.2010; Quinlan et al. 2012), we found that La CYP71D582 was localized to the endoplasmic reticulum (ER) (Figure 3), close to plastids within plant cells. This enabled putative communication between these two organelles, which was also shown by Ginglinger et al.(2013). LaCYP71D582 shared 57.68 and 7.98% similarity with limonene-6-hydroxylase (M. canadensis ) (Accession number: QDF63370.1) and limonene-6-hydroxylase (M . x gracilis ) (Accession number: AAQ18706.1), respectively, indicating that determining the function of a CYP only by its sequence may be unreliable.
TheLaCYP71D582 promoter possesses elements for response to environmental stimuli, such as the GT1-motif responsible for responding to light conditions, and the LTR sequence responsible for responding to low temperature. This indicates that CYPs are modulated by environmental factors. Unfortunately, studies on how environmental factors influence CYPs and terpenoid regulation are rare. CYPs participate in various catalytic and metabolic activities in plants, including metabolism of fatty acids and alkane compounds mediated by CYP72, 701, and 90, flavonoid metabolism mediated by CYP73, 98, and 84, and phytoalexin metabolism mediated by CYP79, 80, and 93 (Nelson and Werck-Reichhart, 2011). Therefore, it is important to study not only CYP-mediated catalysis but also biotic and environmental factors that influence CYP activity.
In summary, herein we examined the functions of three genes in lavenders. These genes were associated with direct and indirect plant protection during the budding stage. We also examined the interaction among the plants, aphids, and ladybugs. This tritrophic interaction is vital for the successful propagation of lavenders (Figure 8). Collectively, our results show that metabolism in lavenders during the budding stage involves different regulatory mechanism compared with that occurring during the blossoming stage. Our results also indicate that plants used multiple tactics to protect themselves against herbivores and offer a new perspective on biological pest control.