FAD-dependent hexenal isomerases in Lepidoptera evolved convergently with plant-derived hexenal isomerases

FAD-dependent hexenal isomerases in Lepidoptera evolved convergently with plant-derived hexenal isomerases | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article FAD-dependent hexenal isomerases in Lepidoptera evolved convergently with plant-derived hexenal isomerases Yu-Hsien Lin, Bulah Chia-hsiang Wu, Abdoallah Sharaf, Sophie Heijblom, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7163309/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 27 Feb, 2026 Read the published version in Nature Ecology & Evolution → Version 1 posted You are reading this latest preprint version Abstract Green leaf volatiles (GLVs) are six carbon volatile organic compounds that mediate plant responses to environmental stresses. The quantity and composition of emitted GLVs can vary with stress type, allowing plants to fine-tune their volatile blends. Additionally, insect herbivores are capable of modulating these emissions. A key mechanism underlying this plasticity is the conversion of Z-3-hexenal to E-2-hexenal by the enzyme (3Z):(2E)-hexenal isomerase (Hi), which reshapes GLV profiles and influences multi-trophic interactions. Here, we investigate the evolutionary origin, functional diversification, and catalytic mechanisms of Lepidopteran Hi homologs, which belong to the GMC oxidoreductase family. Phylogenetic analysis of 34 lepidopteran species identified a distinct GMCβ subclade enriched in Hi homologs, largely confined to the Apoditrysia lineage. Functional assays showed species-specific variation in Hi activity, with Manduca sexta Hi-1 displaying the highest activity under identical protein concentrations, both in vitro and in planta. Structural modeling and site-directed mutagenesis revealed that Hi activity requires a flavin adenine dinucleotide (FAD) cofactor enabling the identification of key residues critical for FAD binding. Comparative phylogenetics further suggests that Hi enzymes in plants and Lepidoptera evolved independently from unrelated enzyme families, representing a case of functional convergence during the Cretaceous angiosperm radiation. Biological sciences/Evolution/Molecular evolution Biological sciences/Ecology/Ecophysiology Biological sciences/Plant sciences/Plant evolution Biological sciences/Zoology/Entomology Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Plants interact with their environment through the emission of volatile organic compounds (VOCs) (Dudareva et al., 2006). Among these, green leaf volatiles (GLVs) form a group of six‑carbon (C 6 ) molecules that impart the “grassy” scent to foliage (Hatanaka et al., 1987). Produced by most green plants, GLVs are emitted within seconds of mechanical wounding, herbivore feeding, or various abiotic stresses (Matsui and Engelberth, 2022). They contribute to a plant’s direct defense by deterring herbivores and inhibiting pathogenic infection, and to a plant’s indirect defense, by attracting natural enemies of herbivores (Matsui and Engelberth, 2022). Owing to this near‑instantaneous release, GLVs also serve as ecological “alarm signals” that prime, and in some cases directly trigger, plant defenses in neighboring tissues and nearby plants (Ameye et al., 2018). The biosynthesis of GLVs is initiated from polyunsaturated fatty acids, primarily α-linolenic and linoleic acid, via the enzymatic action of lipoxygenases (LOX) (Matsui, 2006). LOXs oxygenate these fatty acids, forming hydroperoxides, subsequently cleaved by hydroperoxide lyases (HPL) into Z -3-hexenal (Z3AL) or hexanal, depending on the initial fatty acid substrate. Within the GLV biosynthesis pathway, the re-arrangement of Z3AL to E ‑2‑hexenal (E2AL) is particularly important because it reshapes the Z 3/ E 2 ratio and thereby modulates the downstream formation of the corresponding alcohols and acetates (Fig. 3e) (Engelberth and Engelberth, 2020). Although Z3AL is emitted first, E2AL exhibits stronger antimicrobial activity and a more pronounced signaling capacity owing to its higher reactivity toward nucleophilic biomolecules (Alméras et al., 2003; Croft et al., 1993; Hyun et al., 2022). This re-arrangement from Z3AL to E2AL could occur spontaneously due to the intrinsic instability of Z3AL but becomes far more efficient when catalyzed by the (3 Z ):(2 E )-hexenal isomerase (Hi). Plant-derived Hi enzymes have been identified in various species, including cucumber, tomato, and rice, and belong to the cupin superfamily (Chen et al., 2022; Kunishima et al., 2016; Phillips et al., 1979; Spyropoulou et al., 2017) Intriguingly, larvae of the hawk moth ( Manduca sexta ) secrete a functionally analogous but phylogenetically distinct Hi protein in their oral secretions, affiliated with the GMC oxidoreductase that converts plant-produced Z3AL into E2AL while feeding (Lin et al., 2023). Such conversion alters the composition of GLVs, which guides female moths in choosing oviposition sites (Allmann et al., 2013), but paradoxically also serves as the cue that attracts their natural enemies (Allmann and Baldwin, 2010). Hi activity independently emerged in plants and Lepidoptera via distinct protein families, representing a compelling case of convergent evolution. Notably, in vitro assays reveal that Hi activity varies among Lepidopteran species (Jones et al., 2021; Lin et al., 2023). Semi‑field trials confirm this pattern: Manduca sexta releases a pronounced burst of E2AL when feeding on solanaceous hosts, whereas Chloridea virescens produces only a negligible Z3AL to E2AL conversion (Paudel Timilsena et al., 2020). Despite this ecological and evolutionary significance, the origins and functional diversity of Hi are still poorly understood in Lepidoptera. Only a single Hi protein from Manduca sexta has been biochemically and functionally characterized to date (Lin et al., 2023), and inter‑species variation in Hi activity remains unexplored. In this study, we combine phylogenetic and functional analysis to reconstruct the evolutionary history of Lepidopteran Hi. We map its taxonomic distribution and compare the enzymatic activity of Hi homologs from multiple Lepidopteran taxa. We also identify the FAD‑binding and catalytic motifs that are crucial for Lepidopteran Hi activity. Finally, we chart Hi evolution across 34 Lepidopteran species and 183 species in the green lineage (Viridiplantae), revealing that both plant and Lepidopteran Hi arose during the Cretaceous angiosperm radiation. Results Phylogenetic and functional analysis of putative (3 Z) :(2 E )-hexenal isomerase (Hi) genes in Lepidoptera The first Lepidopteran Hi characterized in our earlier study (Lin et al., 2023), M . sexta Hi (MsHi-1), is a member of the GMC oxidoreductase family. This protein family is defined by conserved N-terminal (PF00732) and C-terminal (PF05199) domains. To identify related homologs in Lepidoptera, we used profile hidden Markov models (HMMs) based on these domains. Subsequently, we constructed a maximum-likelihood (ML) phylogram comprising 1251 GMC oxidoreductases from 34 lepidopteran species, spanning both non-Ditrysia and Ditrysia lineages (Supplementary Table 1). Consistent with previous studies (Iida et al., 2007; Sun et al., 2012), the Lepidopteran GMC genes grouped into distinct subfamilies, showing significant expansion within the GMCβ subfamily (Fig. 1a). MsHi-1, along with its gene duplicate lacking Hi activity, MsHi-like, clustered within a well-supported subclade (bootstrap value = 100) within the GMCβ subfamily (Fig. 1a, and Supplementary Data 1, highlighted in orange). To determine whether this subclade might represent a broader set of Hi homologs, we selected all homologous genes from four species - Bombyx mori , Manduca sexta , Chloridea virescens and Danaus plexippus - that cluster within a monophyletic group (bootstrap value = 84) containing MsHi-1 (XP_030035814) (Supplementary Data 2), for further functional characterization. Notably, oral secretions (OS) of these four species had previously been shown to exhibit Hi activity (Lin et al., 2023). Since the salivary glands (SG) and midgut (MG) are primary sources of enzymes in OS, we first investigated whether the selected genes potentially encode Hi enzymes in OS by comparing their expression levels in the SG and MG to the non-OS source tissue, fat body (FB). The majority of candidate genes were highly expressed in the SG, whereas MsHi-like in M. sexta and DpHi-1 in D. plexippus showed comparatively higher expression in the MG (Fig. 1b-e). To assess the Hi activity of these putative homologs, we cloned the corresponding cDNAs into pGEX vectors and expressed the recombinant proteins in E . coli BL21 (Supplementary Fig. 1). A sole homolog from Pieris rapae, present in the putative Hi clade (Supplementary Data 2), was included as a negative control, as previous studies have shown no Hi activity in the OS of this species (Jones et al., 2021; Lin et al., 2023). After expression and purification, we tested three protein quantities (0.01, 0.1, and 1 µg) for their ability to convert Z -3-hexenal (Z3AL) into E -2-hexenal (E2AL) in vitro using SPME-GC-qToF-MS. At least one protein from each species showed a concentration-dependent increase in Hi activity (Fig. 2a). In M . sexta , a second homolog (MsHi-2) also displayed Hi activity, albeit at lower levels than MsHi-1. Representative chromatograms of the 1 µg protein reactions confirm E2AL formation in active homologs (Fig. 2b). As expected, the P . rapae’ s homolog showed no detectable Hi activity. Species-specific Hi activity determines the magnitude of E -2-GLV emissions from wounded host plants Hi activity has been confirmed in the OS of numerous lepidopteran species through ex vivo approaches (Jones et al., 2021; Lin et al., 2023), and here we further identify active Hi homologs. However, only M . sexta has been reported to induce a pronounced increase in E -2-GLVs when feeding on host plants, thereby influencing multitrophic interactions (Allmann and Baldwin, 2010; Allmann et al., 2013). To explore this discrepancy, we applied equal amounts of each species’ recombinant Hi protein to wounded leaves of their respective host plants and measured the emission of GLVs. Treatment with MsHi-1 on wounded tomato leaves led to a significant rise in E2AL emissions, compared to water or heat-inactivated MsHi-1 controls (Fig. 3a; Supplementary Fig. 2a). Furthermore, a notable increase in E -2-hexenol (E2OL)—a reduction product of E2AL by the plant cinnamaldehyde and hexenal reductase (CHR) (Fig. 3e)—was also observed (Fig. 3a’; Supplementary Fig. 2a’). By contrast, treatment with CvHi-1 on wounded tomato leaves induced only modest increases in both E2AL and E2OL (Fig. 3b and 3b’; Supplementary Fig. 2b and 2b’). Similarly, DpHi-1 on wounded milkweed leaves and BmHi-1 on wounded mulberry leaves elicited only slight increases in E2AL; E2OL was undetectable in these samples (Fig. 3c and 3d; Supplementary Fig. 2c and 2d). Taken together, these results indicate that while Hi homologs are widespread in Lepidoptera, MsHi-1 in M . sexta appears uniquely efficient at converting Z3AL into E2AL during feeding on one of their host plants, resulting in a pronounced increase in other E -2-GLVs. Lepidopteran (3 Z ):(2 E )-hexenal isomerase activity is FAD-dependent GMC oxidoreductases typically catalyze redox reactions using flavin adenine dinucleotide (FAD) as a cofactor (Cavener, 1992). Although the conversion of Z3AL to E2AL does not involve a net redox change, we hypothesized that the Lepidopteran Hi still requires FAD for its isomerase activity. To test this, we measured Hi activity in the presence or absence of supplemental FAD. Control reactions showed that FAD alone did not affect the non-enzymatic conversion of Z3AL to E2AL (Fig. 4a, a'). However, the addition of FAD to enzymatic reactions moderately increased Hi activity across all tested homologs (Fig. 4b–e), as indicated by higher E2AL peak areas in the chromatograms (Fig. 4b'–e'). Since Hi activity was still detectable even without the addition of extra FAD, we inferred that FAD was already bound to the enzyme during expression in E . coli and remained associated through the purification process. To further validate the importance of FAD in Hi activity, we generated site-directed mutants at key residues predicted to be involved in FAD binding or catalytic activity. Using AlphaFold 3 (Abramson et al., 2024) to model MsHi-1 with docked FAD, we identified putative FAD-binding residues. We then compared the predicted structure of MsHi-1 with two experimentally determined GMC oxidoreductase structures: aryl-alcohol oxidase from Pleurotus eryngii (PDB: 3FIM) and glucose dehydrogenase from Aspergillus flavus (PDB: 4YNT) (Fig. 4f, g). Superimposing their FAD-binding domains onto the predicted FAD-binding domain of MsHi-1 revealed high structural similarity, with root-mean-square deviation (RMSD) of 0.22 Å and 0.48 Å, respectively (Fig. 4f', g'). Y92 in 3FIM and G94 in 4YNT, previously reported as essential residues for FAD binding (Fernández et al., 2009; Yoshida et al., 2015) were found to correspond to a conserved histidine (H135) in MsHi-1 (Fig. 4f', g') and other Lepidopteran Hi homologs (Fig. 4h and Supplementary Fig. 3). This histidine could potentially form a hydrogen bond with the O4 atom on the isoalloxazine ring of FAD, similar to Y92 and G94 in 3FIM and 4YNT (Fig. 4f', g'), respectively. Additionally, comparative analysis of conserved C-terminal GMC oxidoreductase active-site motifs (Liu et al., 2013) showed MsHi-1 and other Lepidopteran Hi generally contain H–N pairs (e.g., H521/N559 in MsHi-1) (Fig. 4h, and Supplementary Fig. 3) where H–H pairs occur in 3FIM (H502/H546) and 4YNT (H505/H548). These two conserved residues are known to be crucial for substrate positioning and electron transfer in GMC oxidoreductases (Ohta et al., 2006; Wongnate and Chaiyen, 2013). We further introduced alanine substitutions at H135, H521, and N559 to confirm the role of these FAD-interactive residues in Hi activity (Supplementary Fig. 4). The H135A and N559A mutants completely lost Hi activity, even when increasing the amount of protein five times (0.5 µg) (Fig. 4i). The H521A mutant still converted Z3AL to E2AL, but at reduced rates. Enzyme kinetics showed an increased Km (0.04 → 1.01 mM) and a ~25-fold drop in catalytic efficiency (359.6 → 14.0 mM⁻¹ s⁻¹), while Kcat remained roughly unchanged (13.8 vs. 14.1 s⁻¹) (Fig. 4j). In planta assays, using wounded tomato leaves, confirmed these findings (Supplementary Fig. 5). Mechanically wounded leaves that were treated with recombinant protein of either MsHi H135A or MsHi N559A emitted predominantly Z3AL, resembling water-treated controls. MsHi H521A -treated leaves released more E2AL than those treated with the other two mutants, but still significantly less than wildtype MsHi-1-treated leaves. These findings support the conclusion that the cofactor FAD is essential for Lepidopteran Hi activity. Distribution and evolutionary history of (3 Z ):(2 E )-hexenal isomerase in plants and lepidopterans Previous studies have shown that in plants, an enzyme of the cupin superfamily catalyzes the conversion of Z3AL to E2AL (Kunishima et al., 2016; Spyropoulou et al., 2017). Both plant- and Lepidopteran-derived Hi display promiscuous activity toward other Z -3-aldehydes, such as Z -3-octenal and Z -3-nonenal (Lin et al., 2023; Spyropoulou et al., 2017). This functional convergence in plants and insects raises intriguing questions about when and how Hi evolved independently in these distinct lineages. In plants, the cupin superfamily comprises structurally conserved proteins characterized by a ß-barrel fold, with each member containing either one or two cupin domains (Dunwell et al., 2001). We first searched for proteins containing the cupin domain (Pfam: PF07883) across 183 representative Viridiplantae (green lineage) species (Supplementary Table 2a). This search yielded 3734 sequences, which we used to construct a phylogenetic tree, alongside 31 reference sequences from well-characterized cupin superfamily members, including Germin, Vicilin, Legumin, Globulin, and plant Hi proteins (Supplementary Table 2b). The previously characterized plant Hi proteins clustered within a distinct subclade of the cupin superfamily, which we refer to as the “Hi-potential clade” (Supplementary Fig. 6). Furthermore, our analysis suggests that the proteins in the Hi-potential clade first emerged in embryophyte (land plant) hornworts, were subsequently lost, and later retained in Bryopsida (Supplementary Fig. 7 and Supplementary Table 2c). Subsequently, a rooted phylogenetic tree was constructed focusing specifically on the sequences within the "Hi-potential clade" (260 sequences, Supplementary Table 2c). ). To elucidate their evolutionary relationships (Fig. 5a), we incorporated reference sequences from well-characterized cupin subfamilies—Germin, Vicilin, Globulin, and Legumin—that lie outside the Hi-potential clade and serve as outgroups. The rooted ML tree grouped the proteins in the Hi-potential clade into seven clades (Fig. 5a). The four clades located near the root of the tree (Clade I to IV) include all identified orthologs from ferns, spikemosses, liverworts, and hornworts. The remaining sequences formed three distinct crown clades: Clade V contains all gymnosperm orthologs, Clade VI includes angiosperm orthologs and the Clade Hi, located at the top of the tree, contains orthologs from mesangiosperms (core angiosperms). Our data indicate that only proteins in the pink-shaded “clade Hi” possess the critical catalytic residues (H/K/Y) required for Hi activity (Fig. 5b) (Kunishima et al., 2016). Moreover, all proteins in clade Hi are derived from mesangiosperms, whereas proteins from basal angiosperms ( Amborella trichopoda and Nymphaea colorata , ANA grade) form a separate sister clade (clade VI) that lacks these catalytic residues (Fig. 5a, blue highlight). Interestingly, among 16 monocot species examined, only rice ( Oryza sativa ) possessed proteins within clade Hi (Fig. 5a). Hi proteins were entirely absent in six of the sixteen eudicot families analyzed—including Brassicaceae and Caricaceae from the Brassicales order—across the 31 species examined (Fig. 5c). In parallel, we expanded our analysis of the lepidopteran GMCβ oxidoreductase subclade, enriched in Hi homologs (Fig. 1a). Hi homologs were absent in all non-Ditrysian superfamilies—Micropterigoidea, Hepialoidea, Tischerioidea, and Palaephatoidea—collectively representing the early-diverging lineages of Lepidoptera, as well as in the basal Ditrysian superfamilies Tineoidea and Yponomeutoidea (Fig. 5d and Supplementary Table 1a). Instead, Hi homologs were exclusively identified within the Apoditrysia lineages, including Obtectomera and Macroheterocera. This pattern suggests that Hi arose relatively recently in the Lepidopteran lineage, potentially coinciding with the diversification of Apoditrysia. Given that Apoditrysia is estimated to have a crown-group origin in the Early Cretaceous (~118.5 Ma) (Kawahara et al., 2019), a period marked by rapid angiosperm radiation, it is plausible that the emergence of Lepidopteran Hi reflects an adaptive response to increasingly novel and chemically diverse host plants. Discussion In this study, we provide new insights into the evolutionary history and taxonomic distribution of plant and insect Hi genes and describe the enzymatic versatility of Hi genes in Lepidoptera. Functional assays of Lepidopteran Hi homologs revealed that at least one gene of each examined species exhibited measurable Hi activity, indicating that these enzymes may contribute to the herbivore-driven production of E2AL in various lepidopteran species. Notably, Hi-1 from M . sexta induced significantly higher levels of E2AL and downstream E-2-GLVs in host plants compared to Hi from other three Lepidopteran species (Fig. 3a). These results, obtained in situ by applying recombinant Hi proteins onto mechanically wounded leaves of host plants, support previous observations that plants fed on by M . sexta release significantly higher levels of E2AL overtime compared to plants fed on by C . virescens (Paudel Timilsena et al., 2020). Our findings also suggest that Hi activity is not universally linked to trophic specialization, as Hi proteins from two specialist herbivores, D . plexippus and B . mori , induced much lower E2AL emissions than M . sexta’s Hi-1 (Fig. 3c and 3d). The ecological relevance of Hi is well established: shifts in the ratio of Z -3- to E -2-GLVs, driven by Hi activity, serve as cues for both ovipositing moths and natural enemies of the herbivore (Allmann and Baldwin, 2010; Allmann et al., 2013). However, whether Hi ultimately benefits the insect or the plant remains unclear, as its effects can be both advantageous and detrimental depending on context. Similarly, other OS effectors, such as fatty acid dehydratases (FHDs), hexenal‑trapping molecules (HALTs) and glucose oxidase (GOX), can reduce GLV emissions (Jones et al., 2021), which may dampen the strength of volatile cues in the environment. Notably, suppression of GLVs via FHD has been shown to decrease parasitoid attraction, potentially benefiting the herbivore by reducing attack risk (Takai et al., 2018). Beyond its ecological effects, we previously demonstrated that Hi enzymes act on a range of Z -3-alkenal substrates, and that hi mutants of M. sexta raised on artificial diet lacking GLVs developed more slowly and showed increased rates of adult abnormalities compared to wild-type insects (Lin et al., 2023). This suggests that Hi may also serve internal physiological roles independent of interactions with GLV-producing plants. Together, these findings highlight that variation in Hi activity among lepidopteran species likely reflects a combination of ecological, physiological, and evolutionary factors. Homology analysis indicates that Lepidopteran Hi proteins belong to the GMC oxidoreductase protein family. Given its involvement in an electrophilic isomerization reaction, this enzyme could be classified as an intramolecular oxidoreductase (EC 5.3). A notable example of such an intramolecular oxidoreductase is isopentenyl pyrophosphate (IPP) isomerase from plants, which participates in isoprenoid biosynthesis. This enzyme catalyzes the isomerization of IPP into its electrophilic isomer, dimethylallyl diphosphate (DMAPP), via a protonation/deprotonation mechanism (Berthelot et al., 2012; Reardon and Abeles, 1986). Our results demonstrated that FAD is essential for Lepidopteran Hi activity, even though there is no net redox change between the substrate and the product (Fig. 4). The role of FAD in this reaction may resemble its function in polyunsaturated fatty acid isomerase (PAI) from Propionibacterium acnes (Liavonchanka and Feussner, 2008; Liavonchanka et al., 2009). In PAI, FAD facilitates the double bond isomerization of linoleic acid to conjugated linoleic acid through an ionic mechanism. During this reaction, FAD stabilizes a carbocation intermediate by interacting with the substrate's double bonds via its redox-active isoalloxazine ring. Another example of a FAD-dependent non-redox reaction is carotene cis-trans isomerase (CRTISO) in plants, where FAD facilitates the isomerization of prolycopene to all-trans-lycopene during carotenoid biosynthesis (Eggers et al., 2021; Yu et al., 2011). These examples suggest that FAD in Lepidopteran Hi may facilitate a coupled isomerization and double bond migration from Z -3 to E -2-aldehydes by stabilizing the transient enolate intermediate and aiding proton abstraction. The Z3AL isomerization by plant Hi has previously been reported to involve a keto-enol tautomerism mechanism (Kunishima et al., 2016). The process is mediated by a conserved catalytic histidine (Fig. 5b), which abstracts a proton from the C2 position of Z3AL, forming a transient enolate intermediate. This enolate structure allows the electron density to shift, facilitating the formation of a keto-like tautomer. We identified an N-terminal histidine in Lepidoptera Hi (Fig. 4h and Supplementary Fig. 3), which is critical for Hi activity. As this conserved histidine has been previously reported in other GMC oxidoreductases to abstract the proton from the substrate (Liu et al., 2013; Yoshida et al., 2015), this histidine may perform a role that is similar to a catalytic histidine in plant Hi. Some of the inactive Hi homologs show substitution at catalytic residues such as leucine and aspartic acid replacing the N-terminal histidine in C. virescens ‘s PCG65132 and PCG76483, or alanine, leucine, or tyrosine replacing the C-terminal histidine critical for FAD binding in C . virescens ‘s PCG69794, PCG76483 and P. rapae ’s homolog, respectively (Supplementary Fig. 3). Nevertheless, several Lepidopteran Hi proteins with all three conserved FAD binding and catalytic sites are still unable to rearrange Z3AL to E2AL (Fig. 2 and Supplementary Fig. 3). This includes the gene duplicate of MsHi-1, MsHi-like (Fig. 2), which shares 85% coding sequence identity. A similar observation was made in a plant Hi: one of the Cucumis sativus Hi proteins (Cs033080/ XP_011651276), which showed no activity (Spyropoulou et al., 2017), despite possessing all catalytic HKY residues and clustering within the clade Hi in the phylogenetic tree (Fig. 5a and Supplementary Table 2b). Future studies are needed to compare and test the roles of surrounding residues directly involved in Z3AL interactions between active and inactive Lepidopteran Hi, as well as Hi with varying activity levels. Insights from plant Hi suggest that a tyrosine near the catalytic histidine (Tyr-128 in C . annuum ) may form the substrate binding pocket, highlighting the importance of surrounding residues in determining Hi functionality (Kunishima et al., 2016). A previous study has shown that mechanical damage in Cucurbitaceae and Fabaceae species leads to the emission of a higher proportion of E2AL compared to Z3AL (Engelberth and Engelberth, 2020). Consistent with this, we found that all seven Cucurbitaceae and Fabaceae species in our phylogenetic analysis possess plant Hi homologs containing the catalytic HKY residues (Fig. 5c). In contrast, five Brassicaceae or Caricaceae species within the Brassicales order lack Hi genes (Fig. 5c). This is in agreement with earlier reports that Brassicales species emit Z3AL rather than E2AL upon mechanical damage (Jones et al., 2021). Interestingly, two Lepidopteran specialists that feed on Brassicales— P. rapae and Plutella xylostella —also lack a functional Hi. Although P . rapae possesses a sole homolog clustering within the putative Hi subclade (Fig. 1a and Supplementary Data 2), no activity was detected either from the recombinant protein (Fig. 2a) or from the oral secretions (Lin et al., 2023). Meanwhile, P . xylostella has no representative gene within the putative Hi clade in the phylogenetic tree (Supplementary Tab 1a). One plausible explanation is that the glucosinolate-based defenses of Brassicales reduce the ecological importance of Z3AL to E2AL conversion, thereby relaxing selection for Hi in both the plants and their co-evolved herbivores. While some herbivores, such as M . sexta can rely on GLVs for feeding stimulation (Halitschke et al., 2004) or oviposition cues (Allmann et al., 2013), Brassicales specialists often depend primarily on glucosinolates for both feeding stimulation and oviposition decisions (Hopkins et al., 2009). Similarly, although certain natural enemies use shifts in the Z -3/ E -2-GLVs ratio for prey detection (Allmann and Baldwin, 2010; Arimura et al., 2009; Takai et al., 2018), parasitoid wasps that target Brassicales-feeding caterpillars ( Cotesia rubecula and Hyposoter ebeninus ) instead rely, although not exclusively, on glucosinolate-derived volatiles (e.g. isothiocyanates and nitriles) as their host location cues (Dicke et al., 2009; Kos et al., 2012; Mumm et al., 2008). This may have consequently attenuated the selective pressures driving the maintenance of Hi function in both Brassicales and their specialist herbivores. Phylogenetic analysis (Fig. 5a) and chronogram (Fig. 5d) indicates that plant Hi first arose in the stem lineage of mesangiosperms (192.2–166.4 Ma) (Yang et al., 2020), with no orthologues detectable in basal angiosperms ( A . trichopoda and N . colorata ) (Fig. 5a). This finding implies that the evolution of plant Hi significantly postdates the origin of the GLV pathway, as the capability for GLV biosynthesis and the associated key enzyme, hydroperoxide lyase (HPL), were already established in early vascular plants, including lycophytes and monilophytes (Tanaka et al., 2021). Lepidopteran Hi independently evolved approximately 60 million years later during the early-Cretaceous radiation of Apoditrysia (132.1–105.6 Ma) (Kawahara et al., 2019), coinciding with—and likely facilitated by—the diversification of flowering plants during this period. Notably, several adaptive traits in Apoditrysia, such as a versatile proboscis and expanded detoxification gene families, reflect evolutionary adjustments to novel ecological niches created by angiosperm diversification in the Cretaceous (Breeschoten et al., 2021; Krenn and Kristensen, 2004). This temporal pattern, in which plant metabolic innovations precede and potentially drive insect biochemical adaptations, reflects a broader pattern seen in other evolutionary arms races. A representative example is the glucosinolate detoxification through nitrile-specifier proteins (NSPs): Kelch-type NSPs emerged in Brassicales after a whole-genome duplication event (~85 Ma; At-β event), shortly before insect-specific NSPs independently evolved in Pierinae butterflies (~68 Ma) to redirect host plants’ glucosinolates toward less-toxic nitriles (Edger et al., 2015; Kissen and Bones, 2009; Walden et al., 2020; Wheat et al., 2007). Collectively, the independent emergence of Hi in both mesangiosperms and Lepidoptera represents another example of this evolutionary scenario, potentially enhancing the ecological specificity and impact of GLV-mediated interactions across trophic levels. Materials and Methods Insects and plants Rearing conditions of tobacco hornworm ( Manduca sexta ), tobacco budworm ( Chloridea virescens ), cabbage white ( Pieris rapae ), and monarch butterfly ( Danaus plexippus ) are described in previous articles (Lievers et al., 2020; Lin et al., 2023; Lin et al., 2020). For dissecting different tissues, fourth to fifth instar larvae were dissected in 1x PBS buffer (pH 7.4) to extract midgut, fat body and salivary glands. Tissues from three different individuals were pooled for each biological replicate. All samples were flash-frozen in liquid nitrogen and stored at -80°C. Tomato Micro-Tom ( Solanum lycopersicum ), white mulberry ( Morus alba ) and milkweed ( Asclepias incarnate ) were grown in the greenhouse with a day/night cycle of 16 h (26°C-28°C)/8 h (22°C-24°C) under supplemental light from Master Sun-T PIA Agro 400 or Master Sun-T PIA Plus 600-W sodium lights (Philips). RNA extraction and cDNA synthesis The collected tissues were ground in liquid nitrogen with sterile pestles. Total RNA was extracted using the TRIzol/chloroform method according to the manufacturer’s protocol. The purified RNA was treated with DNase using the Ambion Turbo DNase kit (Thermo Fisher Scientific) to remove genomic DNA. Total RNA concentration was measured by NanoDrop ND-1000 (Thermo Fisher Scientific). One microgram of total RNA was used for cDNA synthesis with RevertAid First Strand cDNA synthesis kit (Thermo Fisher Scientific). Quantitative real-time PCR (ABI 7500 Real-Time PCR System; Applied Biosystems) was performed using the HOT FIREPol® EvaGreen® qPCR Mix Plus (Solis BioDyne). Gene cloning and recombinant protein production The coding regions of putative Hi homologs from C. virescens , M. sexta , and D. plexippus were PCR amplified from a cDNA mixture derived from salivary glands and midgut tissues. The primers used for amplification are listed in Supplementary Table 3. The coding sequences of B. mori were obtained through de novo synthesis (Gene Universal). All coding sequences were cloned into the pGEX-4T-1 vector for GST-fusion protein expression in E. coli . Plasmids were transformed into competent E. coli BL21 (DE3) for recombinant protein expression. A single colony of transformed E. coli was cultured in 10 mL LB medium, shaking overnight at 37 °C. The overnight E. coli culture was then transferred to 1 L of 2×YT medium (16 g tryptone, 10 g yeast extract, 5 g NaCl) and kept shaking at 37 °C until the optical density at 600 nm reached 0.4-0.5. IPTG (1 mM final concentration) was added to induce recombinant protein expression, and the culture was incubated with shaking at 16°C for 24-48 hours. For MsHi-2, induction was performed with 0.2mM IPTG at 10°C for 120 hours. The E. coli pellet was collected by centrifugation at 15000 g. After discarding the supernatant, the pellet was snap-frozen in liquid nitrogen and stored at -20°C until use. For purification of recombinant proteins, the E. coli pellet was first resuspended in 30 mL of lysis buffer containing 1× PBS (pH 7.3), 1 mM EDTA, 10 mg/mL lysozyme, and proteinase inhibitor cocktails (50 mL per tablet). The suspended pellet was sonicated on ice, followed by the addition of 1% Triton X-100 and rotated for 30 minutes. The E. coli lysate was collected by centrifugation at 15000 g and passed through a 0.45 µm filter. The lysate was batch purified using GST Sepharose 4B (GE Healthcare) according to the manufacturer's instructions, and the purified proteins were preserved in 50 mM Tris-HCl buffer (pH 8.0) with 10% glycerol at -80°C until use. Quantitative densitometry of proteins from SDS-PAGE stained with Coomassie blue was used to determine protein concentrations by comparing the relative intensity of bands between recombinant proteins and a BSA standard (Bio-Rad Image Lab). Western blot analysis Recombinant proteins were mixed with 4× loading buffer and boiled at 95°C for 3 minutes. The proteins were separated on a 10% SDS-PAGE gel and transferred to an Immobilon-E PVDF membrane (Millipore) by semi-dry blotting. The membrane was washed three times with 1× PBST (0.05% Tween 20) for 15 minutes each, then blocked with 5% BSA at room temperature for one hour. Subsequently, the membrane was incubated with GST-HRP conjugated antibody (1:2000, Santa Cruz Biotechnology) on a rotator overnight at 4°C. After three additional washes with 1× PBST for 15 minutes each, the membrane was treated with 1 mL of chemiluminescence solution (100 mM Tris-HCl pH8.5, 9 mL of H 2 O, 1 mL of 1M Tris-HCl pH 8.5, 22 μL of 90 mM p-coumaric acid, 50 μL of 200 mM luminol, and 3 μL of 30% H 2 O 2 ). Images were captured using the Odyssey® Fc Imaging System (LI-COR) and analyzed with Image Studio Lite, or the ChemiDoc MP Imaging System and analyzed with Bio-Rad Image Lab. Homolog identification and phylogenetic analysis in Lepidoptera Thirty-four lepidopteran species representing a broad range of superfamilies were selected to capture the phylogenetic diversity of the order. In addition, three species from Trichoptera (caddisflies), the sister group of Lepidoptera , and Drosophila melanogaster (Diptera) were included to assist in the classification of GMC oxidoreductase subfamilies (Iida et al., 2007; Sun et al., 2012). The proteome sequences of studied organisms were obtained from the NCBI (https://www.ncbi.nlm.nih.gov/) and Ensembl (https://beta.ensembl.org/) databases (Supplementary Table 1a). Genome assemblies were prioritized based on completeness and the availability of annotated protein-coding genes. For lineages lacking annotated genomes, specifically Hepialoidea, Tischerioidea, and Palaephatoidea, transcriptome shotgun assemblies (TSAs) were used as alternative sources of protein coding sequences. From these transcriptome assemblies, open reading frames (ORFs) were predicted using the getorf program in the EMBOSS suite (version 6.5.7) (Rice et al., 2000). To identify GMC oxidoreductase homologs, we employed a domain-based search strategy using profile hidden Markov models (HMMs). HMM profiles for the N-terminal (PF00732) and C-terminal (PF05199) domains of the GMC oxidoreductase family were downloaded from the Pfam database hosted by InterPro (https://www.ebi.ac.uk/interpro/). These profiles were combined into a single database and indexed using hmmpress from the HMMER suite (v3.3.2) (Eddy, 1998). These validated sequences were then extracted from the original protein FASTA files, yielding a refined set of full-length GMC homologs. To reduce redundancy from alternative splicing, isoforms were collapsed by retaining only the longest protein variant per gene. Three fungal GMC sequences (AAF59929.2, XP_001727544.1, and AAD01493.1) were incorporated as outgroup taxa. The final dataset comprised 1,251 curated GMC domain sequences used for phylogenetic analysis. Species names, taxonomic classifications, and sequence sources are detailed in Supplementary Table 1b. For phylogenetic inference, protein sequences were aligned using MAFFT (v7.520) (Katoh and Standley, 2013) with the E-INS-i algorithm. The resulting multiple sequence alignment was trimmed with ClipKIT (v2.0.1) (Steenwyk et al., 2020). The best-fitting amino acid substitution model was selected using ModelTest-NG (v0.1.6) (Darriba et al., 2019). Maximum likelihood phylogenetic analysis was conducted with IQ-TREE2 (v2.3.4) (Minh et al., 2020). Branch support was assessed with 3,000 ultrafast bootstrap replicates and 3,000 SH-aLRT tests. The final phylogenetic tree was rooted with the predefined fungal outgroup sequences. Homolog identification and phylogenetic analysis in the green lineage (Viridiplantae) The complete predicted proteome sequences (Supplementary Table 2a) were obtained from the NCBI GeneBank (https://ncbi.nlm.nih.gov), UniProt-Proteomes database (https://www.uniprot.org), and JGI (http://genome.jgi.doe.gov). The 1000 Plants project (OneKP) database (http://www.onekp.com) was an additional source for predicted proteome sequences inferred from transcriptomic data. All cupin-domain (PF07883) containing proteins were identified on the previously retrieved predicted proteome sequences using the Hidden Markov model (HMM)-based tool hmmsearch (Johnson et al., 2010). All identified sequences (3734 sequences) were combined with 31 representative sequences from Cupin protein subfamilies (including Germin, Vicilin, Legumin, Globulin, and Hi) (Supplementary Table 2b) into one single dataset and analyzed using a phylogenetic approach (Supplementary Fig. 6). Only sequences that clustered within the Hi-potential clade in the phylogeny were identified as potential Hi orthologs (Supplementary Table 2c). SMART and Pfam databases were employed to identify conserved domains present in potential Hi orthologs (El-Gebali et al., 2018; Letunic and Bork, 2017). The results from both databases were merged, redundant domains were filtered-out and domain architecture was analyzed using the HMM based tool hmmscan (Finn et al., 2011). All identified potential Hi orthologs and representative sequences from cupin protein subfamilies were aligned using MAFFT, and ambiguously aligned regions were removed with trimAl (Capella-Gutiérrez et al., 2009). The resulting alignments were evaluated with ProtTest v3 (Darriba et al., 2011) to determine the best-fit model for amino acid substitution. Two separate Maximum likelihood (ML) phylogenetic analyses were computed using RAxML-NG (Kozlov et al., 2019) and IQ-TREE 2 (Minh et al., 2020) (Fig. 5a), each with 1000 bootstrap replicates. Bootstrap support values from both analyses were mapped onto the IQ-TREE 2 ML tree, which was rooted at its midpoint (Fig. 5a). (3 Z ):(2 E )-hexenal isomerase (Hi) activity assays by SPME-GC-qToF-MS A 200 µL solution containing recombinant proteins in a 20 mM Tris-HCl buffer (pH 8.5) was first transferred to a 1.5 mL GC vial equipped with a 200 µL insert. The reaction was initiated by adding Z-3-hexenal (0.2mM final concentration) to the solution. The GC vial was gently vortexed for 1 min. Subsequently, 100 µL mixture was transferred to a 20 mL glass headspace vial (SureSTART, Thermo Scientific) and immediately closed with a cap crimper. Headspace volatiles were collected using the Solid Phase Micro Extraction (SPME) fiber (Carboxen/Polydimethylsiloxane coated) for 10 min at 35 °C and analyzed by GC-qToF-MS. After sampling, the fiber was desorbed for 1 min in the injection port which was constantly kept at 250 °C. Compounds were separated on a HP-5ms column (30 m x 250 µm, 0.25 µm film thickness; Agilent) in an Agilent 7890A gas chromatograph with a temperature program set to 40 °C for 5 min, increasing to 140 °C at a rate of 5 °C per min, followed by increasing temperature to 250 °C at a rate of 15 °C per min and an additional 5 min at 250 °C. Helium was used as the carrier gas with the transfer column flow set to 3 mL per minute and a flow rate of 1 mL per min thereafter. Mass spectra were generated by an Agilent 7200 accurate-mass quadrupole time-of-flight mass spectrometer, operating in electron ionization mode (70 eV) at 230 °C and collected with an acquisition rate of 5 scans per second. Volatiles were identified and quantified using standard volatiles listed in Supplementary Table 3. The conversion rate of E -2-hexenal was calculated as the proportion of its intensity relative to the total aldehyde intensity ( Z -3-hexenal + E -2-hexenal). This value was then corrected by subtracting the non-enzymatic conversion rate determined from a buffer-only control, yielding the relative E -2-hexenal level (Fig. 2a). To account for differences in detector sensitivity, response factors for Z -3-hexenal and E -2-hexenal were calculated using measured intensities of 2 nmol of each standard compound. Effect of additional FAD on Hi activity To compare the activity of different Hi enzymes with and without additional FAD, the following amounts of recombinant proteins were used: 3 µg BmHi-1, 1.38 µg DpHi-1, 1 µg CvHi-1, and 0.0625 µg MsHi-1. These protein concentrations were carefully selected to ensure that, without FAD, the conversion of Z -3-hexenal to E -2-hexenal would not reach completion, allowing any potential increase in activity upon FAD addition to be clearly observed. Each enzyme was tested in a 200 µL reaction mixture containing 20 µg BSA, with or without 1 mM FAD, in 20 mM Tris-HCl buffer (pH 7.0). This mixture was transferred to a 1.5 mL GC vial equipped with a 200 µL insert, after which 0.2 mM Z -3-hexenal was added to initiate the reaction. The vial was gently vortexed for 1 min. Subsequently, 100 µL of the mixture was transferred to a 20 mL glass headspace vial and sealed immediately using a cap crimper. Headspace volatiles were collected using the SPME fiber for 10 min at 35 °C and analyzed by GC-qToF-MS. Enzyme kinetics For the determination of the kinetic parameters of the MsHi wildtype and H521A mutant, a substrate concentration range of 5 – 4000 µM Z -3-hexenal was used for the MsHi wildtype and 125 – 4000 µM Z -3-hexenal for the mutant. Recombinant proteins were diluted with 20 mM Tris-HCl (pH 8.5) buffer to a final volume of 200 µL. 37.5 ng of MsHi wildtype and 112.5 ng of H521A mutant were used for measurement. The mixture was incubated at room temperature for 2 min, subsequently transferred to 20 mL glass headspace vial and analyzed by GC-qToF-MS. The Km, Kcat, and enzyme efficiency (Kcat/Km) were calculated with nonlinear least-square regression using GraphPad Prism 10. Analysis of volatiles in planta A 24 mm diameter of leaf disc was punched out from the lamina of the second or third pair of true leaves from the top of four-week-old plants. The leaf disc was mechanically wounded on the adaxial surface by rolling a fabric pattern wheel to produce two parallel rows of punctures on either side of the midvein. 10 µL of recombinant protein, heat-inactivated recombinant protein (boiled at 95 °C for 3 minutes), or water was applied to the wounds and gently dispersed across the leaf surface. After 20 seconds, the leaf disc was transferred to a 20 mL glass headspace vial, closed with a crimp cap and volatiles were immediately collected with a SPME fiber for 10 minutes at 35 °C and measured by GC-qToF-MS analysis. Statistical analysis Statistical analyses were performed using GraphPad Prism 10. The Shapiro–Wilk test was used to assess normality of data distributions. For comparisons among multiple groups, one-way ANOVA followed by Tukey’s multiple comparisons test was used. For non-parametric data, the Kruskal–Wallis test followed by Dunn’s multiple comparisons test was applied. Different letters above bars in the graphs indicate statistically significant differences between groups ( p < 0.05). Two-tailed t-tests were used for comparisons between two groups. Error bars represent mean values ± SD. * p < 0.05, ** p < 0.01, *** p < 0.001, n.s., not significant. Declarations Acknowledgement We thank Prof. Erik Poelman (Wageningen University) who kindly provided us with Pieris rapae . We appreciate the help of our colleagues at University of Amsterdam, Ludek Tikovsky and Harold Lemereis for taking care of all plants in the glasshouse, Dr. Jocelyne Vreede and Dr. Sandra Eltschkner for the discussion of protein structure analysis, Prof. Astrid Groot provided us with Chloridea virescens . We thank Dr. Heiko Vogel (Max Planck Institute), Dr. David Doležel (Biology Centre CAS) and Dr. Smýkal Vlastimil (Biology Centre CAS) for valuable discussion on this work. Computational resources for homolog identification and phylogenetic analysis in Lepidoptera were provided by the e-INFRA CZ project (ID: 90254), supported by the Ministry of Education, Youth and Sports of the Czech Republic, and the ELIXIR-CZ project (ID: 90255), part of the international ELIXIR infrastructure. A.S. was supported by the SequAna Sequencing Analysis Core Facility at the Department of Biology, University of Konstanz. C.W.T, R.S., and J.A. were supported by U.S. National Science Foundation (NSF‐IOS 1754996). This work was supported by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement no. 805074) (S. A. and Y. H. L.), Contributions Y.H.L. conceived and designed the study, performed experiments, analyzed data, and drafted the manuscript. B.C.H.W. and A.S. analyzed the phylogeny. S.M.E.H. performed cloning and qPCR for D . plexippus , C . virescens and M . sexta Hi. I.P. performed GC/MS analysis and kinetic measurements of mutant Hi. C.W.T., J.A. and R.S. conducted qPCR analysis for D . plexippus Hi. S.A. conceived the study, supervised the project, acquired funding, and drafted the manuscript. All authors read and approved the final manuscript. References Abramson, J., Adler, J., Dunger, J., Evans, R., Green, T., Pritzel, A., Ronneberger, O., Willmore, L., Ballard, A.J., Bambrick, J., Bodenstein, S.W., Evans, D.A., Hung, C.-C., O’Neill, M., Reiman, D., Tunyasuvunakool, K., Wu, Z., Žemgulytė, A., Arvaniti, E., Beattie, C., Bertolli, O., Bridgland, A., Cherepanov, A., Congreve, M., Cowen-Rivers, A.I., Cowie, A., Figurnov, M., Fuchs, F.B., Gladman, H., Jain, R., Khan, Y.A., Low, C.M.R., Perlin, K., Potapenko, A., Savy, P., Singh, S., Stecula, A., Thillaisundaram, A., Tong, C., Yakneen, S., Zhong, E.D., Zielinski, M., Žídek, A., Bapst, V., Kohli, P., Jaderberg, M., Hassabis, D., Jumper, J.M., 2024. Accurate structure prediction of biomolecular interactions with AlphaFold 3. Nature 630, 493-500. Allmann, S., Baldwin, I.T., 2010. Insects betray themselves in nature to predators by rapid isomerization of green leaf volatiles. Science 329, 1075-1078. Allmann, S., Spathe, A., Bisch-Knaden, S., Kallenbach, M., Reinecke, A., Sachse, S., Baldwin, I.T., Hansson, B.S., 2013. Feeding-induced rearrangement of green leaf volatiles reduces moth oviposition. Elife 2, e00421. Alméras, E., Stolz, S., Vollenweider, S., Reymond, P., Mène-Saffrané, L., Farmer, E.E., 2003. Reactive electrophile species activate defense gene expression in Arabidopsis. The Plant Journal 34, 205-216. Ameye, M., Allmann, S., Verwaeren, J., Smagghe, G., Haesaert, G., Schuurink, R.C., Audenaert, K., 2018. Green leaf volatile production by plants: a meta-analysis. New Phytologist 220, 666-683. Arimura, G.-i., Matsui, K., Takabayashi, J., 2009. Chemical and Molecular Ecology of Herbivore-Induced Plant Volatiles: Proximate Factors and Their Ultimate Functions. Plant and Cell Physiology 50, 911-923. Berthelot, K., Estevez, Y., Deffieux, A., Peruch, F., 2012. Isopentenyl diphosphate isomerase: A checkpoint to isoprenoid biosynthesis. Biochimie 94, 1621-1634. Breeschoten, T., van der Linden, C.F.H., Ros, V.I.D., Schranz, M.E., Simon, S., 2021. Expanding the Menu: Are Polyphagy and Gene Family Expansions Linked across Lepidoptera? Genome Biology and Evolution 14. Capella-Gutiérrez, S., Silla-Martínez, J.M., Gabaldón, T., 2009. trimAl: a tool for automated alignment trimming in large-scale phylogenetic analyses. Bioinformatics 25, 1972-1973. Cavener, D.R., 1992. GMC oxidoreductases: A newly defined family of homologous proteins with diverse catalytic activities. Journal of Molecular Biology 223, 811-814. Chen, C., Yu, F., Wen, X., Chen, S., Wang, K., Wang, F., Zhang, J., Wu, Y., He, P., Tu, Y., Li, B., 2022. Characterization of a new (Z)-3:(E)-2-hexenal isomerase from tea (Camellia sinensis) involved in the conversion of (Z)-3-hexenal to (E)-2-hexenal. Food Chemistry 383, 132463. Croft, K., Juttner, F., Slusarenko, A.J., 1993. Volatile Products of the Lipoxygenase Pathway Evolved from Phaseolus vulgaris (L.) Leaves Inoculated with Pseudomonas syringae pv phaseolicola. Plant Physiol 101, 13-24. Darriba, D., Posada, D., Kozlov, A.M., Stamatakis, A., Morel, B., Flouri, T., 2019. ModelTest-NG: A New and Scalable Tool for the Selection of DNA and Protein Evolutionary Models. Molecular Biology and Evolution 37, 291-294. Darriba, D., Taboada, G.L., Doallo, R., Posada, D., 2011. ProtTest 3: fast selection of best-fit models of protein evolution. Bioinformatics 27, 1164-1165. Dicke, M., van Loon, J.J.A., Soler, R., 2009. Chemical complexity of volatiles from plants induced by multiple attack. Nature Chemical Biology 5, 317-324. Dudareva, N., Florence, N., A., N.D., and Orlova, I., 2006. Plant Volatiles: Recent Advances and Future Perspectives. Critical Reviews in Plant Sciences 25, 417-440. Dunwell, J.M., Culham, A., Carter, C.E., Sosa-Aguirre, C.R., Goodenough, P.W., 2001. Evolution of functional diversity in the cupin superfamily. Trends in Biochemical Sciences 26, 740-746. Eddy, S.R., 1998. Profile hidden Markov models. Bioinformatics 14, 755-763. Edger, P.P., Heidel-Fischer, H.M., Bekaert, M., Rota, J., Glöckner, G., Platts, A.E., Heckel, D.G., Der, J.P., Wafula, E.K., Tang, M., Hofberger, J.A., Smithson, A., Hall, J.C., Blanchette, M., Bureau, T.E., Wright, S.I., dePamphilis, C.W., Eric Schranz, M., Barker, M.S., Conant, G.C., Wahlberg, N., Vogel, H., Pires, J.C., Wheat, C.W., 2015. The butterfly plant arms-race escalated by gene and genome duplications. Proceedings of the National Academy of Sciences 112, 8362-8366. Eggers, R., Jammer, A., Jha, S., Kerschbaumer, B., Lahham, M., Strandback, E., Toplak, M., Wallner, S., Winkler, A., Macheroux, P., 2021. The scope of flavin-dependent reactions and processes in the model plant Arabidopsis thaliana. Phytochemistry 189. El-Gebali, S., Mistry, J., Bateman, A., Eddy, S.R., Luciani, A., Potter, S.C., Qureshi, M., Richardson, L.J., Salazar, G.A., Smart, A., Sonnhammer, E.L L., Hirsh, L., Paladin, L., Piovesan, D., Tosatto, S.C E., Finn, R.D., 2018. The Pfam protein families database in 2019. Nucleic Acids Research 47, D427-D432. Engelberth, J., Engelberth, M., 2020. Variability in the Capacity to Produce Damage-Induced Aldehyde Green Leaf Volatiles among Different Plant Species Provides Novel Insights into Biosynthetic Diversity. Plants 9, 213. Fernández, I.S., Ruíz-Dueñas, F.J., Santillana, E., Ferreira, P., Martínez, M.J., Martínez, A.T., Romero, A., 2009. Novel structural features in the GMC family of oxidoreductases revealed by the crystal structure of fungal aryl-alcohol oxidase. Acta crystallographica. Section D, Biological crystallography 65, 1196-1205. Finn, R.D., Clements, J., Eddy, S.R., 2011. HMMER web server: interactive sequence similarity searching. Nucleic Acids Research 39, W29-W37. Halitschke, R., Ziegler, J., Keinänen, M., Baldwin, I.T., 2004. Silencing of hydroperoxide lyase and allene oxide synthase reveals substrate and defense signaling crosstalk in Nicotiana attenuata. The Plant Journal 40, 35-46. Hatanaka, A., Kajiwara, T., Sekiya, J., 1987. Biosynthetic pathway for C6-aldehydes formation from linolenic acid in green leaves. Chemistry and Physics of Lipids 44, 341-361. Hopkins, R.J., van Dam, N.M., van Loon, J.J.A., 2009. Role of Glucosinolates in Insect-Plant Relationships and Multitrophic Interactions. Annual Review of Entomology 54, 57-83. Hyun, J., Lee, J.G., Yang, K.-Y., Lim, S., Lee, E.J., 2022. Postharvest Fumigation of (E)-2-Hexenal on Kiwifruit (Actinidia chinensis cv. ‘Haegeum’) Enhances Resistance to Botrytis cinerea. Postharvest Biology and Technology 187, 111854. Iida, K., Cox-Foster, D.L., Yang, X., Ko, W.-Y., Cavener, D.R., 2007. Expansion and evolution of insect GMC oxidoreductases. BMC Evolutionary Biology 7, 75. Johnson, L.S., Eddy, S.R., Portugaly, E., 2010. Hidden Markov model speed heuristic and iterative HMM search procedure. BMC Bioinformatics 11, 431. Jones, A.C., Cofer, T.M., Engelberth, J., Tumlinson, J.H., 2021. Herbivorous Caterpillars and the Green Leaf Volatile (GLV) Quandary. J Chem Ecol. Katoh, K., Standley, D.M., 2013. MAFFT Multiple Sequence Alignment Software Version 7: Improvements in Performance and Usability. Molecular Biology and Evolution 30, 772-780. Kawahara, A.Y., Plotkin, D., Espeland, M., Meusemann, K., Toussaint, E.F.A., Donath, A., Gimnich, F., Frandsen, P.B., Zwick, A., Dos Reis, M., Barber, J.R., Peters, R.S., Liu, S., Zhou, X., Mayer, C., Podsiadlowski, L., Storer, C., Yack, J.E., Misof, B., Breinholt, J.W., 2019. Phylogenomics reveals the evolutionary timing and pattern of butterflies and moths. Proc Natl Acad Sci U S A 116, 22657-22663. Kissen, R., Bones, A.M., 2009. Nitrile-specifier Proteins Involved in Glucosinolate Hydrolysis in Arabidopsis thaliana*. Journal of Biological Chemistry 284, 12057-12070. Kos, M., Houshyani, B., Wietsma, R., Kabouw, P., Vet, L.E.M., van Loon, J.J.A., Dicke, M., 2012. Effects of glucosinolates on a generalist and specialist leaf-chewing herbivore and an associated parasitoid. Phytochemistry 77, 162-170. Kozlov, A.M., Darriba, D., Flouri, T., Morel, B., Stamatakis, A., 2019. RAxML-NG: a fast, scalable and user-friendly tool for maximum likelihood phylogenetic inference. Bioinformatics 35, 4453-4455. Krenn, H.W., Kristensen, N.P., 2004. Evolution of proboscis musculature in Lepidoptera. Eur J Entomol 101, 565-575. Kunishima, M., Yamauchi, Y., Mizutani, M., Kuse, M., Takikawa, H., Sugimoto, Y., 2016. Identification of (Z)-3:(E)-2-Hexenal Isomerases Essential to the Production of the Leaf Aldehyde in Plants. J Biol Chem 291, 14023-14033. Letunic, I., Bork, P., 2017. 20 years of the SMART protein domain annotation resource. Nucleic Acids Research 46, D493-D496. Liavonchanka, A., Feussner, I., 2008. Biochemistry of PUFA Double Bond Isomerases Producing Conjugated Linoleic Acid. ChemBioChem 9, 1867-1872. Liavonchanka, A., Rudolph, M.G., Tittmann, K., Hamberg, M., Feussner, I., 2009. On the Mechanism of a Polyunsaturated Fatty Acid Double Bond Isomerase from. Journal of Biological Chemistry 284, 8005-8012. Lievers, R., Kuperus, P., Groot, A.T., 2020. DNA methylation patterns in the tobacco budworm, Chloridea virescens. Insect Biochemistry and Molecular Biology 121, 103370. Lin, Y.-H., Silven, J.J.M., Wybouw, N., Fandino, R.A., Dekker, H.L., Vogel, H., Wu, Y.-L., de Koster, C., Große-Wilde, E., Haring, M.A., Schuurink, R.C., Allmann, S., 2023. A salivary GMC oxidoreductase of Manduca sexta re-arranges the green leaf volatile profile of its host plant. Nature Communications 14, 3666. Lin, Y.H., Tai, C.C., Broz, V., Tang, C.K., Chen, P., Wu, C.P., Li, C.H., Wu, Y.L., 2020. Adenosine Receptor Modulates Permissiveness of Baculovirus (Budded Virus) Infection via Regulation of Energy Metabolism in Bombyx mori. Front Immunol 11, 763. Liu, X., Ohta, T., Kawabata, T., Kawai, F., 2013. Catalytic Mechanism of Short Ethoxy Chain Nonylphenol Dehydrogenase Belonging to a Polyethylene Glycol Dehydrogenase Group in the GMC Oxidoreductase Family. International Journal of Molecular Sciences 14, 1218-1231. Matsui, K., 2006. Green leaf volatiles: hydroperoxide lyase pathway of oxylipin metabolism. Current Opinion in Plant Biology 9, 274-280. Matsui, K., Engelberth, J., 2022. Green Leaf Volatiles—The Forefront of Plant Responses Against Biotic Attack. Plant and Cell Physiology 63, 1378-1390. Minh, B.Q., Schmidt, H.A., Chernomor, O., Schrempf, D., Woodhams, M.D., von Haeseler, A., Lanfear, R., 2020. IQ-TREE 2: New Models and Efficient Methods for Phylogenetic Inference in the Genomic Era. Molecular Biology and Evolution 37, 1530-1534. Mumm, R., Burow, M., Bukovinszkine’Kiss, G., Kazantzidou, E., Wittstock, U., Dicke, M., Gershenzon, J., 2008. Formation of Simple Nitriles upon Glucosinolate Hydrolysis Affects Direct and Indirect Defense Against the Specialist Herbivore, Pieris rapae. Journal of Chemical Ecology 34, 1311-1321. Ohta, T., Kawabata, T., Nishikawa, K., Tani, A., Kimbara, K., Kawai, F., 2006. Analysis of Amino Acid Residues Involved in Catalysis of Polyethylene Glycol Dehydrogenase from Sphingopyxis terrae, Using Three-Dimensional Molecular Modeling-Based Kinetic Characterization of Mutants. Applied and Environmental Microbiology 72, 4388-4396. Paudel Timilsena, B., Seidl-Adams, I., Tumlinson, J.H., 2020. Herbivore-specific plant volatiles prime neighboring plants for nonspecific defense responses. Plant Cell Environ 43, 787-800. Peris, D., Condamine, F.L., 2024. The angiosperm radiation played a dual role in the diversification of insects and insect pollinators. Nature Communications 15, 552. Phillips, D.R., Matthew, J.A., Reynolds, J., Fenwick, G.R., 1979. Partial purification and properties of a cis-3: trans-2-enal isomerase from cucumber fruit. Phytochemistry 18, 401-404. Reardon, J.E., Abeles, R.H., 1986. Mechanism of action of isopentenyl pyrophosphate isomerase: evidence for a carbonium ion intermediate. Biochemistry 25, 5609-5616. Rice, P., Longden, I., Bleasby, A., 2000. EMBOSS: The European Molecular Biology Open Software Suite. Trends in Genetics 16, 276-277. Spyropoulou, E.A., Dekker, H.L., Steemers, L., van Maarseveen, J.H., de Koster, C.G., Haring, M.A., Schuurink, R.C., Allmann, S., 2017. Identification and Characterization of (3Z):(2E)-Hexenal Isomerases from Cucumber. Front Plant Sci 8, 1342. Steenwyk, J.L., Buida, T.J., III, Li, Y., Shen, X.-X., Rokas, A., 2020. ClipKIT: A multiple sequence alignment trimming software for accurate phylogenomic inference. PLOS Biology 18, e3001007. Sun, W., Shen, Y.H., Yang, W.J., Cao, Y.F., Xiang, Z.H., Zhang, Z., 2012. Expansion of the silkworm GMC oxidoreductase genes is associated with immunity. Insect Biochem Mol Biol 42, 935-945. Takai, H., Ozawa, R., Takabayashi, J., Fujii, S., Arai, K., Ichiki, R.T., Koeduka, T., Dohra, H., Ohnishi, T., Taketazu, S., Kobayashi, J., Kainoh, Y., Nakamura, S., Fujii, T., Ishikawa, Y., Kiuchi, T., Katsuma, S., Uefune, M., Shimada, T., Matsui, K., 2018. Silkworms suppress the release of green leaf volatiles by mulberry leaves with an enzyme from their spinnerets. Sci Rep 8, 11942. Walden, N., German, D.A., Wolf, E.M., Kiefer, M., Rigault, P., Huang, X.-C., Kiefer, C., Schmickl, R., Franzke, A., Neuffer, B., Mummenhoff, K., Koch, M.A., 2020. Nested whole-genome duplications coincide with diversification and high morphological disparity in Brassicaceae. Nature Communications 11, 3795. Wheat, C.W., Vogel, H., Wittstock, U., Braby, M.F., Underwood, D., Mitchell-Olds, T., 2007. The genetic basis of a plant–insect coevolutionary key innovation. Proceedings of the National Academy of Sciences 104, 20427-20431. Wongnate, T., Chaiyen, P., 2013. The substrate oxidation mechanism of pyranose 2-oxidase and other related enzymes in the glucose–methanol–choline superfamily. The FEBS Journal 280, 3009-3027. Yang, L.X., Su, D.Y., Chang, X., Foster, C.S.P., Sun, L.H., Huang, C.H., Zhou, X.F., Zeng, L.P., Ma, H., Zhong, B.J., 2020. Phylogenomic Insights into Deep Phylogeny of Angiosperms Based on Broad Nuclear Gene Sampling. Plant Commun 1. Yoshida, H., Sakai, G., Mori, K., Kojima, K., Kamitori, S., Sode, K., 2015. Structural analysis of fungus-derived FAD glucose dehydrogenase. Sci Rep-Uk 5. Yu, Q.J., Ghisla, S., Hirschberg, J., Mann, V., Beyer, P., 2011. Plant Carotene Cis-Trans Isomerase CRTISO. Journal of Biological Chemistry 286, 8666-8676. Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryInformation20250719Final.docx Supplementary Information SupplementaryData1.pdf Supplementary Data 1 SupplementaryData2.pdf Supplementary Data 2 SupplementaryTable1InsectGMC.xlsx Supplementary Table 1 SupplementaryTable2Plantcupin.xlsx Supplementary Table 2 SupplementaryTable3primersandchemicalreagents.xlsx Supplementary Table 3 Cite Share Download PDF Status: Published Journal Publication published 27 Feb, 2026 Read the published version in Nature Ecology & Evolution → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7163309","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":488374353,"identity":"d97a01f1-3b60-42e1-b7dd-bef62c9f99f5","order_by":0,"name":"Yu-Hsien Lin","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5UlEQVRIiWNgGAWjYDAC9uaDDz9USPDwszcAeQYWRGjhOZZsLHHGRkay5wBIiwQRWiRyzAR429JsDG4kgLmEdZjPyDFjkGA7zMNw8/nVDT8KJBj427sT8GqROfOs7EEBz2Eextk5ZTd7gA6TOHN2A353sSdvN5CQOMzDLJ2TdoMHqMVAIpeAFoYEMwkeg8M8bJJn0m7+IUoLRwpQS0IaD48E+7HbxNkCDuQDNjwSPDlst2UMJHiI8AswKj/+k7C3P3782c03f2zk+Nt78WtBAjwGYJJY5SDA/oAU1aNgFIyCUTCCAACgMEUk5kSGjwAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-2647-0495","institution":"Swammerdam Institute for Life Sciences, University of Amsterdam","correspondingAuthor":true,"prefix":"","firstName":"Yu-Hsien","middleName":"","lastName":"Lin","suffix":""},{"id":488374354,"identity":"286c4aeb-599a-4f00-80af-8d420d55dee2","order_by":1,"name":"Bulah Chia-hsiang Wu","email":"","orcid":"","institution":"Biology Centre of the Czech Academy of Sciences, Institute of Entomology","correspondingAuthor":false,"prefix":"","firstName":"Bulah","middleName":"Chia-hsiang","lastName":"Wu","suffix":""},{"id":488374355,"identity":"5eb8cdc2-f9e0-4767-b213-6210a7754d73","order_by":2,"name":"Abdoallah Sharaf","email":"","orcid":"https://orcid.org/0000-0002-3436-9290","institution":"SequAna Core Facility, Department of Biology, University of Konstanz","correspondingAuthor":false,"prefix":"","firstName":"Abdoallah","middleName":"","lastName":"Sharaf","suffix":""},{"id":488374356,"identity":"fe5c0371-8f13-4225-be7a-8badf556f209","order_by":3,"name":"Sophie Heijblom","email":"","orcid":"","institution":"Swammerdam Institute for Life Sciences, University of Amsterdam","correspondingAuthor":false,"prefix":"","firstName":"Sophie","middleName":"","lastName":"Heijblom","suffix":""},{"id":488374357,"identity":"f46c2d2f-aab9-4a59-a5f9-26cb3698152f","order_by":4,"name":"Ilias Prattis","email":"","orcid":"","institution":"Swammerdam Institute for Life Sciences, University of Amsterdam","correspondingAuthor":false,"prefix":"","firstName":"Ilias","middleName":"","lastName":"Prattis","suffix":""},{"id":488374358,"identity":"e665afc0-f984-41dc-b400-982294d74a12","order_by":5,"name":"Ching-Wen Tan","email":"","orcid":"","institution":"Department of Entomology, National Chung Hsing University","correspondingAuthor":false,"prefix":"","firstName":"Ching-Wen","middleName":"","lastName":"Tan","suffix":""},{"id":488374359,"identity":"52777b4e-21d8-4d39-9b25-61c20468bb0c","order_by":6,"name":"Rudolf Schilder","email":"","orcid":"https://orcid.org/0000-0003-1229-1274","institution":"The Pennsylvania State University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Rudolf","middleName":"","lastName":"Schilder","suffix":""},{"id":488374360,"identity":"47ab1a5f-b127-48e3-82d2-9ebc35ecae67","order_by":7,"name":"Jared Ali","email":"","orcid":"https://orcid.org/0000-0001-9870-0299","institution":"Pennsylvania State University","correspondingAuthor":false,"prefix":"","firstName":"Jared","middleName":"","lastName":"Ali","suffix":""},{"id":488374361,"identity":"eba590e7-d1b2-4492-b4fb-a0840cae2892","order_by":8,"name":"Silke Allmann","email":"","orcid":"","institution":"Swammerdam Institute for Life Sciences, University of Amsterdam","correspondingAuthor":false,"prefix":"","firstName":"Silke","middleName":"","lastName":"Allmann","suffix":""}],"badges":[],"createdAt":"2025-07-19 09:15:27","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7163309/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7163309/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41559-026-02999-2","type":"published","date":"2026-02-27T05:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":88100057,"identity":"e62b2480-4ab0-432f-93e9-37df97aa290a","added_by":"auto","created_at":"2025-08-01 11:19:54","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":734911,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePhylogeny of Lepidopteran GMC oxidoreductase proteins and tissue-specific expression profiles of putative Hi genes.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e, A phylogenetic tree was inferred from the alignment of 1251 protein sequences belonging to the GMC oxidoreductase family from 34 lepidopteran species, with three fungal GMC genes considered as outgroup. To assess the monophyly of the Lepidoptera GMC genes within their respective subfamilies, sequences from three Trichopteran species were included. Sequences from \u003cem\u003eDrosophila melanogaster\u003c/em\u003ewere included to serve as reference sequences for accurate classification of GMC subfamilies. The well-supported subclade (bootstrap = 100) within the GMCβ subfamily, referred to as putative Hi clade, which includes two previously reported genes from \u003cem\u003eM. sexta\u003c/em\u003e, Hi-1 (with Hi activity) and its ortholog Hi-like (without Hi activity), is highlighted in orange. The tree was analyzed by using the IQ-TREE maximum likelihood model. For the detailed tree with accession numbers and taxon, see supplementary Data 1. \u003cstrong\u003eb-e,\u003c/strong\u003eThe gene expression of putative Hi genes, which are clustered in a monophyletic group (bootstrap value = 84) within putative Hi clade (Supplementary Data 2), was measured in the fat body (FB), midgut (MG) and salivary glands (SG) of \u003cem\u003eB. mori\u003c/em\u003e (\u003cstrong\u003eb\u003c/strong\u003e), \u003cem\u003eM. sexta\u003c/em\u003e(\u003cstrong\u003ec\u003c/strong\u003e), \u003cem\u003eC. virescens\u003c/em\u003e (\u003cstrong\u003ed\u003c/strong\u003e) and \u003cem\u003eD. plexippus \u003c/em\u003e(\u003cstrong\u003ee\u003c/strong\u003e). Expression values were normalized relative to the lowest expression level within the species. Genes with Hi activity, based on the results from figure 2, are referred to as Hi next to the accession number. \u003cem\u003en\u003c/em\u003e = 3 (\u003cstrong\u003eb\u003c/strong\u003e-\u003cstrong\u003ed\u003c/strong\u003e) and \u003cem\u003en\u003c/em\u003e = 6 (\u003cstrong\u003ee\u003c/strong\u003e) biologically independent samples. Error bars are presented as mean values ± SD. n.d., not detectable.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7163309/v1/a23c662379857eef10d1cfab.png"},{"id":88101239,"identity":"3b5aa7de-5539-4f31-b6f4-0392b477b9c3","added_by":"auto","created_at":"2025-08-01 11:27:54","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":331326,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSPME-GC/MS in vitro assay for measuring (3\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eZ\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e):(2\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eE\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e)-hexenal isomerase activity using purified recombinant proteins.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea, \u003c/strong\u003eThree different quantities (0.01, 0.1, or 1 µg) of purified recombinant putative Hi proteins were incubated with \u003cem\u003eZ\u003c/em\u003e-3-hexenal (0.2 mM), and the proportion of \u003cem\u003eE\u003c/em\u003e-2-hexenal emitted from total aldehydes (\u003cem\u003eZ\u003c/em\u003e-3-hexenal + \u003cem\u003eE\u003c/em\u003e-2-hexenal) was calculated. The buffer control was used to estimate the non-enzymatic conversion rate of \u003cem\u003eZ\u003c/em\u003e-3-hexenal to \u003cem\u003eE\u003c/em\u003e-2-hexenal. The values were normalized to the buffer control and are shown as relative fold-change levels. Kruskal-Wallis nonparametric test was performed to assess significant differences between the buffer control and the three different quantities of recombinant proteins. A significant increase in \u003cem\u003eE\u003c/em\u003e-2-hexenal with increasing recombinant protein quantity is marked with an asterisk (* \u003cem\u003ep\u003c/em\u003e\u0026lt; 0.05). \u003cem\u003en\u003c/em\u003e = 3 biologically independent samples. The gray line indicates median value. Ms, \u003cem\u003eManduca sexta\u003c/em\u003e. Cv, \u003cem\u003eChloridea\u003c/em\u003e \u003cem\u003evirescen\u003c/em\u003es. Bm, \u003cem\u003eBombyx mori\u003c/em\u003e. Dp, \u003cem\u003eDanaus Plexippus\u003c/em\u003e, Pr, \u003cem\u003ePieris rapae\u003c/em\u003e. \u003cstrong\u003eb\u003c/strong\u003e, Representative extracted ion chromatograms (ion 55) from the SPME-guided assay with 1 µg of purified protein. The blue highlight on the \u003cem\u003eE\u003c/em\u003e-2-hexenal peak area indicates a distinct conversion of \u003cem\u003eZ\u003c/em\u003e-3-hexenal to \u003cem\u003eE\u003c/em\u003e-2-hexenal. \u0026nbsp;Standard compounds of \u003cem\u003eZ\u003c/em\u003e-3-hexenal (20 nmol) and \u003cem\u003eE\u003c/em\u003e-2-hexenal (5 nmol) were used to identify peak areas.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7163309/v1/f18fc2420233510f33ee3243.png"},{"id":88100059,"identity":"2f247f4f-072f-4876-a4a1-bc48e8c3a1e9","added_by":"auto","created_at":"2025-08-01 11:19:54","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":443399,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eChanges in the emission of green leaf volatiles from leaf discs of host plants following treatment with recombinant Hi proteins.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e-\u003cstrong\u003ed\u003c/strong\u003e, Leaf discs (2.4 mm diameter) from the corresponding host plants were mechanically wounded and treated with 10 µL of Milli-Q water (control), recombinant Hi protein (1 µg), or heat inactive recombinant Hi protein (1 µg). (\u003cstrong\u003ea-b\u003c/strong\u003e) Tomato; (\u003cstrong\u003ec\u003c/strong\u003e) Milkweed; (\u003cstrong\u003ed\u003c/strong\u003e) White mulberry. The composition of the \u003cem\u003eZ\u003c/em\u003e3/\u003cem\u003eE\u003c/em\u003e2\u003cem\u003e \u003c/em\u003eform of aldehydes and alcohols was determined using SPME-GC-qToF-MS. Lognormal ordinary one-way ANOVA test was performed to assess significant differences between treatments, \u003cstrong\u003ea\u003c/strong\u003e, \u003cem\u003eF\u003c/em\u003e\u003csub\u003e2, 12\u003c/sub\u003e = 240, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001; \u003cstrong\u003eb\u003c/strong\u003e, \u003cem\u003eF\u003c/em\u003e\u003csub\u003e2, 13\u003c/sub\u003e = 87,26, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001; \u003cstrong\u003ec\u003c/strong\u003e, \u003cem\u003eF\u003c/em\u003e\u003csub\u003e2, 13\u003c/sub\u003e = 12,19, \u003cem\u003ep\u003c/em\u003e = 0.001; \u003cstrong\u003ed\u003c/strong\u003e, \u003cem\u003eF\u003c/em\u003e\u003csub\u003e2, 13\u003c/sub\u003e = 8,477, \u003cem\u003ep\u003c/em\u003e = 0.004. Different letters in the center of each pie chart indicate significant differences (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05) by Tukey post-hoc test. \u003cem\u003en\u003c/em\u003e = 5-6 biologically independent samples. The corresponding dot plot is shown in supplementary Fig. 2. \u003cstrong\u003ee\u003c/strong\u003e, Illustration of a section of GLV biosynthesis pathway. The conversion of \u003cem\u003eZ\u003c/em\u003e-3-hexenal to \u003cem\u003eE\u003c/em\u003e-2-hexenal can occur either spontaneously or through the catalytic action of (3\u003cem\u003eZ\u003c/em\u003e):(2\u003cem\u003eE\u003c/em\u003e)-hexenal isomerase (Hi). Both aldehydes can be further reduced to their corresponding alcohols by cinnamaldehyde and hexenal reductase (CHR).\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7163309/v1/2a6e74f0551bc4bb24d9fcc9.png"},{"id":88100066,"identity":"521e2a21-e406-47d7-a1ff-75e5ef90eacb","added_by":"auto","created_at":"2025-08-01 11:19:54","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1141585,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFAD-dependent rearrangement of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eZ\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e-3- to \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eE\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e-2-hexenal by Lepidopteran Hi\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea-e\u003c/strong\u003e, Increased conversion rate of \u003cem\u003eE\u003c/em\u003e-2-hexenal with the addition of FAD (1 mM). Purified recombinant Hi proteins were incubated with \u003cem\u003eZ\u003c/em\u003e-3-hexenal (0.2 mM) with or without FAD (1 mM). The proportion of \u003cem\u003eE\u003c/em\u003e-2-hexenal emitted from total aldehydes (\u003cem\u003eZ\u003c/em\u003e-3-hexenal + \u003cem\u003eE\u003c/em\u003e-2-hexenal) was calculated. Statistically significant differences between treatments were assessed using ratio paired two-tailed t-tests. (\u003cstrong\u003ea\u003c/strong\u003e) \u003cem\u003en\u003c/em\u003e = 6 biologically independent samples, t = 1.01, \u003cem\u003ep\u003c/em\u003e = 0.36; (\u003cstrong\u003eb\u003c/strong\u003e) \u003cem\u003en\u003c/em\u003e = 6, t = 3.55, \u003cem\u003ep\u003c/em\u003e = 0.017; (\u003cstrong\u003ec\u003c/strong\u003e) \u003cem\u003en\u003c/em\u003e = 5, t = 7.16, \u003cem\u003ep\u003c/em\u003e = 0.002 (\u003cstrong\u003ed\u003c/strong\u003e) \u003cem\u003en\u003c/em\u003e = 5, t = 5.94, \u003cem\u003ep\u003c/em\u003e = 0.004; (\u003cstrong\u003ee\u003c/strong\u003e) \u003cem\u003en\u003c/em\u003e = 3, t = 8.66, \u003cem\u003ep\u003c/em\u003e = 0.013. \u003cstrong\u003ea’-e’\u003c/strong\u003e, Representative extracted ion chromatograms (ion 69) from the SPME-GC/MS. \u003cstrong\u003ef-g\u003c/strong\u003e, Superposition of the MsHi-1 protein structure with FAD docking (predicted using AlphaFold 3) with experimental structures of two reference GMC oxidoreductases, aryl-alcohol oxidase from \u003cem\u003ePleurotus eryngii\u003c/em\u003e (PDB: 3FIM) and glucose dehydrogenase from \u003cem\u003eAspergillus flavus\u003c/em\u003e (PDB: 4YNT). \u003cstrong\u003ef’-g’\u003c/strong\u003e, Detailed view of the FAD binding pocket shows an N-terminal residue that binds FAD, and two conserved C-terminal catalytic residues of GMC oxidoreductases. These N-terminal residues H135 (MsHI-1), Y92 (3FIM) and G94 (4YNT) form a hydrogen bond (green dashed line) with the O4 atom of the FAD isoalloxazine ring. Structural alignment and RMSD calculations were performed using Mol* (https://molstar.org/). \u003cstrong\u003eh\u003c/strong\u003e, Sequence logos representing the conservation of catalytic residues among Lepidopteran Hi homologs. The x-axis shows the residue position in MsHi-1. Catalytic residues selected for mutagenesis are highlighted in blue. Sequence logos were generated using WebLogo 3. \u003cstrong\u003ei\u003c/strong\u003e, Proportion of \u003cem\u003eE\u003c/em\u003e-2-hexenal emitted from total aldehydes (\u003cem\u003eZ\u003c/em\u003e-3-hexenal +\u003cem\u003e E\u003c/em\u003e-2-hexenal) after incubation of \u003cem\u003eZ\u003c/em\u003e-3-hexenal (0.2 mM) with either wild-type (WT) or mutant recombinant MsHi-1 proteins (0.1 µg or 0.5 µg, as indicated by 5×). Significant differences between the buffer control, WT, and mutant proteins were determined using one-way ANOVA (F\u003csub\u003e7,25\u003c/sub\u003e=144.3, \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.0001), followed by Tukey's HSD post-hoc test. Different letters indicate significant differences (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05). \u003cem\u003en\u003c/em\u003e = 6 biologically independent samples for buffer control; \u003cem\u003en\u003c/em\u003e = 5 for WT and H521A; \u003cem\u003en\u003c/em\u003e = 3 for H521A and H559A. \u003cstrong\u003ej\u003c/strong\u003e, Kinetic parameters of wild-type and H521A mutant MsHi-1.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7163309/v1/bc90e4d0ffda0042bb8fe47c.png"},{"id":88100064,"identity":"9742eff8-cbd1-4be2-bc96-63424ddd0b3a","added_by":"auto","created_at":"2025-08-01 11:19:54","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1339367,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDivergence time of Hi proteins in plants and Lepidopterans\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e, A maximum likelihood phylogenetic tree was inferred from the protein sequences that clustered within the Hi-potential branch of the cupin superfamily tree (Highlighted in Supplementary Fig. 6), alongside representative cupin superfamily proteins from outside this branch (labeled as Germin, Vicilin, Legumin, Globulin). The clade Hi (highlighted in red) includes proteins with conserved Hi catalytic residues (His/Lys/Tyr). This clade is found exclusively in mesangiosperms, based on evidence that proteins from basal angiosperms (ANA grade), highlighted in blue, lack these catalytic residues. The bootstrap values were inferred from RAxML-NG and IQ-TREE 2. \u003cstrong\u003eb\u003c/strong\u003e, Sequence logos represent the alignment of proteins from each subclade. The Hi catalytic residues (His/Lys/Tyr) in proteins of the Hi clade are highlighted in blue. \u003cstrong\u003ec\u003c/strong\u003e, Distribution survey of Hi proteins across eudicot families. Filled circles indicate Hi presence in all analyzed species within a family, half-filled circles show presence in some species, and empty circles indicate complete absence. Notably, all analyzed species from both Brassicaceae and Caricaceae families (highlighted in green), members of the Brassicales order, lack Hi proteins. Sequence logos were generated using WebLogo 3. \u003cstrong\u003ed\u003c/strong\u003e, The chronogram illustrates the emergence and distribution of Lepidopteran Hi across lineages in relation to major geological events. Hi proteins appeared in the Apoditrysia lineage of Lepidoptera (132.1–105.6 Ma), coinciding with the angiosperm radiation (125–90 Ma). In plants, Hi emerged in Mesangiosperms (192.2–166.4 Ma). The presence of Hi in Lepidopteran species was inferred from proteins that clustered within the putative Hi clade in the phylogenetic tree (Fig. 1a). Presence of Hi in different Lepidopteran superfamilies is indicated using filled, half-filled, and empty circles, similar to panel \u003cstrong\u003ec\u003c/strong\u003e. Estimates of Lepidopteran and plant divergence times are based on Kawahara et al. (2019), Peris and Condamine (2024), and Yang et al. (2020).\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7163309/v1/d3f28efa8a13335940e2220c.png"},{"id":103637096,"identity":"43022eda-19d9-4ce8-b4b7-bd27161c09ff","added_by":"auto","created_at":"2026-02-28 08:05:25","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4974762,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7163309/v1/46f4efa2-4205-4192-9dbb-0bd76f7108c9.pdf"},{"id":88101241,"identity":"47290a6e-9114-4159-bab9-9a4aa80cf307","added_by":"auto","created_at":"2025-08-01 11:27:54","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3323459,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"SupplementaryInformation20250719Final.docx","url":"https://assets-eu.researchsquare.com/files/rs-7163309/v1/692deaf46e4ee47381bafb4c.docx"},{"id":88101238,"identity":"5282f652-349f-4dd6-afbf-f67411717699","added_by":"auto","created_at":"2025-08-01 11:27:54","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":377628,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Data 1\u003c/p\u003e","description":"","filename":"SupplementaryData1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7163309/v1/0048d20facfb6ee6c1eae2e8.pdf"},{"id":88100058,"identity":"3c02d2f2-06d4-413f-8fdd-fe1f058a2b18","added_by":"auto","created_at":"2025-08-01 11:19:54","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":58394,"visible":true,"origin":"","legend":"Supplementary Data 2","description":"","filename":"SupplementaryData2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7163309/v1/f3be71d1c3e391837b6acde6.pdf"},{"id":88100062,"identity":"91d517f8-2211-41a5-bf48-b399082c7e89","added_by":"auto","created_at":"2025-08-01 11:19:54","extension":"xlsx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":37114,"visible":true,"origin":"","legend":"Supplementary Table 1","description":"","filename":"SupplementaryTable1InsectGMC.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7163309/v1/4c0d66a8ad78301be4b6c3a5.xlsx"},{"id":88101240,"identity":"6e0e82ec-4024-4e90-96c0-8265b47a1832","added_by":"auto","created_at":"2025-08-01 11:27:54","extension":"xlsx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":175468,"visible":true,"origin":"","legend":"Supplementary Table 2","description":"","filename":"SupplementaryTable2Plantcupin.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7163309/v1/7bb2ea4f2246fbff496bfcc4.xlsx"},{"id":88100063,"identity":"741bd87c-e1c9-401e-9560-d2cc505ad531","added_by":"auto","created_at":"2025-08-01 11:19:54","extension":"xlsx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":15071,"visible":true,"origin":"","legend":"Supplementary Table 3","description":"","filename":"SupplementaryTable3primersandchemicalreagents.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7163309/v1/48feb14309d8e738226c905a.xlsx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"FAD-dependent hexenal isomerases in Lepidoptera evolved convergently with plant-derived hexenal isomerases","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePlants interact with their environment through the emission of volatile organic compounds (VOCs) (Dudareva et al., 2006). \u0026nbsp;Among these, green leaf volatiles (GLVs) form a group of six‑carbon (C\u003csub\u003e6\u003c/sub\u003e) molecules that impart the\u0026nbsp;\u0026ldquo;grassy\u0026rdquo;\u0026nbsp;scent to foliage\u0026nbsp;(Hatanaka et al., 1987). Produced by most green plants, GLVs are emitted within seconds of mechanical wounding, herbivore feeding, or various abiotic stresses\u0026nbsp;(Matsui and Engelberth, 2022). They contribute to a plant\u0026rsquo;s direct defense by deterring herbivores and inhibiting pathogenic infection, and to a plant\u0026rsquo;s indirect defense, by attracting natural enemies of herbivores\u0026nbsp;(Matsui and Engelberth, 2022). Owing to this near‑instantaneous release, GLVs also serve as ecological\u0026nbsp;\u0026ldquo;alarm signals\u0026rdquo;\u0026nbsp;that prime, and in some cases directly trigger, plant defenses in neighboring tissues and nearby plants\u0026nbsp;(Ameye et al., 2018).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe biosynthesis of GLVs is initiated from polyunsaturated fatty acids, primarily\u0026nbsp;\u0026alpha;-linolenic and linoleic acid, via the enzymatic action of lipoxygenases (LOX) (Matsui, 2006). LOXs oxygenate these fatty acids, forming hydroperoxides, subsequently cleaved by hydroperoxide lyases (HPL) into \u003cem\u003eZ\u003c/em\u003e-3-hexenal (Z3AL) or hexanal, depending on the initial fatty acid substrate. Within the GLV biosynthesis pathway, the re-arrangement of Z3AL to \u003cem\u003eE\u003c/em\u003e‑2‑hexenal (E2AL) is particularly important because it reshapes the \u003cem\u003eZ\u003c/em\u003e3/\u003cem\u003eE\u003c/em\u003e2 ratio and thereby modulates the downstream formation of the corresponding alcohols and acetates (Fig.\u0026nbsp;3e)\u0026nbsp;(Engelberth and Engelberth, 2020). Although Z3AL is emitted first, E2AL exhibits stronger antimicrobial activity and a more pronounced signaling capacity owing to its higher reactivity toward nucleophilic biomolecules\u0026nbsp;(Alm\u0026eacute;ras et al., 2003; Croft et al., 1993; Hyun et al., 2022). This re-arrangement from Z3AL to E2AL could occur spontaneously due to the intrinsic instability of Z3AL but becomes far more efficient when catalyzed by the (3\u003cem\u003eZ\u003c/em\u003e):(2\u003cem\u003eE\u003c/em\u003e)-hexenal isomerase (Hi). Plant-derived Hi enzymes have been identified in various species, including cucumber, tomato, and rice, and belong to the cupin superfamily\u0026nbsp;(Chen et al., 2022; Kunishima et al., 2016; Phillips et al., 1979; Spyropoulou et al., 2017)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIntriguingly, larvae of the hawk moth (\u003cem\u003eManduca sexta\u003c/em\u003e) secrete a functionally analogous but phylogenetically distinct Hi protein in their oral secretions, affiliated with the GMC oxidoreductase that converts plant-produced Z3AL into E2AL while feeding (Lin et al., 2023). Such conversion alters the composition of GLVs, which guides female moths in choosing oviposition sites (Allmann et al., 2013), but paradoxically also serves as the cue that attracts their natural enemies (Allmann and Baldwin, 2010). Hi activity independently emerged in plants and Lepidoptera via distinct protein families, representing a compelling case of convergent evolution. Notably, in vitro assays reveal that Hi activity varies among Lepidopteran species (Jones et al., 2021; Lin et al., 2023). Semi‑field trials confirm this pattern: \u003cem\u003eManduca sexta\u003c/em\u003e releases a pronounced burst of E2AL when feeding on solanaceous hosts, whereas \u003cem\u003eChloridea virescens\u003c/em\u003e produces only a negligible Z3AL to E2AL conversion (Paudel Timilsena et al., 2020).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDespite this ecological and evolutionary significance, the origins and functional diversity of Hi are still poorly understood in Lepidoptera. Only a single Hi protein from \u003cem\u003eManduca sexta\u0026nbsp;\u003c/em\u003ehas been biochemically and functionally characterized to date (Lin et al., 2023), and inter‑species variation in Hi activity remains unexplored. In this study, we combine phylogenetic and functional analysis to reconstruct the evolutionary history of Lepidopteran Hi. We map its taxonomic distribution and compare the enzymatic activity of Hi homologs from multiple Lepidopteran taxa. We also identify the FAD‑binding and catalytic motifs that are crucial for Lepidopteran Hi activity. Finally, we chart Hi evolution across 34 Lepidopteran species and 183 species in the green lineage (Viridiplantae), revealing that both plant and Lepidopteran Hi arose during the Cretaceous angiosperm radiation.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003ePhylogenetic and functional analysis of putative (3\u003cem\u003eZ)\u003c/em\u003e:(2\u003cem\u003eE\u003c/em\u003e)-hexenal isomerase (Hi) genes in Lepidoptera\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe first Lepidopteran Hi characterized in our earlier study\u0026nbsp;(Lin et al., 2023), \u003cem\u003eM\u003c/em\u003e. \u003cem\u003esexta\u0026nbsp;\u003c/em\u003eHi (MsHi-1), is a member of the GMC oxidoreductase family. This protein family is defined by conserved N-terminal (PF00732) and C-terminal (PF05199) domains. To identify related homologs in Lepidoptera, we used profile hidden Markov models (HMMs) based on these domains. Subsequently, we constructed a maximum-likelihood (ML) phylogram comprising 1251 GMC oxidoreductases from 34 lepidopteran species, spanning both non-Ditrysia and Ditrysia lineages (Supplementary Table 1). Consistent with previous studies\u0026nbsp;(Iida et al., 2007; Sun et al., 2012), the Lepidopteran GMC genes grouped into distinct subfamilies, showing significant expansion within the GMC\u0026beta; subfamily (Fig. 1a). MsHi-1, along with its gene duplicate lacking Hi activity, MsHi-like, clustered within a well-supported subclade (bootstrap value = 100) within the GMC\u0026beta; subfamily (Fig. 1a, and Supplementary Data 1, highlighted in orange). To determine whether this subclade might represent a broader set of Hi homologs, we selected all homologous genes from four species -\u0026nbsp;\u003cem\u003eBombyx mori\u003c/em\u003e, \u003cem\u003eManduca sexta\u003c/em\u003e, \u003cem\u003eChloridea virescens\u003c/em\u003e and \u003cem\u003eDanaus plexippus\u003c/em\u003e - that cluster within a monophyletic group (bootstrap value = 84) containing MsHi-1 (XP_030035814) (Supplementary Data 2), for further functional characterization. Notably, oral secretions (OS) of these four species had previously been shown to exhibit Hi activity\u0026nbsp;(Lin et al., 2023).\u003c/p\u003e\n\u003cp\u003eSince the salivary glands (SG) and midgut (MG) are primary sources of enzymes in OS, we first investigated whether the selected genes potentially encode Hi enzymes in OS by comparing their expression levels in the SG and MG to the non-OS source tissue, fat body (FB). The majority of candidate genes were highly expressed in the SG, whereas MsHi-like in \u003cem\u003eM. sexta\u003c/em\u003e and DpHi-1 in \u003cem\u003eD. plexippus\u003c/em\u003e showed comparatively higher expression in the MG (Fig. 1b-e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo assess the Hi activity of these putative homologs, we cloned the corresponding cDNAs into pGEX vectors and expressed the recombinant proteins in \u003cem\u003eE\u003c/em\u003e. \u003cem\u003ecoli\u0026nbsp;\u003c/em\u003eBL21 (Supplementary Fig. 1). A sole homolog from \u003cem\u003ePieris rapae,\u003c/em\u003e present in the putative Hi clade (Supplementary Data 2), was included as a negative control, as previous studies have shown no Hi activity in the OS of this species (Jones et al., 2021; Lin et al., 2023). After expression and purification, we tested three protein quantities (0.01, 0.1, and 1 \u0026micro;g) for their ability to convert \u003cem\u003eZ\u003c/em\u003e-3-hexenal (Z3AL) into \u003cem\u003eE\u003c/em\u003e-2-hexenal (E2AL) in vitro using SPME-GC-qToF-MS. At least one protein from each species showed a concentration-dependent increase in Hi activity (Fig. 2a). In \u003cem\u003eM\u003c/em\u003e. \u003cem\u003esexta\u003c/em\u003e, a second homolog (MsHi-2) also displayed Hi activity, albeit at lower levels than MsHi-1. Representative chromatograms of the 1 \u0026micro;g protein reactions confirm E2AL formation in active homologs (Fig. 2b). As expected, the \u003cem\u003eP\u003c/em\u003e. \u003cem\u003erapae\u0026rsquo;\u003c/em\u003es homolog showed no detectable Hi activity.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSpecies-specific Hi activity determines the magnitude of \u003cem\u003eE\u003c/em\u003e-2-GLV emissions from wounded host plants\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHi activity has been confirmed in the OS of numerous lepidopteran species through ex vivo approaches (Jones et al., 2021; Lin et al., 2023), and here we further identify active Hi homologs. However, only \u003cem\u003eM\u003c/em\u003e. \u003cem\u003esexta\u003c/em\u003e has been reported to induce a pronounced increase in \u003cem\u003eE\u003c/em\u003e-2-GLVs when feeding on host plants, thereby influencing multitrophic interactions (Allmann and Baldwin, 2010; Allmann et al., 2013). To explore this discrepancy, we applied equal amounts of each species\u0026rsquo; recombinant Hi protein to wounded leaves of their respective host plants and measured the emission of GLVs. Treatment with MsHi-1 on wounded tomato leaves led to a significant rise in E2AL emissions, compared to water or heat-inactivated MsHi-1 controls (Fig. 3a; Supplementary Fig. 2a). Furthermore, a notable increase in \u003cem\u003eE\u003c/em\u003e-2-hexenol (E2OL)\u0026mdash;a reduction product of E2AL by the plant cinnamaldehyde and hexenal reductase (CHR) (Fig. 3e)\u0026mdash;was also observed (Fig. 3a\u0026rsquo;; Supplementary Fig. 2a\u0026rsquo;). By contrast, treatment with CvHi-1 on wounded tomato leaves induced only modest increases in both E2AL and E2OL (Fig. 3b and 3b\u0026rsquo;; Supplementary Fig. 2b and 2b\u0026rsquo;). Similarly, DpHi-1 on wounded milkweed leaves and BmHi-1 on wounded mulberry leaves elicited only slight increases in E2AL; E2OL was undetectable in these samples (Fig. 3c and 3d; Supplementary Fig. 2c and 2d). Taken together, these results indicate that while Hi homologs are widespread in Lepidoptera, MsHi-1 in \u003cem\u003eM\u003c/em\u003e. \u003cem\u003esexta\u003c/em\u003e appears uniquely efficient at converting Z3AL into E2AL during feeding on one of their host plants, resulting in a pronounced increase in other \u003cem\u003eE\u003c/em\u003e-2-GLVs.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLepidopteran (3\u003cem\u003eZ\u003c/em\u003e):(2\u003cem\u003eE\u003c/em\u003e)-hexenal isomerase activity is FAD-dependent\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGMC oxidoreductases typically catalyze redox reactions using flavin adenine dinucleotide (FAD) as a cofactor (Cavener, 1992). Although the conversion of Z3AL to E2AL does not involve a net redox change, we hypothesized that the Lepidopteran Hi still requires FAD for its isomerase activity. To test this, we measured Hi activity in the presence or absence of supplemental FAD. Control reactions showed that FAD alone did not affect the non-enzymatic conversion of Z3AL to E2AL (Fig. 4a, a\u0026apos;). However, the addition of FAD to enzymatic reactions moderately increased Hi activity across all tested homologs (Fig.\u0026nbsp;4b\u0026ndash;e), as indicated by higher E2AL peak areas in the chromatograms (Fig.\u0026nbsp;4b\u0026apos;\u0026ndash;e\u0026apos;).\u003c/p\u003e\n\u003cp\u003eSince Hi activity was still detectable even without the addition of extra FAD, we inferred that FAD was already bound to the enzyme during expression in \u003cem\u003eE\u003c/em\u003e. \u003cem\u003ecoli\u003c/em\u003e and remained associated through the purification process. To further validate the importance of FAD in Hi activity, we generated site-directed mutants at key residues predicted to be involved in FAD binding or catalytic activity. Using AlphaFold 3 (Abramson et al., 2024) to model MsHi-1 with docked FAD, we identified putative FAD-binding residues. We then compared the predicted structure of MsHi-1 with two experimentally determined GMC oxidoreductase structures: aryl-alcohol oxidase from \u003cem\u003ePleurotus eryngii\u003c/em\u003e (PDB: 3FIM) and glucose dehydrogenase from \u003cem\u003eAspergillus flavus\u003c/em\u003e (PDB: 4YNT) (Fig. 4f, g). Superimposing their FAD-binding domains onto the predicted FAD-binding domain of MsHi-1 revealed high structural similarity, with root-mean-square deviation (RMSD) of 0.22 \u0026Aring; and 0.48 \u0026Aring;, respectively (Fig. 4f\u0026apos;, g\u0026apos;). Y92 in 3FIM and G94 in 4YNT, previously reported as essential residues for FAD binding (Fern\u0026aacute;ndez et al., 2009; Yoshida et al., 2015) were found to correspond to a conserved histidine (H135) in MsHi-1 (Fig. 4f\u0026apos;, g\u0026apos;) and other Lepidopteran Hi homologs (Fig. 4h and Supplementary Fig. 3). This histidine could potentially form a hydrogen bond with the O4 atom on the isoalloxazine ring of FAD, similar to Y92 and G94 in 3FIM and 4YNT (Fig. 4f\u0026apos;, g\u0026apos;), respectively. Additionally, comparative analysis of conserved C-terminal GMC oxidoreductase active-site motifs (Liu et al., 2013) showed MsHi-1 and other Lepidopteran Hi generally contain H\u0026ndash;N pairs (e.g., H521/N559 in MsHi-1) (Fig. 4h, and Supplementary Fig. 3) where H\u0026ndash;H pairs occur in 3FIM (H502/H546) and 4YNT (H505/H548). These two conserved residues are known to be crucial for substrate positioning and electron transfer in GMC oxidoreductases (Ohta et al., 2006; Wongnate and Chaiyen, 2013).\u003c/p\u003e\n\u003cp\u003eWe further introduced alanine substitutions at H135, H521, and N559 to confirm the role of these FAD-interactive residues in Hi activity (Supplementary Fig. 4). The H135A and N559A mutants completely lost Hi activity, even when increasing the amount of protein five times (0.5\u0026nbsp;\u0026micro;g) (Fig. 4i). The H521A mutant still converted Z3AL to E2AL, but at reduced rates. Enzyme kinetics showed an increased Km (0.04\u0026nbsp;\u0026rarr;\u0026nbsp;1.01 mM) and a ~25-fold drop in catalytic efficiency (359.6\u0026nbsp;\u0026rarr;\u0026nbsp;14.0 mM⁻\u0026sup1; s⁻\u0026sup1;), while Kcat remained roughly unchanged (13.8 vs. 14.1 s⁻\u0026sup1;) (Fig. 4j). In planta assays, using wounded tomato leaves, confirmed these findings (Supplementary Fig. 5). Mechanically wounded leaves that were treated with recombinant protein of either MsHi\u003csub\u003eH135A\u003c/sub\u003e or MsHi\u003csub\u003eN559A\u003c/sub\u003e emitted predominantly Z3AL, resembling water-treated controls. MsHi\u003csub\u003eH521A\u003c/sub\u003e-treated leaves released more E2AL than those treated with the other two mutants, but still significantly less than wildtype MsHi-1-treated leaves. These findings support the conclusion that the cofactor FAD is essential for Lepidopteran Hi activity.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDistribution and evolutionary history of (3\u003cem\u003eZ\u003c/em\u003e):(2\u003cem\u003eE\u003c/em\u003e)-hexenal isomerase in plants and lepidopterans\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePrevious studies have shown that in plants, an enzyme of the cupin superfamily catalyzes the conversion of Z3AL to E2AL (Kunishima et al., 2016; Spyropoulou et al., 2017). Both plant- and Lepidopteran-derived Hi display promiscuous activity toward other \u003cem\u003eZ\u003c/em\u003e-3-aldehydes, such as \u003cem\u003eZ\u003c/em\u003e-3-octenal and \u003cem\u003eZ\u003c/em\u003e-3-nonenal (Lin et al., 2023; Spyropoulou et al., 2017). This functional convergence in plants and insects raises intriguing questions about when and how Hi evolved independently in these distinct lineages.\u003c/p\u003e\n\u003cp\u003eIn plants, the cupin superfamily comprises structurally conserved proteins characterized by a\u0026nbsp;\u0026szlig;-barrel fold, with each member containing either one or two cupin domains (Dunwell et al., 2001). We first searched for proteins containing the cupin domain (Pfam: PF07883) across 183 representative Viridiplantae (green lineage) species (Supplementary Table 2a). This search yielded 3734 sequences, which we used to construct a phylogenetic tree, alongside 31 reference sequences from well-characterized cupin superfamily members, including Germin, Vicilin, Legumin, Globulin, and plant Hi proteins (Supplementary Table 2b). The previously characterized plant Hi proteins clustered within a distinct subclade of the cupin superfamily, which we refer to as the \u0026ldquo;Hi-potential clade\u0026rdquo; (Supplementary Fig. 6). Furthermore, our analysis suggests that the proteins in the Hi-potential clade first emerged in embryophyte (land plant) hornworts, were subsequently lost, and later retained in Bryopsida (Supplementary Fig. 7 and Supplementary Table 2c).\u003c/p\u003e\n\u003cp\u003eSubsequently, a rooted phylogenetic tree was constructed focusing specifically on the sequences within the \u0026quot;Hi-potential clade\u0026quot; (260 sequences, Supplementary Table 2c). ). To elucidate their evolutionary relationships (Fig. 5a), we incorporated reference sequences from well-characterized cupin subfamilies\u0026mdash;Germin, Vicilin, Globulin, and Legumin\u0026mdash;that lie outside the Hi-potential clade and serve as outgroups. The rooted ML tree grouped the proteins in the Hi-potential clade into seven clades (Fig. 5a). The four clades located near the root of the tree (Clade I to IV) include all identified orthologs from ferns, spikemosses, liverworts, and hornworts. The remaining sequences formed three distinct crown clades: Clade V contains all gymnosperm orthologs, Clade VI includes angiosperm orthologs and the Clade Hi, located at the top of the tree, contains orthologs from mesangiosperms (core angiosperms).\u003c/p\u003e\n\u003cp\u003eOur data indicate that only proteins in the pink-shaded \u0026ldquo;clade Hi\u0026rdquo; possess the critical catalytic residues (H/K/Y) required for Hi activity (Fig. 5b) \u0026nbsp;(Kunishima et al., 2016). Moreover, all proteins in clade Hi are derived from mesangiosperms, whereas proteins from basal angiosperms (\u003cem\u003eAmborella trichopoda and Nymphaea colorata\u003c/em\u003e, ANA grade) form a separate sister clade (clade VI) that lacks these catalytic residues (Fig.\u0026nbsp;5a, blue highlight). Interestingly, among 16 monocot species examined, only rice (\u003cem\u003eOryza sativa\u003c/em\u003e) possessed proteins within clade Hi (Fig.\u0026nbsp;5a). Hi proteins were entirely absent in six of the sixteen eudicot families analyzed\u0026mdash;including Brassicaceae and Caricaceae from the Brassicales order\u0026mdash;across the 31 species examined (Fig. 5c).\u003c/p\u003e\n\u003cp\u003eIn parallel, we expanded our analysis of the lepidopteran GMC\u0026beta; oxidoreductase subclade, enriched in Hi homologs (Fig. 1a). Hi homologs were absent in all non-Ditrysian superfamilies\u0026mdash;Micropterigoidea, Hepialoidea, Tischerioidea, and Palaephatoidea\u0026mdash;collectively representing the early-diverging lineages of Lepidoptera, as well as in the basal Ditrysian superfamilies Tineoidea and Yponomeutoidea (Fig. 5d and Supplementary Table 1a). Instead, Hi homologs were exclusively identified within the Apoditrysia lineages, including Obtectomera and Macroheterocera. This pattern suggests that Hi arose relatively recently in the Lepidopteran lineage, potentially coinciding with the diversification of Apoditrysia. Given that Apoditrysia is estimated to have a crown-group origin in the Early Cretaceous (~118.5 Ma) (Kawahara et al., 2019), a period marked by rapid angiosperm radiation, it is plausible that the emergence of Lepidopteran Hi reflects an adaptive response to increasingly novel and chemically diverse host plants. \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we provide new insights into the evolutionary history and taxonomic distribution of plant and insect Hi genes and describe the enzymatic versatility of Hi genes in Lepidoptera. Functional assays of Lepidopteran Hi homologs revealed that at least one gene of each examined species exhibited measurable Hi activity, indicating that these enzymes may contribute to the herbivore-driven production of E2AL in various lepidopteran species. Notably, Hi-1 from \u003cem\u003eM\u003c/em\u003e. \u003cem\u003esexta\u003c/em\u003e induced significantly higher levels of E2AL and downstream E-2-GLVs in host plants compared to Hi from other three Lepidopteran species (Fig. 3a). These results, obtained in situ by applying recombinant Hi proteins onto mechanically wounded leaves of host plants, support previous observations that plants fed on by \u003cem\u003eM\u003c/em\u003e. \u003cem\u003esexta\u003c/em\u003e release significantly higher levels of E2AL overtime compared to plants fed on by \u003cem\u003eC\u003c/em\u003e. \u003cem\u003evirescens\u003c/em\u003e (Paudel Timilsena et al., 2020). Our findings also suggest that Hi activity is not universally linked to trophic specialization, as Hi proteins from two specialist herbivores, \u003cem\u003eD\u003c/em\u003e. \u003cem\u003eplexippus\u003c/em\u003e and \u003cem\u003eB\u003c/em\u003e. \u003cem\u003emori\u003c/em\u003e, induced much lower E2AL emissions than \u003cem\u003eM\u003c/em\u003e. \u003cem\u003esexta\u0026rsquo;s\u003c/em\u003e Hi-1 (Fig. 3c and 3d).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe ecological relevance of Hi is well established: shifts in the ratio of \u003cem\u003eZ\u003c/em\u003e-3- to \u003cem\u003eE\u003c/em\u003e-2-GLVs, driven by Hi activity, serve as cues for both ovipositing moths and natural enemies of the herbivore (Allmann and Baldwin, 2010; Allmann et al., 2013). However, whether Hi ultimately benefits the insect or the plant remains unclear, as its effects can be both advantageous and detrimental depending on context. Similarly, other OS effectors, such as fatty acid dehydratases (FHDs), hexenal‑trapping molecules (HALTs) and glucose oxidase (GOX), can reduce GLV emissions (Jones et al., 2021), which may dampen the strength of volatile cues in the environment. Notably, suppression of GLVs via FHD has been shown to decrease parasitoid attraction, potentially benefiting the herbivore by reducing attack risk (Takai et al., 2018). Beyond its ecological effects, we previously demonstrated that Hi enzymes act on a range of \u003cem\u003eZ\u003c/em\u003e-3-alkenal substrates, and that \u003cem\u003ehi\u003c/em\u003e mutants of \u003cem\u003eM. sexta\u003c/em\u003e raised on artificial diet lacking GLVs developed more slowly and showed increased rates of adult abnormalities compared to wild-type insects (Lin et al., 2023). This suggests that Hi may also serve internal physiological roles independent of interactions with GLV-producing plants. Together, these findings highlight that variation in Hi activity among lepidopteran species likely reflects a combination of ecological, physiological, and evolutionary factors.\u003c/p\u003e\n\u003cp\u003eHomology analysis indicates that Lepidopteran Hi proteins belong to the GMC oxidoreductase protein family. Given its involvement in an electrophilic isomerization reaction, this enzyme could be classified as an intramolecular oxidoreductase (EC 5.3). A notable example of such an intramolecular oxidoreductase is isopentenyl pyrophosphate (IPP) isomerase from plants, which participates in isoprenoid biosynthesis. This enzyme catalyzes the isomerization of IPP into its electrophilic isomer, dimethylallyl diphosphate (DMAPP), via a protonation/deprotonation mechanism (Berthelot et al., 2012; Reardon and Abeles, 1986). Our results demonstrated that FAD is essential for Lepidopteran Hi activity, even though there is no net redox change between the substrate and the product (Fig. 4). The role of FAD in this reaction may resemble its function in polyunsaturated fatty acid isomerase (PAI) from \u003cem\u003ePropionibacterium acnes\u003c/em\u003e (Liavonchanka and Feussner, 2008; Liavonchanka et al., 2009). In PAI, FAD facilitates the double bond isomerization of linoleic acid to conjugated linoleic acid through an ionic mechanism. During this reaction, FAD stabilizes a carbocation intermediate by interacting with the substrate\u0026apos;s double bonds via its redox-active isoalloxazine ring. Another example of a FAD-dependent non-redox reaction is carotene cis-trans isomerase (CRTISO) in plants, where FAD facilitates the isomerization of prolycopene to all-trans-lycopene during carotenoid biosynthesis (Eggers et al., 2021; Yu et al., 2011). These examples suggest that FAD in Lepidopteran Hi may facilitate a coupled isomerization and double bond migration from \u003cem\u003eZ\u003c/em\u003e-3 to \u003cem\u003eE\u003c/em\u003e-2-aldehydes by stabilizing the transient enolate intermediate and aiding proton abstraction.\u003c/p\u003e\n\u003cp\u003eThe Z3AL isomerization by plant Hi has previously been reported to involve a keto-enol tautomerism mechanism (Kunishima et al., 2016). The process is mediated by a conserved catalytic histidine (Fig. 5b), which abstracts a proton from the C2 position of Z3AL, forming a transient enolate intermediate. This enolate structure allows the electron density to shift, facilitating the formation of a keto-like tautomer. We identified an N-terminal histidine in Lepidoptera Hi (Fig. 4h and Supplementary Fig. 3), which is critical for Hi activity. As this conserved histidine has been previously reported in other GMC oxidoreductases to abstract the proton from the substrate (Liu et al., 2013; Yoshida et al., 2015), this histidine may perform a role that is similar to a catalytic histidine in plant Hi. Some of the inactive Hi homologs show substitution at catalytic residues such as leucine and aspartic acid replacing the N-terminal histidine in C. \u003cem\u003evirescens\u003c/em\u003e\u0026lsquo;s PCG65132 and PCG76483, or alanine, leucine, or tyrosine replacing the C-terminal histidine critical for FAD binding in \u003cem\u003eC\u003c/em\u003e. \u003cem\u003evirescens\u003c/em\u003e\u0026lsquo;s PCG69794, PCG76483 and \u003cem\u003eP.\u003c/em\u003e \u003cem\u003erapae\u003c/em\u003e\u0026rsquo;s homolog, respectively (Supplementary Fig. 3). Nevertheless, several Lepidopteran Hi proteins with all three conserved FAD binding and catalytic sites are still unable to rearrange Z3AL to E2AL (Fig. 2 and Supplementary Fig. 3). This includes the gene duplicate of MsHi-1, MsHi-like (Fig. 2), which shares 85% coding sequence identity. A similar observation was made in a plant Hi: one of the \u003cem\u003eCucumis sativus\u003c/em\u003e Hi proteins (Cs033080/ XP_011651276), which showed no activity (Spyropoulou et al., 2017), despite possessing all catalytic HKY residues and clustering within the clade Hi in the phylogenetic tree (Fig. 5a and Supplementary Table 2b). Future studies are needed to compare and test the roles of surrounding residues directly involved in Z3AL interactions between active and inactive Lepidopteran Hi, as well as Hi with varying activity levels. Insights from plant Hi suggest that a tyrosine near the catalytic histidine (Tyr-128 in \u003cem\u003eC\u003c/em\u003e. \u003cem\u003eannuum\u003c/em\u003e) may form the substrate binding pocket, highlighting the importance of surrounding residues in determining Hi functionality (Kunishima et al., 2016).\u003c/p\u003e\n\u003cp\u003eA previous study has shown that mechanical damage in Cucurbitaceae and Fabaceae species leads to the emission of a higher proportion of E2AL compared to Z3AL (Engelberth and Engelberth, 2020). Consistent with this, we found that all seven Cucurbitaceae and Fabaceae species in our phylogenetic analysis possess plant Hi homologs containing the catalytic HKY residues (Fig. 5c). In contrast, five Brassicaceae or Caricaceae species within the Brassicales order lack Hi genes (Fig. 5c). This is in agreement with earlier reports that Brassicales species emit Z3AL rather than E2AL upon mechanical damage (Jones et al., 2021). Interestingly, two Lepidopteran specialists that feed on Brassicales\u0026mdash;\u003cem\u003eP.\u003c/em\u003e \u003cem\u003erapae\u003c/em\u003e and \u003cem\u003ePlutella\u003c/em\u003e \u003cem\u003exylostella\u003c/em\u003e\u0026mdash;also lack a functional Hi. Although \u003cem\u003eP\u003c/em\u003e. \u003cem\u003erapae\u003c/em\u003e possesses a sole homolog clustering within the putative Hi subclade (Fig. 1a and Supplementary Data 2), no activity was detected either from the recombinant protein (Fig. 2a) or from the oral secretions (Lin et al., 2023). Meanwhile, \u003cem\u003eP\u003c/em\u003e. \u003cem\u003exylostella\u003c/em\u003e has no representative gene within the putative Hi clade in the phylogenetic tree (Supplementary Tab 1a). One plausible explanation is that the glucosinolate-based defenses of Brassicales reduce the ecological importance of Z3AL to E2AL conversion, thereby relaxing selection for Hi in both the plants and their co-evolved herbivores. While some herbivores, such as \u003cem\u003eM\u003c/em\u003e. \u003cem\u003esexta\u003c/em\u003e can rely on GLVs for feeding stimulation (Halitschke et al., 2004) or oviposition cues (Allmann et al., 2013), Brassicales specialists often depend primarily on glucosinolates for both feeding stimulation and oviposition decisions (Hopkins et al., 2009). Similarly, although certain natural enemies use shifts in the \u003cem\u003eZ\u003c/em\u003e-3/\u003cem\u003eE\u003c/em\u003e-2-GLVs ratio for prey detection (Allmann and Baldwin, 2010; Arimura et al., 2009; Takai et al., 2018), parasitoid wasps that target Brassicales-feeding caterpillars (\u003cem\u003eCotesia rubecula\u003c/em\u003e and \u003cem\u003eHyposoter ebeninus\u003c/em\u003e) instead rely, although not exclusively, on glucosinolate-derived volatiles (e.g. isothiocyanates and nitriles) as their host location cues (Dicke et al., 2009; Kos et al., 2012; Mumm et al., 2008). This may have consequently attenuated the selective pressures driving the maintenance of Hi function in both Brassicales and their specialist herbivores.\u003c/p\u003e\n\u003cp\u003ePhylogenetic analysis (Fig. 5a) and chronogram (Fig. 5d) indicates that plant Hi first arose in the stem lineage of mesangiosperms (192.2\u0026ndash;166.4 Ma) (Yang et al., 2020), with no orthologues detectable in basal angiosperms (\u003cem\u003eA\u003c/em\u003e.\u003cem\u003e\u0026nbsp;trichopoda\u0026nbsp;\u003c/em\u003eand \u003cem\u003eN\u003c/em\u003e.\u003cem\u003e\u0026nbsp;colorata\u003c/em\u003e) (Fig. 5a). This finding implies that the evolution of plant Hi significantly postdates the origin of the GLV pathway, as the capability for GLV biosynthesis and the associated key enzyme, hydroperoxide lyase (HPL), were already established in early vascular plants, including lycophytes and monilophytes (Tanaka et al., 2021). Lepidopteran Hi independently evolved approximately 60 million years later during the early-Cretaceous radiation of Apoditrysia (132.1\u0026ndash;105.6 Ma) (Kawahara et al., 2019), coinciding with\u0026mdash;and likely facilitated by\u0026mdash;the diversification of flowering plants during this period. Notably, several adaptive traits in Apoditrysia, such as a versatile proboscis and expanded detoxification gene families, reflect evolutionary adjustments to novel ecological niches created by angiosperm diversification in the Cretaceous (Breeschoten et al., 2021; Krenn and Kristensen, 2004). This temporal pattern, in which plant metabolic innovations precede and potentially drive insect biochemical adaptations, reflects a broader pattern seen in other evolutionary arms races. A representative example is the glucosinolate detoxification through nitrile-specifier proteins (NSPs): Kelch-type NSPs emerged in Brassicales after a whole-genome duplication event (~85 Ma; At-\u0026beta; event), shortly before insect-specific NSPs independently evolved in Pierinae butterflies (~68 Ma) to redirect host plants\u0026rsquo; glucosinolates toward less-toxic nitriles (Edger et al., 2015; Kissen and Bones, 2009; Walden et al., 2020; Wheat et al., 2007). Collectively, the independent emergence of Hi in both mesangiosperms and Lepidoptera represents another example of this evolutionary scenario, potentially enhancing the ecological specificity and impact of GLV-mediated interactions across trophic levels.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003e\u003cstrong\u003eInsects and plants\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRearing conditions of tobacco hornworm (\u003cem\u003eManduca sexta\u003c/em\u003e), tobacco budworm (\u003cem\u003eChloridea virescens\u003c/em\u003e), cabbage white (\u003cem\u003ePieris rapae\u003c/em\u003e), and monarch butterfly (\u003cem\u003eDanaus plexippus\u003c/em\u003e) are described in previous articles (Lievers et al., 2020; Lin et al., 2023; Lin et al., 2020). For dissecting different tissues, fourth to fifth instar larvae were dissected in 1x PBS buffer (pH 7.4) to extract midgut, fat body and salivary glands. Tissues from three different individuals were pooled for each biological replicate. All samples were flash-frozen in liquid nitrogen and stored at -80\u0026deg;C. Tomato Micro-Tom (\u003cem\u003eSolanum lycopersicum\u003c/em\u003e), white mulberry (\u003cem\u003eMorus alba\u003c/em\u003e) and milkweed (\u003cem\u003eAsclepias incarnate\u003c/em\u003e) were grown in the greenhouse with a day/night cycle of 16 h (26\u0026deg;C-28\u0026deg;C)/8 h (22\u0026deg;C-24\u0026deg;C) under supplemental light from Master Sun-T PIA Agro 400 or Master Sun-T PIA Plus 600-W sodium lights (Philips). \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRNA extraction and cDNA synthesis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe collected tissues were ground in liquid nitrogen with sterile pestles. Total RNA was extracted using the TRIzol/chloroform method according to the manufacturer\u0026rsquo;s protocol. The purified RNA was treated with DNase using the Ambion Turbo DNase kit (Thermo Fisher Scientific) to remove genomic DNA. Total RNA concentration was measured by NanoDrop ND-1000 (Thermo Fisher Scientific). One microgram of total RNA was used for cDNA synthesis with RevertAid First Strand cDNA synthesis kit (Thermo Fisher Scientific). Quantitative real-time PCR (ABI 7500 Real-Time PCR System; Applied Biosystems) was performed using the HOT FIREPol\u0026reg; EvaGreen\u0026reg; qPCR Mix Plus (Solis BioDyne).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGene cloning and recombinant protein production\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe coding regions of putative Hi homologs from \u003cem\u003eC. virescens\u003c/em\u003e, \u003cem\u003eM. sexta\u003c/em\u003e, and \u003cem\u003eD. plexippus\u003c/em\u003e were PCR amplified from a cDNA mixture derived from salivary glands and midgut tissues. The primers used for amplification are listed in Supplementary Table 3. The coding sequences of \u003cem\u003eB. mori\u003c/em\u003e were obtained through \u003cem\u003ede novo\u003c/em\u003e synthesis (Gene Universal). All coding sequences were cloned into the pGEX-4T-1 vector for GST-fusion protein expression in \u003cem\u003eE. coli\u003c/em\u003e. Plasmids were transformed into competent \u003cem\u003eE. coli\u003c/em\u003e BL21 (DE3) for recombinant protein expression. A single colony of transformed \u003cem\u003eE. coli\u003c/em\u003e was cultured in 10 mL LB medium, shaking overnight at 37 \u0026deg;C. The overnight \u003cem\u003eE. coli\u003c/em\u003e culture was then transferred to 1 L of 2\u0026times;YT medium (16 g tryptone, 10 g yeast extract, 5 g NaCl) and kept shaking at 37 \u0026deg;C until the optical density at 600 nm reached 0.4-0.5. IPTG (1 mM final concentration) was added to induce recombinant protein expression, and the culture was incubated with shaking at 16\u0026deg;C for 24-48 hours. For MsHi-2, induction was performed with 0.2mM IPTG at 10\u0026deg;C for 120 hours. The \u003cem\u003eE. coli\u003c/em\u003e pellet was collected by centrifugation at 15000 g. After discarding the supernatant, the pellet was snap-frozen in liquid nitrogen and stored at -20\u0026deg;C until use. For purification of recombinant proteins, the \u003cem\u003eE. coli\u003c/em\u003e pellet was first resuspended in 30 mL of lysis buffer containing 1\u0026times; PBS (pH 7.3), 1 mM EDTA, 10 mg/mL lysozyme, and proteinase inhibitor cocktails (50 mL per tablet). The suspended pellet was sonicated on ice, followed by the addition of 1% Triton X-100 and rotated for 30 minutes. The \u003cem\u003eE. coli\u003c/em\u003e lysate was collected by centrifugation at 15000 g and passed through a 0.45 \u0026micro;m filter. The lysate was batch purified using GST Sepharose 4B (GE Healthcare) according to the manufacturer\u0026apos;s instructions, and the purified proteins were preserved in 50 mM Tris-HCl buffer (pH 8.0) with 10% glycerol at -80\u0026deg;C until use. Quantitative densitometry of proteins from SDS-PAGE stained with Coomassie blue was used to determine protein concentrations by comparing the relative intensity of bands between recombinant proteins and a BSA standard (Bio-Rad Image Lab).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWestern blot analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRecombinant proteins were mixed with 4\u0026times; loading buffer and boiled at 95\u0026deg;C for 3 minutes. The proteins were separated on a 10% SDS-PAGE gel and transferred to an Immobilon-E PVDF membrane (Millipore) by semi-dry blotting. The membrane was washed three times with 1\u0026times; PBST (0.05% Tween 20) for 15 minutes each, then blocked with 5% BSA at room temperature for one hour. Subsequently, the membrane was incubated with GST-HRP conjugated antibody (1:2000, Santa Cruz Biotechnology) on a rotator overnight at 4\u0026deg;C. After three additional washes with 1\u0026times; PBST for 15 minutes each, the membrane was treated with 1 mL of chemiluminescence solution (100 mM Tris-HCl pH8.5, 9 mL of H\u003csub\u003e2\u003c/sub\u003eO, 1 mL of 1M Tris-HCl pH 8.5, 22 \u0026mu;L of 90 mM p-coumaric acid, 50 \u0026mu;L of 200 mM luminol, and 3 \u0026mu;L of 30% H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e). Images were captured using the Odyssey\u0026reg; Fc Imaging System (LI-COR) and analyzed with Image Studio Lite, or the ChemiDoc MP Imaging System and analyzed with Bio-Rad Image Lab.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHomolog identification and phylogenetic analysis in Lepidoptera\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThirty-four lepidopteran species representing a broad range of superfamilies were selected to capture the phylogenetic diversity of the order. In addition, three species from \u003cem\u003eTrichoptera\u003c/em\u003e (caddisflies), the sister group of \u003cem\u003eLepidoptera\u003c/em\u003e, and \u003cem\u003eDrosophila melanogaster\u003c/em\u003e (Diptera) were included to assist in the classification of GMC oxidoreductase subfamilies (Iida et al., 2007; Sun et al., 2012). The proteome sequences of studied organisms were obtained from the NCBI (https://www.ncbi.nlm.nih.gov/) and Ensembl (https://beta.ensembl.org/) databases (Supplementary Table 1a). Genome assemblies were prioritized based on completeness and the availability of annotated protein-coding genes. For lineages lacking annotated genomes, specifically Hepialoidea, Tischerioidea, and Palaephatoidea, transcriptome shotgun assemblies (TSAs) were used as alternative sources of protein coding sequences. From these transcriptome assemblies, open reading frames (ORFs) were predicted using the getorf program in the EMBOSS suite (version 6.5.7) (Rice et al., 2000). \u003c/p\u003e\n\u003cp\u003eTo identify GMC oxidoreductase homologs, we employed a domain-based search strategy using profile hidden Markov models (HMMs). HMM profiles for the N-terminal (PF00732) and C-terminal (PF05199) domains of the GMC oxidoreductase family were downloaded from the Pfam database hosted by InterPro (https://www.ebi.ac.uk/interpro/). These profiles were combined into a single database and indexed using hmmpress from the HMMER suite (v3.3.2) (Eddy, 1998). These validated sequences were then extracted from the original protein FASTA files, yielding a refined set of full-length GMC homologs. To reduce redundancy from alternative splicing, isoforms were collapsed by retaining only the longest protein variant per gene. Three fungal GMC sequences (AAF59929.2, XP_001727544.1, and AAD01493.1) were incorporated as outgroup taxa. The final dataset comprised 1,251 curated GMC domain sequences used for phylogenetic analysis. Species names, taxonomic classifications, and sequence sources are detailed in Supplementary Table 1b.\u003c/p\u003e\n\u003cp\u003eFor phylogenetic inference, protein sequences were aligned using MAFFT (v7.520) (Katoh and Standley, 2013) with the E-INS-i algorithm. The resulting multiple sequence alignment was trimmed with ClipKIT (v2.0.1) (Steenwyk et al., 2020). The best-fitting amino acid substitution model was selected using ModelTest-NG (v0.1.6) (Darriba et al., 2019). Maximum likelihood phylogenetic analysis was conducted with IQ-TREE2 (v2.3.4) (Minh et al., 2020). Branch support was assessed with 3,000 ultrafast bootstrap replicates and 3,000 SH-aLRT tests. The final phylogenetic tree was rooted with the predefined fungal outgroup sequences.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHomolog identification and phylogenetic analysis in the green lineage (Viridiplantae)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe complete predicted proteome sequences (Supplementary Table 2a) were obtained from the NCBI GeneBank (https://ncbi.nlm.nih.gov), UniProt-Proteomes database (https://www.uniprot.org), and JGI (http://genome.jgi.doe.gov). The 1000 Plants project (OneKP) database (http://www.onekp.com) was an additional source for predicted proteome sequences inferred from transcriptomic data. All cupin-domain (PF07883) containing proteins were identified on the previously retrieved predicted proteome sequences using the Hidden Markov model (HMM)-based tool hmmsearch (Johnson et al., 2010). All identified sequences (3734 sequences) were combined with 31 representative sequences from Cupin protein subfamilies (including Germin, Vicilin, Legumin, Globulin, and Hi) (Supplementary Table 2b) into one single dataset and analyzed using a phylogenetic approach (Supplementary Fig. 6). Only sequences that clustered within the Hi-potential clade in the phylogeny were identified as potential Hi orthologs (Supplementary Table 2c). SMART and Pfam databases were employed to identify conserved domains present in potential Hi orthologs (El-Gebali et al., 2018; Letunic and Bork, 2017). The results from both databases were merged, redundant domains were filtered-out and domain architecture was analyzed using the HMM based tool hmmscan (Finn et al., 2011).\u003c/p\u003e\n\u003cp\u003eAll identified potential Hi orthologs and representative sequences from cupin protein subfamilies were aligned using MAFFT, and ambiguously aligned regions were removed with trimAl (Capella-Guti\u0026eacute;rrez et al., 2009). The resulting alignments were evaluated with ProtTest v3 (Darriba et al., 2011) to determine the best-fit model for amino acid substitution. Two separate Maximum likelihood (ML) phylogenetic analyses were computed using RAxML-NG (Kozlov et al., 2019) and IQ-TREE 2 (Minh et al., 2020) (Fig. 5a), each with 1000 bootstrap replicates. Bootstrap support values from both analyses were mapped onto the IQ-TREE 2 ML tree, which was rooted at its midpoint (Fig. 5a).\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003e(3\u003cem\u003eZ\u003c/em\u003e):(2\u003cem\u003eE\u003c/em\u003e)-hexenal isomerase (Hi) activity assays by SPME-GC-qToF-MS \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA 200 \u0026micro;L solution containing recombinant proteins in a 20 mM Tris-HCl buffer (pH 8.5) was first transferred to a 1.5 mL GC vial equipped with a 200 \u0026micro;L insert. The reaction was initiated by adding Z-3-hexenal (0.2mM final concentration) to the solution. The GC vial was gently vortexed for 1 min. Subsequently, 100 \u0026micro;L mixture was transferred to a 20 mL glass headspace vial (SureSTART, Thermo Scientific) and immediately closed with a cap crimper. Headspace volatiles were collected using the Solid Phase Micro Extraction (SPME) fiber (Carboxen/Polydimethylsiloxane coated) for 10 min at 35 \u0026deg;C and analyzed by GC-qToF-MS. After sampling, the fiber was desorbed for 1 min in the injection port which was constantly kept at 250 \u0026deg;C. Compounds were separated on a HP-5ms column (30 m x 250 \u0026micro;m, 0.25 \u0026micro;m film thickness; Agilent) in an Agilent 7890A gas chromatograph with a temperature program set to 40 \u0026deg;C for 5 min, increasing to 140 \u0026deg;C at a rate of 5 \u0026deg;C per min, followed by increasing temperature to 250 \u0026deg;C at a rate of 15 \u0026deg;C per min and an additional 5 min at 250 \u0026deg;C. Helium was used as the carrier gas with the transfer column flow set to 3 mL per minute and a flow rate of 1 mL per min thereafter. Mass spectra were generated by an Agilent 7200 accurate-mass quadrupole time-of-flight mass spectrometer, operating in electron ionization mode (70 eV) at 230 \u0026deg;C and collected with an acquisition rate of 5 scans per second. Volatiles were identified and quantified using standard volatiles listed in Supplementary Table 3. The conversion rate of \u003cem\u003eE\u003c/em\u003e-2-hexenal was calculated as the proportion of its intensity relative to the total aldehyde intensity (\u003cem\u003eZ\u003c/em\u003e-3-hexenal + \u003cem\u003eE\u003c/em\u003e-2-hexenal). This value was then corrected by subtracting the non-enzymatic conversion rate determined from a buffer-only control, yielding the relative \u003cem\u003eE\u003c/em\u003e-2-hexenal level (Fig. 2a). To account for differences in detector sensitivity, response factors for \u003cem\u003eZ\u003c/em\u003e-3-hexenal and \u003cem\u003eE\u003c/em\u003e-2-hexenal were calculated using measured intensities of 2 nmol of each standard compound. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEffect of additional FAD on Hi activity\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo compare the activity of different Hi enzymes with and without additional FAD, the following amounts of recombinant proteins were used: 3 \u0026micro;g BmHi-1, 1.38 \u0026micro;g DpHi-1, 1 \u0026micro;g CvHi-1, and 0.0625 \u0026micro;g MsHi-1. These protein concentrations were carefully selected to ensure that, without FAD, the conversion of \u003cem\u003eZ\u003c/em\u003e-3-hexenal to \u003cem\u003eE\u003c/em\u003e-2-hexenal would not reach completion, allowing any potential increase in activity upon FAD addition to be clearly observed. Each enzyme was tested in a 200 \u0026micro;L reaction mixture containing 20 \u0026micro;g BSA, with or without 1 mM FAD, in 20 mM Tris-HCl buffer (pH 7.0). This mixture was transferred to a 1.5 mL GC vial equipped with a 200 \u0026micro;L insert, after which 0.2 mM \u003cem\u003eZ\u003c/em\u003e-3-hexenal was added to initiate the reaction. The vial was gently vortexed for 1 min. Subsequently, 100 \u0026micro;L of the mixture was transferred to a 20 mL glass headspace vial and sealed immediately using a cap crimper. Headspace volatiles were collected using the SPME fiber for 10 min at 35 \u0026deg;C and analyzed by GC-qToF-MS.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEnzyme kinetics\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor the determination of the kinetic parameters of the MsHi wildtype and H521A mutant, a substrate concentration range of 5 \u0026ndash; 4000 \u0026micro;M \u003cem\u003eZ\u003c/em\u003e-3-hexenal was used for the MsHi wildtype and 125 \u0026ndash; 4000 \u0026micro;M \u003cem\u003eZ\u003c/em\u003e-3-hexenal for the mutant. Recombinant proteins were diluted with 20 mM Tris-HCl (pH 8.5) buffer to a final volume of 200 \u0026micro;L. 37.5 ng of MsHi wildtype and 112.5 ng of H521A mutant were used for measurement. The mixture was incubated at room temperature for 2 min, subsequently transferred to 20 mL glass headspace vial and analyzed by GC-qToF-MS. The Km, Kcat, and enzyme efficiency (Kcat/Km) were calculated with nonlinear least-square regression using GraphPad Prism 10. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAnalysis of volatiles in planta\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA 24\u0026thinsp;mm diameter of leaf disc was punched out from the lamina of the second or third pair of true leaves from the top of four-week-old plants. The leaf disc was mechanically wounded on the adaxial surface by rolling a fabric pattern wheel to produce two parallel rows of punctures on either side of the midvein. 10\u0026thinsp;\u0026micro;L of recombinant protein, heat-inactivated recombinant protein (boiled at 95\u0026thinsp;\u0026deg;C for 3 minutes), or water was applied to the wounds and gently dispersed across the leaf surface. After 20 seconds, the leaf disc was transferred to a 20 mL glass headspace vial, closed with a crimp cap and volatiles were immediately collected with a SPME fiber for 10\u0026thinsp;minutes at 35\u0026thinsp;\u0026deg;C and measured by GC-qToF-MS analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStatistical analyses were performed using GraphPad Prism 10. The Shapiro\u0026ndash;Wilk test was used to assess normality of data distributions. For comparisons among multiple groups, one-way ANOVA followed by Tukey\u0026rsquo;s multiple comparisons test was used. For non-parametric data, the Kruskal\u0026ndash;Wallis test followed by Dunn\u0026rsquo;s multiple comparisons test was applied. Different letters above bars in the graphs indicate statistically significant differences between groups (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Two-tailed t-tests were used for comparisons between two groups. Error bars represent mean values\u0026thinsp;\u0026plusmn;\u0026thinsp;SD. *\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, **\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ***\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, n.s., not significant.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Prof. Erik Poelman (Wageningen University) who kindly provided us with \u003cem\u003ePieris rapae\u003c/em\u003e. We appreciate the help of our colleagues at University of Amsterdam, Ludek Tikovsky and Harold Lemereis for taking care of all plants in the glasshouse, Dr. Jocelyne Vreede and Dr. Sandra Eltschkner for the discussion of protein structure analysis, Prof. Astrid Groot provided us with \u003cem\u003eChloridea virescens\u003c/em\u003e. We thank Dr. Heiko Vogel (Max Planck Institute), Dr. David Doležel (Biology Centre CAS) and Dr. Sm\u0026yacute;kal Vlastimil (Biology Centre CAS) for valuable discussion on this work. Computational resources for homolog identification and phylogenetic analysis in Lepidoptera were provided by the e-INFRA CZ project (ID: 90254), supported by the Ministry of Education, Youth and Sports of the Czech Republic, and the ELIXIR-CZ project (ID: 90255), part of the international ELIXIR infrastructure. A.S. was supported by the SequAna Sequencing Analysis Core Facility at the Department of Biology, University of Konstanz. C.W.T, R.S., and J.A. were supported by U.S. National Science Foundation (NSF‐IOS 1754996). This work was supported by the European Research Council (ERC) under the European Union\u0026rsquo;s Horizon 2020 research and innovation programme (grant agreement no. 805074) (S. A. and Y. H. L.), \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eContributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eY.H.L. conceived and designed the study, performed experiments, analyzed data, and drafted the manuscript. B.C.H.W. and A.S. analyzed the phylogeny. S.M.E.H. performed cloning and qPCR for \u003cem\u003eD\u003c/em\u003e. \u003cem\u003eplexippus\u003c/em\u003e, \u003cem\u003eC\u003c/em\u003e. \u003cem\u003evirescens\u003c/em\u003e and \u003cem\u003eM\u003c/em\u003e. \u003cem\u003esexta\u003c/em\u003e Hi. I.P. performed GC/MS analysis and kinetic measurements of mutant Hi. C.W.T., J.A. and R.S. conducted qPCR analysis for \u003cem\u003eD\u003c/em\u003e. \u003cem\u003eplexippus\u003c/em\u003e Hi. S.A. conceived the study, supervised the project, acquired funding, and drafted the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\n"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAbramson, J., Adler, J., Dunger, J., Evans, R., Green, T., Pritzel, A., Ronneberger, O., Willmore, L., Ballard, A.J., Bambrick, J., Bodenstein, S.W., Evans, D.A., Hung, C.-C., O\u0026rsquo;Neill, M., Reiman, D., Tunyasuvunakool, K., Wu, Z., Žemgulytė, A., Arvaniti, E., Beattie, C., Bertolli, O., Bridgland, A., Cherepanov, A., Congreve, M., Cowen-Rivers, A.I., Cowie, A., Figurnov, M., Fuchs, F.B., Gladman, H., Jain, R., Khan, Y.A., Low, C.M.R., Perlin, K., Potapenko, A., Savy, P., Singh, S., Stecula, A., Thillaisundaram, A., Tong, C., Yakneen, S., Zhong, E.D., Zielinski, M., Ž\u0026iacute;dek, A., Bapst, V., Kohli, P., Jaderberg, M., Hassabis, D., Jumper, J.M., 2024. Accurate structure prediction of biomolecular interactions with AlphaFold 3. Nature 630, 493-500.\u003c/li\u003e\n\u003cli\u003eAllmann, S., Baldwin, I.T., 2010. Insects betray themselves in nature to predators by rapid isomerization of green leaf volatiles. Science 329, 1075-1078.\u003c/li\u003e\n\u003cli\u003eAllmann, S., Spathe, A., Bisch-Knaden, S., Kallenbach, M., Reinecke, A., Sachse, S., Baldwin, I.T., Hansson, B.S., 2013. Feeding-induced rearrangement of green leaf volatiles reduces moth oviposition. Elife 2, e00421.\u003c/li\u003e\n\u003cli\u003eAlm\u0026eacute;ras, E., Stolz, S., Vollenweider, S., Reymond, P., M\u0026egrave;ne-Saffran\u0026eacute;, L., Farmer, E.E., 2003. Reactive electrophile species activate defense gene expression in Arabidopsis. The Plant Journal 34, 205-216.\u003c/li\u003e\n\u003cli\u003eAmeye, M., Allmann, S., Verwaeren, J., Smagghe, G., Haesaert, G., Schuurink, R.C., Audenaert, K., 2018. Green leaf volatile production by plants: a meta-analysis. New Phytologist 220, 666-683.\u003c/li\u003e\n\u003cli\u003eArimura, G.-i., Matsui, K., Takabayashi, J., 2009. Chemical and Molecular Ecology of Herbivore-Induced Plant Volatiles: Proximate Factors and Their Ultimate Functions. Plant and Cell Physiology 50, 911-923.\u003c/li\u003e\n\u003cli\u003eBerthelot, K., Estevez, Y., Deffieux, A., Peruch, F., 2012. Isopentenyl diphosphate isomerase: A checkpoint to isoprenoid biosynthesis. Biochimie 94, 1621-1634.\u003c/li\u003e\n\u003cli\u003eBreeschoten, T., van der Linden, C.F.H., Ros, V.I.D., Schranz, M.E., Simon, S., 2021. Expanding the Menu: Are Polyphagy and Gene Family Expansions Linked across Lepidoptera? Genome Biology and Evolution 14.\u003c/li\u003e\n\u003cli\u003eCapella-Guti\u0026eacute;rrez, S., Silla-Mart\u0026iacute;nez, J.M., Gabald\u0026oacute;n, T., 2009. trimAl: a tool for automated alignment trimming in large-scale phylogenetic analyses. Bioinformatics 25, 1972-1973.\u003c/li\u003e\n\u003cli\u003eCavener, D.R., 1992. GMC oxidoreductases: A newly defined family of homologous proteins with diverse catalytic activities. Journal of Molecular Biology 223, 811-814.\u003c/li\u003e\n\u003cli\u003eChen, C., Yu, F., Wen, X., Chen, S., Wang, K., Wang, F., Zhang, J., Wu, Y., He, P., Tu, Y., Li, B., 2022. Characterization of a new (Z)-3:(E)-2-hexenal isomerase from tea (Camellia sinensis) involved in the conversion of (Z)-3-hexenal to (E)-2-hexenal. Food Chemistry 383, 132463.\u003c/li\u003e\n\u003cli\u003eCroft, K., Juttner, F., Slusarenko, A.J., 1993. Volatile Products of the Lipoxygenase Pathway Evolved from Phaseolus vulgaris (L.) Leaves Inoculated with Pseudomonas syringae pv phaseolicola. Plant Physiol 101, 13-24.\u003c/li\u003e\n\u003cli\u003eDarriba, D., Posada, D., Kozlov, A.M., Stamatakis, A., Morel, B., Flouri, T., 2019. ModelTest-NG: A New and Scalable Tool for the Selection of DNA and Protein Evolutionary Models. Molecular Biology and Evolution 37, 291-294.\u003c/li\u003e\n\u003cli\u003eDarriba, D., Taboada, G.L., Doallo, R., Posada, D., 2011. ProtTest 3: fast selection of best-fit models of protein evolution. Bioinformatics 27, 1164-1165.\u003c/li\u003e\n\u003cli\u003eDicke, M., van Loon, J.J.A., Soler, R., 2009. Chemical complexity of volatiles from plants induced by multiple attack. Nature Chemical Biology 5, 317-324.\u003c/li\u003e\n\u003cli\u003eDudareva, N., Florence, N., A., N.D., and Orlova, I., 2006. Plant Volatiles: Recent Advances and Future Perspectives. Critical Reviews in Plant Sciences 25, 417-440.\u003c/li\u003e\n\u003cli\u003eDunwell, J.M., Culham, A., Carter, C.E., Sosa-Aguirre, C.R., Goodenough, P.W., 2001. Evolution of functional diversity in the cupin superfamily. Trends in Biochemical Sciences 26, 740-746.\u003c/li\u003e\n\u003cli\u003eEddy, S.R., 1998. Profile hidden Markov models. Bioinformatics 14, 755-763.\u003c/li\u003e\n\u003cli\u003eEdger, P.P., Heidel-Fischer, H.M., Bekaert, M., Rota, J., Gl\u0026ouml;ckner, G., Platts, A.E., Heckel, D.G., Der, J.P., Wafula, E.K., Tang, M., Hofberger, J.A., Smithson, A., Hall, J.C., Blanchette, M., Bureau, T.E., Wright, S.I., dePamphilis, C.W., Eric Schranz, M., Barker, M.S., Conant, G.C., Wahlberg, N., Vogel, H., Pires, J.C., Wheat, C.W., 2015. The butterfly plant arms-race escalated by gene and genome duplications. Proceedings of the National Academy of Sciences 112, 8362-8366.\u003c/li\u003e\n\u003cli\u003eEggers, R., Jammer, A., Jha, S., Kerschbaumer, B., Lahham, M., Strandback, E., Toplak, M., Wallner, S., Winkler, A., Macheroux, P., 2021. The scope of flavin-dependent reactions and processes in the model plant Arabidopsis thaliana. Phytochemistry 189.\u003c/li\u003e\n\u003cli\u003eEl-Gebali, S., Mistry, J., Bateman, A., Eddy, S.R., Luciani, A., Potter, S.C., Qureshi, M., Richardson, L.J., Salazar, G.A., Smart, A., Sonnhammer, E.L L., Hirsh, L., Paladin, L., Piovesan, D., Tosatto, S.C E., Finn, R.D., 2018. The Pfam protein families database in 2019. Nucleic Acids Research 47, D427-D432.\u003c/li\u003e\n\u003cli\u003eEngelberth, J., Engelberth, M., 2020. Variability in the Capacity to Produce Damage-Induced Aldehyde Green Leaf Volatiles among Different Plant Species Provides Novel Insights into Biosynthetic Diversity. Plants 9, 213.\u003c/li\u003e\n\u003cli\u003eFern\u0026aacute;ndez, I.S., Ru\u0026iacute;z-Due\u0026ntilde;as, F.J., Santillana, E., Ferreira, P., Mart\u0026iacute;nez, M.J., Mart\u0026iacute;nez, A.T., Romero, A., 2009. Novel structural features in the GMC family of oxidoreductases revealed by the crystal structure of fungal aryl-alcohol oxidase. Acta crystallographica. Section D, Biological crystallography 65, 1196-1205.\u003c/li\u003e\n\u003cli\u003eFinn, R.D., Clements, J., Eddy, S.R., 2011. HMMER web server: interactive sequence similarity searching. Nucleic Acids Research 39, W29-W37.\u003c/li\u003e\n\u003cli\u003eHalitschke, R., Ziegler, J., Kein\u0026auml;nen, M., Baldwin, I.T., 2004. Silencing of hydroperoxide lyase and allene oxide synthase reveals substrate and defense signaling crosstalk in Nicotiana attenuata. The Plant Journal 40, 35-46.\u003c/li\u003e\n\u003cli\u003eHatanaka, A., Kajiwara, T., Sekiya, J., 1987. Biosynthetic pathway for C6-aldehydes formation from linolenic acid in green leaves. Chemistry and Physics of Lipids 44, 341-361.\u003c/li\u003e\n\u003cli\u003eHopkins, R.J., van Dam, N.M., van Loon, J.J.A., 2009. Role of Glucosinolates in Insect-Plant Relationships and Multitrophic Interactions. Annual Review of Entomology 54, 57-83.\u003c/li\u003e\n\u003cli\u003eHyun, J., Lee, J.G., Yang, K.-Y., Lim, S., Lee, E.J., 2022. Postharvest Fumigation of (E)-2-Hexenal on Kiwifruit (Actinidia chinensis cv. \u0026lsquo;Haegeum\u0026rsquo;) Enhances Resistance to Botrytis cinerea. Postharvest Biology and Technology 187, 111854.\u003c/li\u003e\n\u003cli\u003eIida, K., Cox-Foster, D.L., Yang, X., Ko, W.-Y., Cavener, D.R., 2007. Expansion and evolution of insect GMC oxidoreductases. BMC Evolutionary Biology 7, 75.\u003c/li\u003e\n\u003cli\u003eJohnson, L.S., Eddy, S.R., Portugaly, E., 2010. Hidden Markov model speed heuristic and iterative HMM search procedure. BMC Bioinformatics 11, 431.\u003c/li\u003e\n\u003cli\u003eJones, A.C., Cofer, T.M., Engelberth, J., Tumlinson, J.H., 2021. Herbivorous Caterpillars and the Green Leaf Volatile (GLV) Quandary. J Chem Ecol.\u003c/li\u003e\n\u003cli\u003eKatoh, K., Standley, D.M., 2013. MAFFT Multiple Sequence Alignment Software Version 7: Improvements in Performance and Usability. Molecular Biology and Evolution 30, 772-780.\u003c/li\u003e\n\u003cli\u003eKawahara, A.Y., Plotkin, D., Espeland, M., Meusemann, K., Toussaint, E.F.A., Donath, A., Gimnich, F., Frandsen, P.B., Zwick, A., Dos Reis, M., Barber, J.R., Peters, R.S., Liu, S., Zhou, X., Mayer, C., Podsiadlowski, L., Storer, C., Yack, J.E., Misof, B., Breinholt, J.W., 2019. Phylogenomics reveals the evolutionary timing and pattern of butterflies and moths. Proc Natl Acad Sci U S A 116, 22657-22663.\u003c/li\u003e\n\u003cli\u003eKissen, R., Bones, A.M., 2009. Nitrile-specifier Proteins Involved in Glucosinolate Hydrolysis in Arabidopsis thaliana*. Journal of Biological Chemistry 284, 12057-12070.\u003c/li\u003e\n\u003cli\u003eKos, M., Houshyani, B., Wietsma, R., Kabouw, P., Vet, L.E.M., van Loon, J.J.A., Dicke, M., 2012. Effects of glucosinolates on a generalist and specialist leaf-chewing herbivore and an associated parasitoid. Phytochemistry 77, 162-170.\u003c/li\u003e\n\u003cli\u003eKozlov, A.M., Darriba, D., Flouri, T., Morel, B., Stamatakis, A., 2019. RAxML-NG: a fast, scalable and user-friendly tool for maximum likelihood phylogenetic inference. Bioinformatics 35, 4453-4455.\u003c/li\u003e\n\u003cli\u003eKrenn, H.W., Kristensen, N.P., 2004. Evolution of proboscis musculature in Lepidoptera. Eur J Entomol 101, 565-575.\u003c/li\u003e\n\u003cli\u003eKunishima, M., Yamauchi, Y., Mizutani, M., Kuse, M., Takikawa, H., Sugimoto, Y., 2016. Identification of (Z)-3:(E)-2-Hexenal Isomerases Essential to the Production of the Leaf Aldehyde in Plants. J Biol Chem 291, 14023-14033.\u003c/li\u003e\n\u003cli\u003eLetunic, I., Bork, P., 2017. 20 years of the SMART protein domain annotation resource. Nucleic Acids Research 46, D493-D496.\u003c/li\u003e\n\u003cli\u003eLiavonchanka, A., Feussner, I., 2008. Biochemistry of PUFA Double Bond Isomerases Producing Conjugated Linoleic Acid. ChemBioChem 9, 1867-1872.\u003c/li\u003e\n\u003cli\u003eLiavonchanka, A., Rudolph, M.G., Tittmann, K., Hamberg, M., Feussner, I., 2009. On the Mechanism of a Polyunsaturated Fatty Acid Double Bond Isomerase from. Journal of Biological Chemistry 284, 8005-8012.\u003c/li\u003e\n\u003cli\u003eLievers, R., Kuperus, P., Groot, A.T., 2020. DNA methylation patterns in the tobacco budworm, Chloridea virescens. Insect Biochemistry and Molecular Biology 121, 103370.\u003c/li\u003e\n\u003cli\u003eLin, Y.-H., Silven, J.J.M., Wybouw, N., Fandino, R.A., Dekker, H.L., Vogel, H., Wu, Y.-L., de Koster, C., Gro\u0026szlig;e-Wilde, E., Haring, M.A., Schuurink, R.C., Allmann, S., 2023. A salivary GMC oxidoreductase of Manduca sexta re-arranges the green leaf volatile profile of its host plant. Nature Communications 14, 3666.\u003c/li\u003e\n\u003cli\u003eLin, Y.H., Tai, C.C., Broz, V., Tang, C.K., Chen, P., Wu, C.P., Li, C.H., Wu, Y.L., 2020. Adenosine Receptor Modulates Permissiveness of Baculovirus (Budded Virus) Infection via Regulation of Energy Metabolism in Bombyx mori. Front Immunol 11, 763.\u003c/li\u003e\n\u003cli\u003eLiu, X., Ohta, T., Kawabata, T., Kawai, F., 2013. Catalytic Mechanism of Short Ethoxy Chain Nonylphenol Dehydrogenase Belonging to a Polyethylene Glycol Dehydrogenase Group in the GMC Oxidoreductase Family. International Journal of Molecular Sciences 14, 1218-1231.\u003c/li\u003e\n\u003cli\u003eMatsui, K., 2006. Green leaf volatiles: hydroperoxide lyase pathway of oxylipin metabolism. Current Opinion in Plant Biology 9, 274-280.\u003c/li\u003e\n\u003cli\u003eMatsui, K., Engelberth, J., 2022. Green Leaf Volatiles\u0026mdash;The Forefront of Plant Responses Against Biotic Attack. Plant and Cell Physiology 63, 1378-1390.\u003c/li\u003e\n\u003cli\u003eMinh, B.Q., Schmidt, H.A., Chernomor, O., Schrempf, D., Woodhams, M.D., von Haeseler, A., Lanfear, R., 2020. IQ-TREE 2: New Models and Efficient Methods for Phylogenetic Inference in the Genomic Era. Molecular Biology and Evolution 37, 1530-1534.\u003c/li\u003e\n\u003cli\u003eMumm, R., Burow, M., Bukovinszkine\u0026rsquo;Kiss, G., Kazantzidou, E., Wittstock, U., Dicke, M., Gershenzon, J., 2008. Formation of Simple Nitriles upon Glucosinolate Hydrolysis Affects Direct and Indirect Defense Against the Specialist Herbivore, Pieris rapae. Journal of Chemical Ecology 34, 1311-1321.\u003c/li\u003e\n\u003cli\u003eOhta, T., Kawabata, T., Nishikawa, K., Tani, A., Kimbara, K., Kawai, F., 2006. Analysis of Amino Acid Residues Involved in Catalysis of Polyethylene Glycol Dehydrogenase from \u0026lt;i\u0026gt;Sphingopyxis terrae\u0026lt;/i\u0026gt;, Using Three-Dimensional Molecular Modeling-Based Kinetic Characterization of Mutants. Applied and Environmental Microbiology 72, 4388-4396.\u003c/li\u003e\n\u003cli\u003ePaudel Timilsena, B., Seidl-Adams, I., Tumlinson, J.H., 2020. Herbivore-specific plant volatiles prime neighboring plants for nonspecific defense responses. Plant Cell Environ 43, 787-800.\u003c/li\u003e\n\u003cli\u003ePeris, D., Condamine, F.L., 2024. The angiosperm radiation played a dual role in the diversification of insects and insect pollinators. Nature Communications 15, 552.\u003c/li\u003e\n\u003cli\u003ePhillips, D.R., Matthew, J.A., Reynolds, J., Fenwick, G.R., 1979. Partial purification and properties of a cis-3: trans-2-enal isomerase from cucumber fruit. Phytochemistry 18, 401-404.\u003c/li\u003e\n\u003cli\u003eReardon, J.E., Abeles, R.H., 1986. Mechanism of action of isopentenyl pyrophosphate isomerase: evidence for a carbonium ion intermediate. Biochemistry 25, 5609-5616.\u003c/li\u003e\n\u003cli\u003eRice, P., Longden, I., Bleasby, A., 2000. EMBOSS: The European Molecular Biology Open Software Suite. Trends in Genetics 16, 276-277.\u003c/li\u003e\n\u003cli\u003eSpyropoulou, E.A., Dekker, H.L., Steemers, L., van Maarseveen, J.H., de Koster, C.G., Haring, M.A., Schuurink, R.C., Allmann, S., 2017. Identification and Characterization of (3Z):(2E)-Hexenal Isomerases from Cucumber. Front Plant Sci 8, 1342.\u003c/li\u003e\n\u003cli\u003eSteenwyk, J.L., Buida, T.J., III, Li, Y., Shen, X.-X., Rokas, A., 2020. ClipKIT: A multiple sequence alignment trimming software for accurate phylogenomic inference. PLOS Biology 18, e3001007.\u003c/li\u003e\n\u003cli\u003eSun, W., Shen, Y.H., Yang, W.J., Cao, Y.F., Xiang, Z.H., Zhang, Z., 2012. Expansion of the silkworm GMC oxidoreductase genes is associated with immunity. Insect Biochem Mol Biol 42, 935-945.\u003c/li\u003e\n\u003cli\u003eTakai, H., Ozawa, R., Takabayashi, J., Fujii, S., Arai, K., Ichiki, R.T., Koeduka, T., Dohra, H., Ohnishi, T., Taketazu, S., Kobayashi, J., Kainoh, Y., Nakamura, S., Fujii, T., Ishikawa, Y., Kiuchi, T., Katsuma, S., Uefune, M., Shimada, T., Matsui, K., 2018. Silkworms suppress the release of green leaf volatiles by mulberry leaves with an enzyme from their spinnerets. Sci Rep 8, 11942.\u003c/li\u003e\n\u003cli\u003eWalden, N., German, D.A., Wolf, E.M., Kiefer, M., Rigault, P., Huang, X.-C., Kiefer, C., Schmickl, R., Franzke, A., Neuffer, B., Mummenhoff, K., Koch, M.A., 2020. Nested whole-genome duplications coincide with diversification and high morphological disparity in Brassicaceae. Nature Communications 11, 3795.\u003c/li\u003e\n\u003cli\u003eWheat, C.W., Vogel, H., Wittstock, U., Braby, M.F., Underwood, D., Mitchell-Olds, T., 2007. The genetic basis of a plant\u0026ndash;insect coevolutionary key innovation. Proceedings of the National Academy of Sciences 104, 20427-20431.\u003c/li\u003e\n\u003cli\u003eWongnate, T., Chaiyen, P., 2013. The substrate oxidation mechanism of pyranose 2-oxidase and other related enzymes in the glucose\u0026ndash;methanol\u0026ndash;choline superfamily. The FEBS Journal 280, 3009-3027.\u003c/li\u003e\n\u003cli\u003eYang, L.X., Su, D.Y., Chang, X., Foster, C.S.P., Sun, L.H., Huang, C.H., Zhou, X.F., Zeng, L.P., Ma, H., Zhong, B.J., 2020. Phylogenomic Insights into Deep Phylogeny of Angiosperms Based on Broad Nuclear Gene Sampling. Plant Commun 1.\u003c/li\u003e\n\u003cli\u003eYoshida, H., Sakai, G., Mori, K., Kojima, K., Kamitori, S., Sode, K., 2015. Structural analysis of fungus-derived FAD glucose dehydrogenase. Sci Rep-Uk 5.\u003c/li\u003e\n\u003cli\u003eYu, Q.J., Ghisla, S., Hirschberg, J., Mann, V., Beyer, P., 2011. Plant Carotene Cis-Trans Isomerase CRTISO. Journal of Biological Chemistry 286, 8666-8676.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7163309/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7163309/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Green leaf volatiles (GLVs) are six carbon volatile organic compounds that mediate plant responses to environmental stresses. The quantity and composition of emitted GLVs can vary with stress type, allowing plants to fine-tune their volatile blends. Additionally, insect herbivores are capable of modulating these emissions. A key mechanism underlying this plasticity is the conversion of Z-3-hexenal to E-2-hexenal by the enzyme (3Z):(2E)-hexenal isomerase (Hi), which reshapes GLV profiles and influences multi-trophic interactions. Here, we investigate the evolutionary origin, functional diversification, and catalytic mechanisms of Lepidopteran Hi homologs, which belong to the GMC oxidoreductase family. Phylogenetic analysis of 34 lepidopteran species identified a distinct GMCβ subclade enriched in Hi homologs, largely confined to the Apoditrysia lineage. Functional assays showed species-specific variation in Hi activity, with Manduca sexta Hi-1 displaying the highest activity under identical protein concentrations, both in vitro and in planta. Structural modeling and site-directed mutagenesis revealed that Hi activity requires a flavin adenine dinucleotide (FAD) cofactor enabling the identification of key residues critical for FAD binding. Comparative phylogenetics further suggests that Hi enzymes in plants and Lepidoptera evolved independently from unrelated enzyme families, representing a case of functional convergence during the Cretaceous angiosperm radiation.","manuscriptTitle":"FAD-dependent hexenal isomerases in Lepidoptera evolved convergently with plant-derived hexenal isomerases","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-01 11:19:50","doi":"10.21203/rs.3.rs-7163309/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-ecology-and-evolution","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"natecolevol","sideBox":"Learn more about [Nature Ecology \u0026 Evolution](http://www.nature.com/natecolevol/)","snPcode":"","submissionUrl":"","title":"Nature Ecology \u0026 Evolution","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"c5982266-6c06-47d7-bb45-13e9f6e5f784","owner":[],"postedDate":"August 1st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":51839676,"name":"Biological sciences/Evolution/Molecular evolution"},{"id":51839677,"name":"Biological sciences/Ecology/Ecophysiology"},{"id":51839678,"name":"Biological sciences/Plant sciences/Plant evolution"},{"id":51839679,"name":"Biological sciences/Zoology/Entomology"}],"tags":[],"updatedAt":"2026-02-28T08:05:14+00:00","versionOfRecord":{"articleIdentity":"rs-7163309","link":"https://doi.org/10.1038/s41559-026-02999-2","journal":{"identity":"nature-ecology-and-evolution","isVorOnly":false,"title":"Nature Ecology \u0026 Evolution"},"publishedOn":"2026-02-27 05:00:00","publishedOnDateReadable":"February 27th, 2026"},"versionCreatedAt":"2025-08-01 11:19:50","video":"","vorDoi":"10.1038/s41559-026-02999-2","vorDoiUrl":"https://doi.org/10.1038/s41559-026-02999-2","workflowStages":[]},"version":"v1","identity":"rs-7163309","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7163309","identity":"rs-7163309","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}