Red palm weevil (RPW, Rhynchophorus ferrugineus) is a quarantine pest of palms (Arecaceae). It can spread up to 50 km per day and currently 85 countries and regions have been infested by this insect pest. In China, it was recorded in 1997 in Zhongshan City, Guangdong Province, and has now spread to 15 provinces (Ge et al., 2015). Since its invasion in China, it has damaged >10,000 km² area and killed >20,000 coconut (Cocos nucifera) trees. However, the damaged area is rapidly expanding, thus it is not only a danger to the coconut industry but also poses a threat to the ecological security of the Chinese coastal areas. Particularly, the expansion of coconut and A. catechu in Hainan province has brought economic benefits to the farmers. Coconut is cultivated on > 40,000 ha in China, of which 99% is located in Hainan province (∼232 million coconuts) (Hainan coconut industry highly quality development: The 14th Five-Year Plan, http://lyj.hainan.gov.cn, accessed on 20.11.2022). However, such increased plantation has also contributed to the rapid spread of RPW (Peng and Hou, 2017). Additionally, the RPW has been included in the “People’s Republic of China Entry Phytosanitary Pest List” (Ministry of Agriculture of the People’s Republic of China, General Administration of Quality Supervision Inspection Quarantine of the People’s Republic of China, 2007) and “National Forestry Quarantine Pest List (Yanhua, 2013)”, and the comprehensive risk evaluation value is highly dangerous.

Due to its habit of dwelling inside the trunks, its early detection is difficult and its presence is confirmed only after the damage symptoms have become visible (Salama et al., 2009). The prevention and control strategies include understanding its population dynamics, chemical control, bait trapping, biological control, and breeding RPW-tolerant genotypes (Peng and Hou, 2017). However, the success of the prevention as well as control of RPW is directly associated with effective early detection (Faleiro, 2006). To detect RPW, computer-assisted tomography, evaluation of the symptoms of larval mines and chewed material at the leaf basis, and bioacoustics sensors have been tested in this regard. However, the feasibility and cost limit the adaptation of these strategies (Giovino et al., 2015). Nevertheless, host-plant resistance is an ideal long-term strategy. Different palm species have been reported to have varying levels of resistance to RPW. Particularly, the reports that show the presence of palm species and wild palms resistant to RPW in Spain and Iran are interesting since they can help us identify the molecular mechanisms of RPW resistance (Barranco et al., 2000; Farazmand, 2002). Additionally, the exploration of host-plant responses to RPW attack on a molecular scale i.e., metabolomic and transcriptomic responses, can help us identify key genes and pathways that can be manipulated for breeding RPW-resistant plants. Studies on Phoenix canariensis Chabaud (Giovino et al., 2015), have shown the involvement of genes in phenylpropanoid biosynthesis, fatty acid metabolism, tryptophan metabolism, and hormonal crosstalk. However, such information in coconut cultivars grown in Hainan, China is sparse, thus not allowing a basic understanding of the coconut plants’ response to RPW attack, especially during early infestation stages.

During the co-evolution of plants and herbivore insects, both have evolved certain mechanisms with giving them better survival chances. In the case of plants, when herbivore insects attack, the concentrations of Ca²⁺ change across the membranes (Parmagnani and Maffei, 2022), which initiate a cascade of signaling (Wu and Baldwin, 2010). These signals activate key pathways that produce chemicals to repel herbivores through direct toxicity or affecting the digestibility of plant tissues. Secondly, biosynthesis of several substances is induced as a result of plant tissues’ damage by herbivore insects. These compounds include plant growth regulators e.g., phytohormones, while others include those involved in transport, storage, and as well as primary metabolism related compounds. The signaling pathways involved in herbivory/wounding include plant-hormone signaling and related pathways e.g., MAPK signaling-plant pathway (Mello and Silva-Filho, 2002; Wu and Baldwin, 2010). Whereas, the insect response mechanisms involve increased biosynthesis of secondary metabolites which either directly deter insects (feeding inhibitors) such as aldehydes, lignins, tannins, terpenoids, phenolics, flavonoids, quinones, alkaloids, and glucosinolates (Mello and Silva-Filho, 2002). For such increased secondary metabolite biosynthesis, the interactions between photosynthesis, reactive oxygen species (ROS), and hormonal signaling lead to efficient resource management (Kerchev et al., 2012).

However, the knowledge of such pathways is limited in the case of RPW infected coconut plants, which is highly required to understand the key coconut response to RPW attack. Here, we have studied the leaf’s physiological and transcriptomic responses to fill this knowledge gap and provide a way forward. For this, we used the most widely cultivated coconut cultivar i.e., Hainan Tall. The information on the susceptibility or tolerance of Hainan Tall to RPW is not known. However, the earlier surveys and pilot studies by our team have indicated that if RPW infestation can be detected at an early stage, a viable RPW control strategy can be developed. In this regard, Hainan Tall plants were infected with different densities (number of insects (male and female)) on different days after infestation (DAI). First, based on leaf physiological responses, we identified the densities of the RPW that significantly affected the coconut plant. Then, we compared the transcriptomic changes in coconut leaf on 5 and 20 DAI for the selected RPW densities.

The increase in the area damaged by RPW is threatening the coconut industry in China (Ge et al., 2015). Continued efforts are needed for its effective control. In this regard, we attempted to understand changes in the activities of antioxidative enzymes and the global transcriptome profiles of the coconut leaves infected with RPW. Our results provide the basis for understanding the possible defense strategies in the host plant.

Firstly, our results that the RPW attack caused the reduced activity of CAT is interesting. These findings are consistent with those reported in Pinus sylvestris in response to wounding caused by sawfly (Diprion pini) oviposition (Bittner et al., 2017). The changes in expressions of CAT transcripts (COCNU_scaffold021643G000020 and OCNU_08G006310) in CK are consistent with these findings (Figure 4). These results suggest that the RPW attack on different DAI causes changes in ROS generation/degradation. Which is also evident from the changes in the activities of POD and SOD in addition to CAT (particularly) (Figure 2). The observations that we did not find a specific trend between the RPW population density and DAI are not novel and have been previously reported in two coconut cultivars (PANDAN and MATAG from Malaysia) (Hasim and Yusuf, 2017). The general trend is that POD activities were lower in CK as compared to A-D is consistent with reports that plants infested with chewing insects (e.g., Spodoptera) exhibit increased POD activities (Singh et al., 2013). Peroxidases are pathogenesis-related proteins and have been potentially implicated in plant-insect interaction (Kehr, 2006). The increase in POD activities, particularly, on 5 and 20 DAI in C and D as compared to CK, and the consistent expressions of POD transcripts are important results that indicate their role in defense against RPW. It is also a potential indicator of infestation level. Other than these, the higher MDA content on 10, 20, and 25 DAI in RPW-infested coconut leaves as compared to CK is indicative of peroxidation of lipids. The study by (Hasim and Yusuf, 2017) also indicated somewhat similar results. The polar lipids form lipid bilayer and hence act as permeability barrier of the cells. In case of higher ROS levels, lipid peroxidation increases, which affect the normal functioning of cells. Since MDA is the final product of lipid oxidative modification (Hasim and Yusuf, 2017), the higher MDA contents on the said days and consistent expression of two LOXs (novel.642, COCNU_11G010490 and COCNU_08G002490) indicates RPW attack can induce changes in ROS homeostasis, and may result in membrane damage (Figure 4; Supplementary Table S4). Taken together, the RPW attack initiates changes in ROS levels, which trigger differential activities of antioxidant enzymes i.e., CAT, POD, and SOD.

Survival and the performance of herbivores on plants tissues are linked with the phytoalexins i.e., (iso)flavonoids, terpenoids, and alkaloids. These secondary metabolites are directly associated with plant fitness during insect attack (Walling, 2000; Sosa et al., 2004). Thus, any expression changes in the associated genes/pathways are important to understand how coconut leaves respond to RPW attack. The observations that PAL, CAD, and C4H transcripts were highly expressed in C and D population density of RPW are important from defense point of view. PAL is the key enzyme that assists in inducing the synthesis of SA, which causes systemic resistance in many plants (Kim and Hwang, 2014). An earlier investigation showed that wounding by chewing insect (aphid) induced increased PAL expression, which lead to higher SA content; the SA content was positively correlated with the damage level (Chaman et al., 2003). Additionally, the CCR expression was also shown to be related to the SA biosynthesis, thus, a similar mechanism in coconut leaves can be present (Tronchet et al., 2010). Similarly, C4H and CAD are reported to participate in the biosynthesis of phytoalexins (Kochs et al., 1992; Tronchet et al., 2010). Thus, the expression changes in PAL, CAD, and C4H are an indication of defense activation in coconut leaves in response to RPW infestation. Further gene/pathway specific exploration can highlight the roles of PAL, CAD, and related genes in defense response against RPW. Other than these, the expression changes in a large number of transcripts associated with the phenylpropanoid and flavonoid biosynthesis pathway are consistent with a similar study of transcriptome analysis of Phoenix canariensis in response to RPW attack (Giovino et al., 2015). On the contrary, the lower expressions of CHI, CHS, ANS, LAR, and DFRs in C and D on both DAIs indicate that RPW attack possibly causes reduction in anthocyanin biosynthesis. An earlier study in Arabidopsis had shown the relatedness of anthocyanins with the resistance against aphid (Onkokesung et al., 2014). Nevertheless, anthocyanins play multiple roles in plants and their reduced biosynthesis may disturb the balance between photosynthesis and defense against RPW attack in coconut similar to Pseudowintera colorata (Menzies et al., 2016). For instance, overexpression of LAR or ANR resulted in elevation of tannins content and confers cassava resistance to two-spotted spider mite (Chen et al., 2022). Taken together, RPW attack induces expression changes in phenylpropanoid, flavonoid, and anthocyanin biosynthesis genes. Furthermore, the expression changes in higher number of genes in C and D RPW population densities on 20 DAI as compared to five DAI indicate that coconut leaves undergo higher changes in flavonoid and associated pathways when more days have passed after the infestation. In addition, the numerous potential candidate genes could be useful to breeding for resistant coco cultivars.

The higher expression of genes such as CaM, CDPK, HtpG, RPS2, RPM1, and EIX receptor 1/2 in response to both C and D population densities of RPW on 5 and 20 DAI indicate that coconut responds by activating defense mechanisms against RPW (Figure 5). Ca²⁺ signaling is crucial for plant-herbivore interaction (Parmagnani and Maffei, 2022). The higher expression of CDPK indicate that the RPW attack triggered changes in Ca²⁺ are being conveyed via enzymatic reactions (DeFalco et al., 2010) as well as by non-catalytic sensor relay proteins i.e., CaMs (Qiu et al., 2012). These changes are consistent with the earlier suggested roles of CaM in regulating wound-induced signaling in plants (Qiu et al., 2012). The upregulation of RPS2 and RPM1 proteins as well as HSP90 in RPW infested coconut leaves indicate that RPW have established a sustained interaction with host plant. The RPS2 and RPM1 proteins are part of plant-pathogen interaction pathway and can positively trigger increased expression of HSP90, which induce hypersensitive response (HR) (https://www.genome.jp/pathway/ko04626 (Arakawa et al., 2005)). Although HR is induced by fungi, oomycetes, bacteria, and viruses, but its induction by insects through unknown mechanisms is also reported in Salix viminalis against Dasineura marginematorquens (Höglund et al., 2005) and in wheat against Hessian fly (Aljbory et al., 2020). Thus, the transcriptome analysis of coconut leaves indicates that RPW induced changes in Ca²⁺ sensing and signaling, and HR.

Since signal transduction pathways mediated by phytohormones are one of the major defense strategies in plants against abiotic stresses, therefore, our results that multiple genes associated with plant-hormone signal transduction and MAPK-signaling (plant) are important (War et al., 2012). The changes in the expression of PYR/PYLs and PP2C indicate ABA reception and signaling in coconut leaves in response to infestation by both C and D population densities on 5 and 20 DAI (Figure 5B; Supplementary Table S4). These observations are consistent with earlier reports that ABA signaling is primary driver of oil palm against RPW attack (Harith-Fadzilah et al., 2021). However, our results also imply that other phytohormones including auxin, ethylene, SA, JA, and cytokinins are involved in coconut’s leaf responses to RPW infestation (Kunkel and Brooks, 2002). The expression changes in ET receptor genes and EIN3 indicate that ethylene signaling is involved in coconuts’ responses against RPW infestation. Since it is known that ethylene and JA work together in Alnus glutinosa L. (Horiuchi et al., 2003) against invading insect by enhancing terpenes and volatiles, therefore, an interplay of these hormones might also be functional in coconut leaves under RPW attack. A detailed understanding of the role of ethylene in coconut against RPW infestation is needed by exploring expressions of the above-mentioned genes. Whereas the higher expressions of the PR1 transcripts in C and D on 5 and 20 DAIs may be a result of ethylene signaling since PR1 RNA accumulation in tomato was dependent on ethylene signals (Ciardi et al., 2001). However, the changes in PR1 expressions could also be due to SA signaling as reported in Arabidopsis infested with whitefly (Zarate et al., 2007). Apart from these signaling related gene expressions, the upregulation of both MAPKKK4/5 and MAPKKK17/18 transcripts is also an important preliminary data and indicates that RPW infestation can activate these genes in coconut leaves (Figure 5B; Supplementary Table S5). Since MAPK signaling, together with JA, SA, and ethylene signaling, is required for transcriptional activation of herbivore defense-related genes and metabolites (Hettenhausen et al., 2015), therefore, coconut’s defense can be improved by manipulating these genes.

Insect attack alters carbon allocation patterns in plants in such a way that it triggers carbon remobilization from damaged to undamaged tissues, mostly towards the leaves to support defense (Schwachtje et al., 2006; Gómez et al., 2012). Increased sugar biosynthesis is essential to fuel the energy requirements for plant defense, scavenge ROS, initiate signaling cascades for activation/deactivation of defense genes, and change the hormone levels (Trouvelot et al., 2014). Our results that large numbers of transcripts enriched in amino sugar and nucleotide sugar metabolism pathways were increasingly expressed in coconut leaves after RPW infestation are relevant to the above-mentioned reports (Figure 5C; Supplementary Table S4). These expression changes indicate that RPW infestation initiated sugar metabolism both on 5 and 20 DAI in response to an attack by different RPW population densities. The general trend that on expressions on 20 DAI were higher than that of on five DAI indicates that higher sugars are being synthesize when a greater number of days have passed after infestation. These results should be further validated on more coconut varieties and after confirmation can be adapted as markers for RPW infestation (Ray et al., 2016).

The induction of a large number of TF families indicate that TFs are an essential part of the coconut’s responses to RPW infestation. Expression changes in a large number of AP2/ERF transcripts signify that ethylene signaling is an important part of the coconut leaf’s responses towards RPW infestation. Earlier studies have shown that AP2/ERFs were induced after Spodoptera exigua and Pieris rapae L. infestation (Rehrig et al., 2014). Tomato plants overexpressing AP2/ERF TF JRE4 showed increased biosynthesis of steroidal glycoalkaloids, which increased plant defense against Spodoptera litura larvae infestation (Nakayasu et al., 2018). On the other hand, the expression changes in bHLH, MYB, and WRKY TFs (Supplementary Table S5) present novel candidates for improving RPW resistance in coconut. Their upregulation in coconut leaves infested with C and D population densities of RPW on different DAI present them as defense related TFs. These expression changes are consistent with earlier reports on the roles of bHLH in rice against brown planthopper (Wang et al., 2019), MYB in Arabidopsis against Helicoverpa armigora (Shen et al., 2018), and WRKYs in rice against Chilo suppressalis (Li et al., 2015). Taken together, our findings indicate that RPW infestation can induce a large number of TFs belonging to diverse families.

RPW infestation on coconut can trigger changes in ROS levels and antioxidative enzyme activities. Generally, the higher RPW population densities i.e., C and D, were more damaging to coconut leaves as compared to CK. With the increase in the number of days after infestation, the expressions of a relatively higher number of genes are induced. The pathways which are highly regulated in response to RPW attack include amino sugar and nucleotide sugar biosynthesis, flavonoid biosynthesis, plant-pathogen interaction, plant-hormone signaling, MAPK-signaling, and ROS scavenging. We present a large number of candidate genes that can be used to detect early (5 DAI) or late (20 DAI) RPW infestation in coconut. We also report a preliminary understanding of the coconut plant’s transcriptomic responses to RPW attack.