Aquilaria sinensis (Lour.) Gilg (Thymelaeaceae) produces the fragrant agarwood that is widely used in traditional medicine and the incense industry (Jin et al. 2016; Wen et al. 2009). Heortia vitessoides Moore (Lepidoptera: Crambidae) is considered to be the most severe pest of A. sinensis and ranges from India, Nepal, China, and Sri Lanka through Southeast Asia and the East Indies to Queensland, the New Hebrides, and Fiji. In southern China, the insects have seven or eight generations per year, and the larvae feed on the leaves of A. sinensis, causing significant economic losses (Qiao et al. 2012; Wen et al. 2017). Carbamates, pyrethroids, anthranilic diamides, and plant-derived insecticides such as fenoxycarb, beta-cypermethrin, chlorantraniliprole, and matrine have been used to control H. vitessoides infestations. However, these insecticides have become less efficient, even as mixtures or at relatively high doses (Chen et al. 2011; Lu et al. 2014; Su 1994).

The cytochrome P450 (CYP) monooxygenases, a large and complex gene superfamily of heme-thiolate proteins, are ubiquitously expressed in almost all living organisms (Yu et al. 2015). Insect CYPs can be divided into four major clans: three microsomal CYP clans (CYP2, CYP3, and CYP4) and a mitochondrial CYPs clan (Feyereisen 2011). In insects, CYPs perform many important functions including involvement in biosynthetic pathways of juvenile hormones, ecdysteroids, and pheromones which are closely related to insect growth, development, and reproduction (Lao et al. 2015). Moreover, insect CYPs are well-known for their vital role in the detoxification of various types of synthetic insecticides such as chlorantraniliprole (Hu et al. 2014), malathion (Li et al. 2016), and chlorpyrifos (Wang et al. 2017). Increased CYP activity has been reported to be one of the main reasons for insecticide resistance in other lepidopteran species such as Helicoverpa armigera Hübner (Zhou et al. 2010), Cnaphalocrocis medinalis Guenée (Liu et al. 2015), and Cydia pomonella Linnaeus (Bosch et al. 2018).

The identification and functional analysis of candidate CYP genes are important first steps in investigating the mechanisms of insecticide resistance in insects. Next-generation sequencing techniques such as RNA sequencing (RNA-seq), aided by decreasing costs and technical advances, have become valuable tools that allow vast amounts of genetic information to be acquired from nonmodel organisms without prior sequence knowledge (Liu et al. 2016). For H. vitessoides, which does not have a sequenced genome, we deposited three transcriptomes in the National Center for Biotechnology Information (NCBI) Sequence Read Archive (SRA; adult: SRX3035102, female antennae: SRX3136158, male antennae: SRX3136160) (Cheng et al. 2017). However, extensive genomic and transcriptomic sequences are still required for H. vitessoides.

In the present study, we sampled H. vitessoides eggs, larvae, pupae, and adults and used a BGISEQ-500 sequencing platform to generate a large-scale dataset, utilizing bioinformatics analyses to focus on the genes encoding putative CYPs. We identified 46 putative CYP genes in H. vitessoides. Moreover, the expression profiles of genes belonging to CYP4 and CYP6 families exposed to half the lethal concentrations (LC₅₀) of chlorantraniliprole and beta-cypermethrin were determined using real-time reverse transcription-quantitative PCR (RT-qPCR). To the best of our knowledge, this is the first report of the identification and characterization of multiple CYP genes in this forest pest.

Materials and Methods

Insect rearing and sample collection. Heortia vitessoides eggs and larvae were collected in May 2017 from an A. sinensis plantation (N 22°01′, E 110°25′) in Huazhou, Guangdong, China. No chemical treatment was applied before or during collection. All insects were reared in the laboratory under conditions of 26°C with 70 ± 2% relative humidity and maintained at a 14 h:10 h light:dark cycle. Eggs, larvae, pupae, and adults were collected from that colony and were pooled together, then immediately frozen in liquid nitrogen and stored at –80°C for total RNA extraction.

RNA sample preparation. Total RNA was extracted using the E.Z.N.A.™ Total RNA kit II (OMEGA Biotec, Norcross, GA, USA) following the supplier's instructions and then treated with DNase I (Invitrogen, Life Technologies, Carlsbad, CA, USA). A Nanodrop 2000 spectrophotometer (NanoDrop Products, Wilmington, DE, USA) was used to check sample purity while a Qubit 2.0 fluorometer (Life Technologies, Gaithersburg, MD, USA) and Quantifluor-ST fluorometer with Agilent 2100 Bioanalyzer (Promega, Madison, WI, USA) were used to measure the concentration and integrity, respectively.

cDNA library construction. The qualified RNA samples were used for transcriptome sequencing. The first step involved purifying the poly-(A)-containing mRNA molecules using poly-T oligo-attached magnetic beads. Following purification, the mRNA was fragmented using divalent cations under an elevated temperature. The cleaved RNA fragments were reverse transcribed to form the first-strand cDNA using reverse transcriptase and random primers. This was followed by second-strand cDNA synthesis using DNA Polymerase I and RNase H. A single “A” base was added to these cDNA fragments, followed by the ligation of DNA adapters. The products were then purified and enriched with PCR amplification. We then quantified the PCR yield using Qubit and pooled samples to generate a single strand DNA (ssDNA) circle, which formed the final library. The cDNA library was sequenced using the BGISEQ-500 platform (BGI, Shenzhen, China). The raw reads were saved as FASTQ files and deposited in the NCBI SRA with the accession number SRX4045498.

De novo assembly and functional annotation. Prior to assembly, we obtained clean reads from the raw data by removing reads containing adaptor sequences, more than 5% unknown nucleotides, more than 50% bases with Q-value ≤20, and empty reads. These clean reads were then de novo assembled into unigenes using the short reads assembling program Trinity (Grabherr et al. 2011). To acquire comprehensive information on gene functions, assembled unigenes over 150 bp in length were searched against the NCBI nonredundant protein sequences (Nr), NCBI nucleotide (Nt), eukaryotic ortholog groups (KOG), SwissProt, and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases using BLASTx and BLASTn with an E-value <10^–5. Blast2GO (Conesa et al. 2005) was used for gene ontology (GO) annotation with an E-value <10^–5 based on the protein annotation results of the Nr database. InterPro functional annotation was performed using InterProScan5 (Quevillon et al. 2005).

Gene identification and bioinformatic analyses. The BLASTn program was used to identify candidate unigenes encoding putative CYP monooxygenase genes in H. vitessoides (HvCYPs) using available sequences of these proteins from lepidopteran insects. All candidate genes were manually checked using BLASTx on the NCBI website. All putative HvCYPs were named in accordance with David R. Nelson to maintain consistency in the nomenclature.

Open reading frames (ORFs) were predicted using the ORF Finder (http://www.ncbi.nlm.nih.gov/gorf/gorf.html). The functional domains and core catalytic residues were predicted by searching the Conserved Domain Database (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi). Amino acid sequences were aligned with MAFFT (http://mafft.cbrc.jp/alignment/server/clustering.html). Phylogenetic trees were constructed using the MEGA5.0 based on the neighbor-joining method with the p-distance model, including trees of 34 putative HvCYPs (amino acid residues >300aa) and three insects (Drosophila melanogaster Meigen, Apis mellifera L., and Bombyx mori L.) (Tamura et al. 2011). Node support was assessed using a bootstrap procedure based on 1,000 replicates, and node support values <50% are not shown.

Insecticide treatment. Chlorantraniliprole and beta-cypermethrin were purchased from Fengle Agrochemical Co., Ltd., (Hefei, China) and diluted with analytical-grade acetone to make a working solution, 7.7 × 10^–4 mg L^–1 for chlorantraniliprole and 8.9 × 10^–5 mg L^–1 for beta-cypermethrin (LC₅₀ values) (Chen et al. 2011). The freshly-molted fourth-instar larvae were selected and starved for 2 h. The leaf-dipping method was employed to investigate insecticidal activity (Chen and Zhang 2015). Fresh A. sinensis leaves were dipped into the pesticide solutions for 10–15 s and air dried at 26°C, then fed to the starved larvae. Control insects were fed with A. sinensis leaves with acetone only. Insecticide-treated and control insects were collected after 24 h, immediately frozen in liquid nitrogen, and stored at –80°C prior to RNA extraction. Each sample consisted of 15 larvae with three independent replicates.

Real-time reverse transcription-quantitative PCR (RT-qPCR).To determine the transcriptional changes of six CYP4 and 12 CYP6 family genes in the fourth-instar larvae after exposure to the LC₅₀ of chlorantraniliprole and beta-cypermethrin, RT-qPCR was performed using cDNA prepared from insecticide-treated and control insects. Total RNA was extracted as described above. First-strand cDNA was synthesized from 2 µg total RNA using the PrimeScript^® RT reagent kit with gDNA Eraser (Takara Bio, Otsu, Japan) and then immediately stored at –80°C for further use. The RT-qPCR was performed using SYBR^® Premix Ex TaqTM II (Takara Bio, Otsu, Japan). Each reaction (20-µL volume) contained 2 µL cDNA, 10 µL SYBR Premix Ex Taq, 0.4 µL forward and reverse primers (10 µM), and 7.2 µL RNase-free double distilled water. The gene-specific primers (Table 1) were designed using Primer Premier 5.0 (Premier Biosoft International, Palo Alto, CA, USA) and synthesized by TSINGKE Biotech Co., Ltd. (Guangzhou, China). The housekeeping gene β-actin (GenBank accession number MG132199) was used as reference gene in the RT-qPCR experiments (Cheng et al. 2018). RT-qPCR on the cDNA products was carried out in 96-well plates using a Light Cycler 480 (Roche Diagnostics, Indianapolis, IN, USA). The amplification conditions were as follows: initial denaturation at 95°C for 5 min; 40 cycles at 95°C for 10 s and 60°C for 20 s; and cooling at 40°C for 30 s. Negative controls were nontemplate reactions (replacing cDNA with diethyl pyrocarbonate water), and the results were analyzed using the LightCycler^® Real-Time PCR system. Three biological and technical replicates each were set for RT-qPCR analysis. The quantity of CYP mRNAs was calculated using the 2^–ΔΔCt method (Pfaffl 2001).

Table 1 Primers used in this study (F ¼ forward; R ¼ reverse).

View Fullscreen

Table 1 Continued

View Fullscreen

Statistical analysis. Gene expression data are presented as the means ± standard deviation (SD) of three independent replicates. To compare the differences in gene expression between the insecticide-treated and control larvae, a paired Student's t-test was performed. P < 0.05 was considered statistically significant. Data analysis was conducted using SPSS 18.0 (SPSS, Inc., Chicago, IL, USA).

Results

Sequencing and de novo assembly. We performed next-generation sequencing of the cDNA library constructed from the eggs, larvae, pupae, and adults of H. vitessoides using the BGISEQ-500 platform. Transcriptomic sequencing provided 117.04 Mb raw reads. After removing adaptor, low quality, and N-containing sequences, 110.53 Mb clean reads were generated. After assembly, we obtained 42,946 unigenes with an average length of 1,059 bp and an N50 of 1,944 bp (Table 2). The size distribution of the assembled unigenes is shown in Fig. 1.

Table 2 Overview of transcriptome of Heortia vitessoides.

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

Sequence annotation. The number and percentage of matched unigenes at different values is shown in Table 3. In summary, 22,106 (51.47%), 12,455 (29.00%), 15,297 (35.62%), 16,833 (39.20%), 14,837 (34.55%), 15,006 (34.94%), and 4,510 (10.50%) unigenes had homologous sequences in the Nr, Nt, SwissProt, KEGG, KOG, Interpro, and GO databases, respectively. The total unigenes annotated by any of the seven functional databases was 56.11%. Only 4.76% of the unigenes were annotated in all databases. For species distribution, the highest match percentage was to H. armigera (26.18%) sequences followed by Amyelois transitella Walker (24.10%), B. mori (8.98%), and Papilio xuthus Linnaeus (6.64%, Fig. 2).

Table 3 Summary of unigene annotations.

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

GO annotation was used to classify the function of unigenes according to the GO terms (Fig. 3). In biological processing terms, “cellular processes” (1,632), “metabolic processes” (1,403), and “biological regulation” (536) were the most abundant. In cellular component terms, “cell” (1,504), “cell part” (1,481), and “membrane” (1,247) were the highest classified. In molecular function terms, genes involved in binding (2,085), catalytic activity (1,655), and structural molecule activity (307) were the most abundant.

View Figure

• Download as Powerpoint
• Download Figure

In total, 14,837 unigenes were assigned to 25 KOG functional categories (Fig. 4). Of these categories, “general function prediction only” represented the largest group, containing 3,775 unigenes, followed by “signal transduction mechanisms” (3,230) and “function unknown” (1,689). The “nucleotide transport and metabolism” (189), “coenzyme transport and metabolism” (134), and “cell motility” (53) categories were the smallest clusters represented.

View Figure

• Download as Powerpoint
• Download Figure

To understand the biological pathways active in H. vitessoides, the sequences were mapped to reference canonical pathways in KEGG (Fig. 5). In summary, 16,833 unigenes were classified into six groups, “cellular processes,” “environmental information processing,” “genetic information processing,” “human diseases,” “metabolism,” and “organismal systems.” “Transport and catabolism” (1,262), “signal transduction” (2,451), “translation” (1,001), “cancers: overview” (1,574), “global and overview maps” (2,277), and “endocrine system” (1,405) were the dominant pathways in each group, respectively.

View Figure

• Download as Powerpoint
• Download Figure

Identification and characterization of CYP genes from H. vitessoides. In total, 46 putative CYP genes were identified in the H. vitessoides transcriptome. Of these, only six genes had completed ORFs while the remaining 40 genes consisted of incomplete cDNAs, missing a portion of the sequence (Table 4). According to the standard nomenclature, the 46 HvCYPs were divided into 22 families and 34 subfamilies (Table 4). The largest family was the CYP6 family, which included 12 genes. These CYP genes have been deposited in GenBank with the accession numbers MH236440–MH236485 (Table 4).

Table 4 Details of 46 cytochrome P450 (CYP) monooxygenases identified in H. vitessoides.

View Fullscreen

Table 4 Continued

View Fullscreen

Table 4 Continued

View Fullscreen

Table 4 Continued

View Fullscreen

Information from the BLASTx search of the best hits for all 46 CYP genes is provided in Table 4. All HvCYPs genes had a relatively high sequence identity (57–93%) with their respective orthologs from other lepidopteran species. Multiple sequence alignment analysis revealed that these CYP genes had five conserved domains; a helix-C (WxxxR), helix-I (GxE/DTT/S), helix-K (ExLR), PERF (PxxFxPE/DRE) and heme-binding motif (PFxxGxRxCxG/A) (Fig. 6) (Ai et al. 2011). However, one microsomal CYP (CYP49A1) lacked three residues in the heme-binding motif.

View Figure

• Download as Powerpoint
• Download Figure

A phylogenetic analysis was conducted to evaluate the relationships between the HvCYPs and CYPs from three other model insect species. In the phylogenetic tree (Fig. 7), the 34 HvCYPs genes were allocated to four groups representing different CYP clans including nine within the CYP4 clan, 17 in CYP3, two in CYP2, and six within the mitochondrial clan. The HvCYPs from four CYP clans were clustered into different subfamilies, such as CYP6AE and CYP9G from the CYP3 clan, CYP302A and CYP315A from Mito clan, CYP18A from CYP2 clan, and CYP341A and CYP340A from the CYP4 clan (Fig. 7). This tree demonstrated that there is a close relationship between CYP genes from B. mori and H. vitessoides (Fig. 7).

View Figure

• Download as Powerpoint
• Download Figure

Expression of CYP genes in larvae exposed to insecticides. Transcriptional changes of the CYP4 and CYP6 family genes in the fourth-instar larvae, after exposure to LC₅₀ of chlorantraniliprole and beta-cypermethrin, were determined by RT-qPCR. In chlorantraniliprole-treated insects, the expression of eight genes (CYP4M39, CYP6AB49, CYP6AB53, CYP6AB61, CYP6AE17, CYP6AW1, CYP6BD6, and CYP6CV1) was significantly higher than that in the control insects after 24 h of chlorantraniliprole exposure, and the expression of three genes (CYP6AB10, CYP6AB47, and CYP6CT1) was markedly downregulated compared with that in the control insects (Fig. 8). In the beta-cypermethrin-treated insects, the expression of five genes (CYP6AB10, CYP6AB53, CYP6AE12, CYP6AE17, and CYP6BD6) was significantly up-regulated compared with that in the control insects (Fig. 8). Moreover, beta-cypermethrin significantly decreased the mRNA levels of CYP4G24 in the treated insects compared with that in the control insects (Fig. 8).

View Figure

• Download as Powerpoint
• Download Figure

Discussion

In recent years, many CYP genes have been identified in various insect species, especially in insects for which whole genomic sequencing has been performed. Bioinformatic analyses have revealed 84 CYP genes in B. mori (Ai et al. 2011), 90 genes in D. melanogaster (Tijet et al. 2001), 46 genes in A. mellifera (Claudianos et al. 2006), and 143 genes in Tribolium castaneum Herbst (Zhu et al. 2013). Within the order Lepidoptera, 63 CYP genes were identified in Pieris rapae L. (Liu et al. 2018), 85 in Plutella xylostella L. (Yu et al. 2015), and 77 in Chilo suppressalis Walker (Wang et al. 2014). As mentioned above, whole-genome information for H. vitessoides is currently unavailable; searching of transcriptome datasets can be used to identify new genes, including CYP genes. This approach has been used successfully for other insects lacking genomic data, such as C. medinalis (Liu et al. 2015) and Liposcelis entomophila Enderlein (Li et al. 2016). In the current study, we first generated the transcriptome dataset from four developmental stages of H. vitessoides. From the dataset, 46 CYP genes were identified. The number of CYP genes identified in H. vitessoides was clearly lower than in other lepidopteran species. There are two possible reasons for this result; firstly, previous studies investigated CYP expression in virtually all tissue types and throughout insect development (Wang et al. 2018). In contrast, we sequenced limited samples and may have missed CYP genes from other tissues or developmental stages. Secondly, the current sequencing technology might not be sufficiently powerful to screen all CYP genes, especially the transcripts expressed at very low levels (Liu et al. 2015).

Based on their evolutionary relationship, insect CYPs can be classified into four major clans: the CYP2, CYP3, CYP4, and mitochondrial. The genes of CYP2 clan play basic physiological functions in insects. In D. melanogaster, for example, CYP15A1 is involved in juvenile hormone metabolic pathways and CYP307A1 is tightly correlated with the biosynthesis of 20-hydroxyecdysone (Iga and Kataoka 2012; Rewitz et al. 2007). The CYP3 clan is a large group of insect CYPs, and members of the CYP3 clan are mainly related to xenobiotic metabolism and insecticide resistance; for instance, CYP321A1 and CYP6B8 have been found to metabolize plant allelochemicals in Helicoverpa zea Boddie (Rupasinghe et al. 2007). In Heliothis virescens F., CYP9A1 was inducible by thiodicarb, indicating a potential role in insecticide metabolism (Rose et al. 2006). CYP4 genes are involved in odorant and pheromone metabolism, fatty acid hydroxylation, and biosynthesis and metabolism of hormones, such as juvenile hormone and others associated with the gonadotropic cycle, as well as in insecticide resistance (Bergé et al. 1998, Wang et al. 2017). For example, CYP341A2 in P. xuthus acts as a degrading enzyme that plays a role in the chemosensory reception for host plant recognition (Ono et al. 2005). In Bemisia tabaci Gennadius, increased expression of CYP4C64 has been reported to be the main reason for imidacloprid resistance (Yang et al. 2013). Genes within the mitochondrial clan have highly specific functions in insects, such as the CYP302a1, CYP314a1, and CYP315a1 genes which are involved in biosynthesis, activation, and inactivation of 20-hydroxyecdy-sone in Aedes aegypti L. and D. melanogaster (Sztal et al. 2012). In this study, 46 HvCYPs, divided into four major clans and further assigned to 22 families and 34 subfamilies, might be connected with diverse functions. The functional diversity might lead to a better adaptation for H. vitessoides to ecological environments. However, further study is warranted to provide an in-depth confirmation of the above hypothesis.

An important mechanism that gives rise to insecticide tolerance is the metabolism of insecticides by the products of the over-expressed CYP genes and, in particular, CYP4 and CYP6 family genes (Li et al. 2007). For example, CYP4D4v2, CYP4G2, and CYP6A38, related to permethrin resistance, were upregulated by permethrin exposure in Musca domestica L. (Zhu et al. 2008). In Lymantria dispar L., the expression of 12 CYP6 family genes (CYP6AB32, CYP6AB33, CYP6AB34, CYP6AB35, CYP6AB36, CYP6AB37, CYP6AE51, CYP6AE52, CYP6AN15v1, CYP6AN16, CYP6B53, and CYP6CT4) was significantly upregulated by exposure to different insecticides (deltamethrin, omethoate, and carbaryl) (Sun et al. 2014). Similarly, products of CYP6B8 and CYP6B27 from H. zea can detoxify multiple insecticides including aldrin, diazinon, carbaryl, and α-cypermethrin (Wen et al. 2009). Accordingly, determination of insecticide-inducible CYP4 and CYP6 family genes may lead to the identification of candidates that are involved in insecticide tolerance. In this study, three CYP6 family genes (CYP6AB53, CYP6AE17, and CYP6BD6) were significantly upregulated following exposure to LC₅₀ of chlorantraniliprole and beta-cypermethrin. Moreover, CYP4M39, CYP6AB49, CYP6AB61, CYP6AW1, and CYP6CV1 were significantly overexpressed under the stress of chlorantraniliprole whereas CYP6AB10 and CYP6AE12 were upregulated in the beta-cypermethrin-treated larvae compared with the control insects. These genes are, therefore, potential candidates involved in the detoxification of chlorantraniliprole and beta-cypermethrin, and further investigation of their functions using reverse genetic manipulation tools, such as the RNA interference (RNAi) approach, would contribute to enhancing our understanding of these genes.

In contrast, several CYP4 and CYP6 family genes were downregulated by insecticide exposure. For example, the expression of CYP4D47 was downregulated by both malathion and beta-cypermethrin exposure in Bactrocera dorsalis Hendel (Huang et al. 2013). In Leptinotarsa decemlineata Say, an exposure to cyhalothrin markedly reduced the expression levels of three CYP genes (CYP6BU1, CYP9Z10, and CYP12J1) (Wan et al. 2013). Our results demonstrated that the transcript levels of three CYP6 family genes, CYP6AB10, CYP6AB47, and CYP6CT1, were significantly downregulated under stress of chlorantraniliprole. Moreover, beta-cypermethrin exposure also significantly reduced the mRNA level of CYP4G24. Transcriptional suppression of CYP genes was regarded as being related to multiple molecular mechanisms, and these processes often involve complex cascades of transcription factors and other regulatory proteins, which may create adaptive homeostasis in these organisms (Riddick et al. 2004). In addition, the expression levels of the remaining CYP4 and CYP6 family genes had no significant response to insecticide treatment, indicating that these genes might play a minor role in the metabolism of the tested insecticides or may not be sufficiently activated after 24-h exposure.

In summary, a de novo transcriptome was assembled for H. vitessoides, and 46 putative CYP genes were identified from this dataset for the first time in this study. These CYP genes were classified into four clans consisting of 22 families and 34 subfamilies. Furthermore, RT-qPCR results indicated that several CYP4 and CYP6 family genes were upregulated following exposure to LC₅₀ of chlorantraniliprole and beta-cypermethrin and are potential candidates involved in the detoxification of these insecticides.