1. Introduction
Henosepilachna vigintioctopunctata (Fabricius) (Coleoptera: Coccinellidae) is a widely distributed beetle in Asian countries, including China, India, Japan, and Korea [
1,
2,
3,
4]. This species is considered to be a phytophagous pest that mainly causes damage to Solanaceae plants, particularly eggplants, along with other crops such as potatoes, pumpkins, and rock melons [
5]. Both the adult and larval stages of
H. vigintioctopunctata feed on the leaves of these plants by scraping the leaf cuticle, thereby promoting the development of irregular transparent spots or perforations in the injured leaves, which later become brown depression marks. In addition, these beetles can cause notable damage to tender stems, petals, sepals, and fruits, resulting in reduced plant yield, poor fruit quality, or plant withering [
6,
7]. During the past 20 years, a large number of studies have been conducted on the basic ecology and biology of
H. vigintioctopunctata [
8,
9,
10,
11,
12]. At present, chemical pesticides are primarily used for controlling
H. vigintioctopunctata. However, given the current requirement for negative growth in pesticide use in China, as stipulated in the 2019 “No. 1 Central Document,” there is an urgent need to develop effective new pest control methods that target
H. vigintioctopunctata based on novel modes of action.
In this latter regard, RNA interference (RNAi) provides an effective molecular tool for the determination of gene function, and can also be exploited to facilitate the development of targeted and environmentally friendly pest control methods with new modes of action [
13,
14,
15,
16,
17,
18]. In terms of pest control, RNAi-mediated silencing has shown excellent efficacy and has been applied in the control of a number of coleopteran insects, including the Western corn rootworm
Diabrotica virgifera virgifera [
19], red flour beetle
Tribolium castaneum [
20], and the Colorado potato beetle
Leptinotarsa decemlineata [
20,
21]. Moreover, our preliminary experimental results indicated that dietary RNAi targeting multiple genes is particularly effective against
H. vigintioctopunctata, indicating that RNAi has potential utility in the control of this species.
From the perspective of pest control, gaining an understanding of the genes involved in gonadal differentiation and reproduction could make a considerable contribution toward the development of novel control strategies. Numerous genes have been investigated to evaluate their roles in the gonadal differentiation and reproduction of insects, including
transformer 2 in the Asian citrus psyllid
Diaphorina citri [
22];
vATPase A,
vATPase E,
vitellogenin, and
Brahma in the common bed bug
Cimex lectularius [
23,
24,
25];
lipase maturation factor in the grain aphid
Sitobion avenae [
26];
vitellogenin receptor and
boule in
D. virgifera virgifera [
27];
Ran in the brown planthopper,
Nilaparvata lugens [
28];
Fatty acyl-Co A reductase in the cotton plant bug
Adelphocoris suturalis [
29]; and
boule,
Zero growth Population,
double sex male,
fzo, and
gas8 in the fruit fly
Bactrocera dorsalis [
30]. Silencing these genes that are highly expressed in the ovary or testis has been shown to result in markedly reduced egg production and a lower rate of hatching. Although the egg production capacity of
H. vigintioctopunctata is relatively high, with each female capable of laying approximately 618 eggs in its lifetime [
9], the regulatory molecular aspects of reproduction in
H. vigintioctopunctata remain unclear. Accordingly, we reasoned that a molecular investigation of
H. vigintioctopunctata reproduction would provide important insights with regard to identifying novel target sites for population control.
The first step on the road to understanding the molecular mechanisms of reproduction is to identify gonad-related genes. In this study, we employed Illumina-based sequencing and de novo assembly to screen for gonad-related genes by constructing the transcriptomes of ovaries and testes in H. vigintioctopunctata. We accordingly obtained 69,854 unigenes from the six libraries comprising the ovary and testis transcriptomes of H. vigintioctopunctata. Moreover, we performed comparative transcriptome analysis to identify the genes that were differentially expressed between ovaries and testes, among which, 13 and 16 candidate unigenes that were upregulated in the ovary and testis, respectively, were selected to confirm their altered expression levels via RT-qPCR and semi-RT-qPCR analyses. Finally, we investigated the effects of silencing two ovary-specific genes, RC2-like and PSHS-like, on the mortality, percentage infertility, pre-oviposition period, and fecundity of adult females, along with the number of eggs laid daily and egg hatching rate. By comparing unigene sequences in the ovary and testis libraries, we were able to determine gonad-specific differentially expressed genes, which will contribute to a more comprehensive understanding of the regulatory mechanisms involved in gonadal differentiation and reproductive processes in H. vigintioctopunctata. Furthermore, these gonad-specific genes can potentially be used as targets for suppressing the population growth of this species.
3. Discussion
Although H. vigintioctopunctata has become an increasingly serious pest of solanaceous and cucurbitaceous plants in China in recent years, there is still relatively little genomic information available for this beetle. In this regard, transcriptome analysis can provide a basis for future gene expression and functional analyses of H. vigintioctopunctata. The findings of the present study will also augment the limited sequence data currently available for phytophagous ladybugs.
In this study, we performed RNA-seq analysis of the ovaries and testes of newly eclosed adult males and females of
H. vigintioctopunctata in an attempt to identify gonad-related genes. We accordingly obtained 69,854 unigenes with an average length of 780 bp. This length is considerably longer than the unigenes obtained from the gut and salivary glands of the whitefly
Bemisia tabaci [
31,
32] and the intestines and fat bodies of the brown planthopper
N. lugens [
33,
34], although notably shorter than the unigenes from the small white butterfly
Pieris rapae [
35]. The 69,854 unigenes with the highest number of BLASTx hits with the sequences of other species revealed high homology with several other coleopteran species, for which complete genome sequences are currently available. Nevertheless, we succeeded in annotating only 22,472 (32.17%) of the unigenes obtained, possibly owing to the limited published data for phytophagous ladybug transcriptomes. Accordingly, further research is necessary to functionally characterize these genes.
Gene ontology distributions are known to differ substantially, both among the transcriptomes of various body parts within the same insect and between the transcriptomes of different insects [
36,
37,
38]. For example, the GO distribution determined in the present study was found to show a pattern similar to the gonad transcriptomes of other insects, such as the testis transcriptomes from the cockroach
Periplaneta americana [
36] and the oriental fruit fly
Bactrocera dorsalis [
38]. However, it differs significantly from the transcriptomes of many other insects, such as the antennal transcriptome of the Asian long-horned beetle
A. glabripennis [
37] and the intestine-specific transcriptome of the brown planthopper
N. lugens [
33]. We believe that the genes that are highly or specifically expressed in gonads show marked difference from those in other tissues because sex-biased expression tends to be high in the gonads but lower in other tissues [
39,
40,
41,
42]. This should perhaps be expected, given that the gonads show considerably greater sexual dimorphism than other tissues.
Comparisons of testis and ovary transcriptomes have previously been conducted for the green mud crab
Scylla paramamosain [
43], the swimming crab
Portunus trituberculatus [
44], and the Pacific white shrimp
Litopenaeus vannamei [
45]. Similarly, in the present study, we specifically sought to identify the gonad specifically/differentially expressed genes in
H. vigintioctopunctata. To the best of our knowledge, this is the first study to compare the testis- and ovary-specific transcriptomes of the same insect. Indeed, to date, the testes transcriptomes of only a relatively few insect species have been sequenced [
36,
38]. Among those that have been sequenced, 125,390 unigenes were obtained from the testis transcriptome of
P. americana [
36], whereas only 20,921 unigenes were annotated from the testes of
B. dorsalis [
38]. Therefore, we had very little data to compare with the number of testis- and ovary-specific genes identified in
H. vigintioctopunctata. The 1,421 and 5,315 genes identified as being specifically expressed in the ovary and testis of
H. vigintioctopunctata will thus provide important genetic information for the study of reproductive regulatory mechanisms in this species. Our preliminary study showed that, compared with the reproductive system of unmated females (
Figure S1B), that in the mated females in the dsGFP control group developed normally (
Figure S1A). In contrast, we found that the reproductive system of the mated females that did not lay eggs had developed abnormally (
Figure S1C). Furthermore, we noted that there was no obvious difference among the reproductive systems of the egg-laying mated females in the dsGFP control group and the dsRC2-like and dsPSHS-like treatment groups. However, although a large number of genes related to gonadal development have been characterized in the ovary and testis transcriptomes of
H. vigintioctopunctata (
Table S1), the precise role of these genes in the gonadal development and differentiation processes of this species remain to be determined.
In terms of quantity, the number of testis-specific unigenes identified was substantially greater than that of ovary-specific genes, which could be attributable to the fact that the ovaries and testis were dissected from the newly emerged H. vigintioctopunctata adults, indicating that original female development represents a type of “default” state. In order to validate the gene expression data obtained from transcriptomes analysis, we selected 13 and 16 genes that were specifically upregulated in the ovary and testis, respectively, for semi-RT-qPCR and RT-qPCR analysis using the same cDNA samples. We accordingly found that the expression levels of these 29 unigenes in ovary and testis were consistent with the transcriptome analysis, indicating that the transcriptome data were reliable.
In this study, we also randomly selected two genes, namely, RC2-like and PSHS-like, that were highly and specifically expressed in the ovary to investigate their roles based on the dietary delivery of dsRNAs. We accordingly recorded 10.53% and 26.32% mortality among the females in the dsPSHS-like and dsRC2-like treatment group, respectively, although this mortality was not significantly different from that of females in the dsGFP control group. Importantly, however, the fecundity of females was significantly reduced in the dsRC2-like and dsPSHS-like treatment groups compared with the dsGFP control females, and the pre-oviposition period was found to be markedly prolonged in the dsPSHS-like-treated females compared with those in the dsRC2-like and dsGFP groups. Therefore, we can speculate that RC2-like and PSHS-like, particularly the latter, could be used to inhibit the population growth of H. vigintioctopunctata.
4. Materials and Methods
4.1. Insect Rearing and Sample Collection
H. vigintioctopunctata were reared on eggplant leaves in the laboratory as described in our previous study [
46]. The ovaries and testes were dissected from adult males and females (collected during the first 2 days after emergence). All samples were rapidly frozen in liquid nitrogen and stored at −80 °C. Samples were collected in triplicate and these biological replicates were processed independently.
4.2. RNA Isolation, Library Preparation, and Sequencing
Total RNA was extracted using a RNeasy MinElute Cleanup Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Concentrations of the extracted RNA were determined using a NanoDrop 2000 spectrophotometer (Thermo Scientific, Madison, Wisconsin, USA) and the quality was verified using an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, California, USA). Ovary- and testis-specific RNA samples were pooled at an equal concentration. cDNA libraries were sequenced by Majorbio (Majorbio, Shanghai, China) using an Illumina platform (Illumina, San Diego, California, USA) with sequence runs of 2 × 150-bp paired-ends. The raw reads have been deposited in the Short Read Archive (SRA) of NCBI with the accession number PRJNA531518.
4.3. Sequence Assembly and Functional Annotation
The reads were initially assembled using Trinity (V2.2.0) [
47], after which duplicated contigs were removed using cd-hit [
48] to yield a unigene sequence file.
These unigenes were subsequently annotated against NR, Swiss-Prot, Pfam, COG, GO, and KEGG databases using associated software, the details and parameters of which are listed in
Table 5.
4.4. Identification of Differentially Expressed Genes in Ovary and Testis
In order to compare the unigene sequences of the ovary and testis libraries, we performed BLASTN alignments. The best matched (identity ≥90%) sequences were identified as genes that were similarly expressed in both libraries, whereas those with lower matches (identity <90%) were considered to be differentially expressed between the ovary and testis libraries. The Fisher P test was used to determine the differential expression of each pair of common genes, with a Fisher p-value of < 0.05 being considered indicative of a significant difference in gene expression.
4.5. Molecular Marker Detection
Simple sequence repeats (SSRs) were detected using MISA software [
49]. Six types of SSRs were investigated, namely, mono-, di-, tri-, quad-, penta-, and hexa-nucleotide repeats.
4.6. Semi-RT-qPCR and RT-qPCR Confirmation of Gonad-biased Genes
Semi-RT-qPCR and RT-qPCR were used to verify gene expression profiles in testis and ovary, as identified from the Illumina sequencing data. In total, 13 and 16 unigenes showing upregulation in ovaries and testes, respectively, were selected to confirm their altered expression levels using the same RNA samples as used for transcriptome sequencing. The RT-qPCR primers used for amplification are shown in
Table S3. Gene expression level was initially examined by semi-RT-qPCR, with the PCR products being analyzed electrophoretically using 1.0% agarose gels. We subsequently conducted RT-qPCR to verify the expression profiles in the ovary and testis. Reaction mixtures consisted of 5.25 µL of ddH
2O, 7.5 µL of 2× SYBR Green MasterMix (Bio-Rad Laboratories, Hercules, California, USA), 4 μM of each specific primer, and 1.0 µL of first-strand cDNA template. The RT-qPCR program included an initial denaturation for 3 min at 95°C, followed by 40 cycles of denaturation at 95 °C for 10 s, annealing for 30 s at 55 °C, and extension for 30 s at 72 °C. The RT-qPCR reactions were run in triplicate in 96-well format Microseal PCR plates (Bio-Rad Laboratories, Hercules, California, USA) and carried out in a CFX96 Touch™ Real-Time PCR Detection System (Bio-Rad Laboratories, Hercules, California, USA), with the
RPS18 gene being used as internal control [
46]. RT-qPCR was performed using SYBR
® Premix Ex Taq™ (Tli RNaseH Plus) (TaKaRa, Dalian, China) in 15-µL reaction mixture containing 5 ng cDNA as the template.
4.7. Dietary RNAi Toxicity Assay
We determined the role two ovary-specific genes with the assembly IDs TRINITY_DN37464_c0_g3 (REST corepressor 2-like,
RC2-like) and TRINITY_DN34758_c0_g1 (protein snail homolog Sna-like,
PSHS-like) (
Tables S2 and S3). Specific dsRNA primers containing a T7 promoter sequence at the 5ʹ end targeting these two genes were designed using E-RNAi [
50], the sequences of which are listed in
Table S4. After amplification from the cDNA samples of
H. vigintioctopunctata adults and confirmation by sequencing, the amplicons (394 bp for dsRC2-like and 398 bp for dsPSHS-like) were used as templates for in vitro transcription reactions to generate dsRNAs using a T7 MEGAscript kit (Ambion, Austin, Texas, USA).
For the phenotypic investigations, the sex of the newly emerged adults (<12 h) was determined under a stereoscope, and each individual was thereafter reared independently in a petri dish. The adult males were provided with normal leaves, whereas the adult females were provided with leaf discs treated with dsRNAs for five consecutive days. The leaf discs (12 mm diameter) were pre-treated by immersing in one of three 200 ng/µL solutions of dsRNA, namely, dsGFP control, dsRC2-like, or dsPSHS-like. On the fifth day (the first day of pairing), one untreated male and one treated female were paired as a single replicate in individual petri dishes and provided with fresh untreated leaves. As the dsGFP, dsRC2-like, and dsPSHS-like treatment groups, we used 12, 19, and 19 pairs of respectively treated adults. Over the subsequent 14 days, we recorded adult female mortality, the percentage of sterile females, pre-oviposition period, fecundity, number of eggs laid per day, and hatching rate. The ovaries of unmated females, and mated females treated with dsGFP, dsRC2-like, and dsPSHS-like were dissected to evaluate the effects of dsRNA ingestion on ovarian development.
For the evaluation gene expression, adult females were treated with 200 ng/µL dsGFP, dsRC2-like, or dsPSHS-like dsRNA as described above. After 5 days, the ovary was dissected from each adult female. All samples were flash-frozen in liquid nitrogen and stored at −80 °C in 1.5-mL centrifuge tubes prior to total RNA extraction. Each sample was represented by five biological replicates. The reference gene
RPS18 was used as the internal control [
46]. Relative quantification of
RC2-like and
PSHS-like in the various samples was performed using the 2
−ΔΔCt method [
51].
4.8. Data Analysis
Student’s t-test was used to determine the RC2-like and PSHS-like expression levels under different treatments. For each comparison, p < 0.05 was considered to indicate a significant difference. One-way ANOVA was used to compare data relating to the pre-oviposition period, fecundity, and hatching rate under different treatments. Means were compared using Tukey’s test with the significance set at p < 0.05. Proportional data were arcsine square root-transformed prior to analysis. Mortality rates and percentages of sterile females were analyzed using Chi-squared tests. SPSS v. 21.0 (IBM Corp., Armonk, NY, USA) was used for all statistical analyses.