Role of Groucho and Groucho1-like in Regulating Metamorphosis and Ovary Development in Nilaparvata lugens (Stål)

Juvenile hormone and ecdysone are key regulators in the metamorphosis and development. Grocho (Gro) is a highly conserved protein required for metamorphosis and development. Brown planthopper (Nilaparvata lugens) is a major pest affecting rice production in China and many Asian countries. Although the molecular function of Gro has been investigated in holometabolous insects such as Aedes aegypti and Drosophila melanogaster, their role in the hemimetabolous insect, brown planthopper, and the relationship between NlGro/NlGro1-L and JH/ecdysone signaling pathway, remained unknown. In this study, NlGroucho (NlGro) and NlGroucho1-like (NlGro1-L) were cloned. An analysis of the predicted protein sequence showed that NlGro has highly conserved Q domain and WD40 domain, and NlGro1-L has a highly conserved WD40 domain. The expression profiles of both genes were studied by quantitative real-time PCR (qRT-PCR). Their relative expressions were high in egg, head, wing, ovary, and testis. NlGro and NlGro1-L were found to interact genetically with juvenile hormone and ecdysone signaling by hormone treatment and RNAi of JH/ecdysone signaling-related genes. Moreover, when NlGro or NlGro1-L was down-regulated alone, the survival rate was decreased, the ovarian development was delayed, and the oviposition was also affected. All defects were aggravated when NlGro and NlGro1-L were down-regulated together. This study will help to develop new pesticides on the basis of the function of NlGro and NlGro1-L, and provide new possibilities for the control of Nilaparvata lugens.


Introduction
Insects must undergo a process of molting and metamorphosis during their growth and development. The main regulators of insect metamorphosis are juvenile hormone (JH) and ecdysone (or 20-hydroxy-ecdysone, ED) [1,2]. Juvenile Hormone (JH) is a sesquiterpene compound and was first discovered by Sir Wigglesworth in the Rhodnius prolixus [3]. Juvenile hormone controls the basic physiological functions of insect development, reproduction, pheromone production, and antennae formation [4]. Increasing or decreasing the level of juvenile hormone will affect ovarian development [1,2]. Ecdysone is a sterol compound that is synthesized by insect prothoracic gland and secreted into the hemolymph [5]. Both JH and ED are key regulatory hormones in the metamorphosis and development [6][7][8]. Moreover, juvenile hormone signaling regulates ovarian development and maturation, partly through the interaction with ecdysone signaling [6][7][8][9]. Juvenile hormone also regulates the expression of vitellogenin in fat body and hemolymph [1,2].
Thus far, the Groucho (Gro) gene family is only found in metazoans, and is named after the first identified Drosophila mutant phenotype [10]. Drosophila Gro is widely expressed throughout the development [11]. The homologous gene Transducin-Like Enhancer of split (TLE) in medaka fish is mainly expressed during embryo and ear development [12].
The human Gro homologue has high structural similarity with β-transducin and is close to Drosophila Gro (a member of the Enhancer of split (E(spl)) complex), so it is named the TLE protein [13][14][15]. The homologous Gro-related gene Grg protein was also isolated from mice [16]. The homologue of Caenorhabditis elegans is named Unc-37 (ESG) protein and its 612 amino acid sequence protein is the smallest known Gro protein [16,17]. The Drosophila embryonic development requires maternal Gro protein, and post-embryonic development depends on the expression of zygotic Gro [11]. The primary structure of Gro has five domains, which are defined by its evolutionary conservation, namely Q, GP, CcN, SP, and WD. The most highly conserved and strictly characterized domains of the Gro family are the amino terminal Q domain and the carboxy terminal WD repeat domain [16]. The other three domains, GP, CcN, and SP, are the middle variable domains [16]. Despite the different nomenclature in different species, Gro still shows high conservation, especially in the carboxy terminal WD domain, most of which have at least 80% amino acid identity [16]. Gro plays an important role in embryonic development, neurogenesis, and sex determination [17][18][19][20].
The Drosophila Gro interacts genetically with the E(spl) complex [21]. Gro does not directly bind to DNA, but it can bind to transcription factors to regulate transcription [21]. Both Drosophila Gro and human TLE1 proteins have been shown to physically interact with Hairy (human homologue HES) [18]. The Trp-Arg-Pro-Trp (WRPW) motif found at the carboxyl terminal of Hairy is necessary and sufficient for interaction with Gro [18]. Furthermore, this interaction between the WRPW motif and Gro is necessary for transcriptional repression [16,22]. The Drosophila Gro also interacts with Hairy and Deadpan regulatory proteins in segmentation and sex determination [23,24]. Aedes aegypti Hairy is a downstream gene of Met, and Gro directly interacts with Hairy's conserved C-terminal tetrapeptide WPRW to regulate the signal transduction of JH-Met [25]. There are short Grg/TLE family members in both humans and mice; these short members only include a Q domain responsible for the dimerization of Grg/TLEs and a glycine-proline-rich GP domain to repress transcription [16,22,26].
The brown planthopper (Nilaparvata lugens Stål) is a monophagous pest that can only feed and reproduce on rice and common wild rice. The brown planthopper has a migratory habit and is a major threat to rice production in China and many other Asian countries. Although the molecular function of Gro has been investigated in holometabolous insects, such as Aedes aegypti and Drosophila melanogaster, their role in the hemimetabolous insect, brown planthopper, and the relationship between NlGro/NlGro1-L and the JH/ecdysone signaling pathway, especially key players of the JH/ecdysone signaling pathway, such as E93, remain unknown. In this study, NlGro and NlGro1-L were cloned, and their expression at different developmental stages and in different tissues was measured; the relationship between NlGro/NlGro1-L and JH/ecdysone signaling pathway was also investigated through survival rate, development, and reproduction.

Insect Culture
The experimental insects were reared in our laboratory with rice of the IIyou023 variety. The rice was cultured with nutrient solution at 25 • C. The relative humidity is around 65%, and the photoperiod is 14L:10D (h). The nutrient solution used was changed daily. The nutrient solution contained NH 4

RNA Extraction, cDNA Synthesis, and qRT-PCR
Brown planthopper individuals of every stage of development were collected over a 3-day period for total RNA extraction, including 40 first instar nymphs, 20-30 of each of second, third, fourth, and fifth instar nymphs, 15 brachypterous 7-day old adults, and 15 macropterous 7-day old adults. Forty embryos were collected on each day of the 9-day embryonic development stage. Three biological replicates were used. Total RNA was extracted with RNAiso Plus (TaKaRa, Dalian, China). The precipitate was dissolved in DEPC-treated water, and then the concentration and purity were measured with a microspectrophotometer (Allsheng, Hangzhou, China). The Transcriptor First Strand cDNA Synthesis Kit (Roche, Shanghai, China) was used for synthesis of cDNA. A 20 µL reaction was used: 1 µg total RNA, 1 µL Oligo dT primer, water added to 13 µL, and incubated in a water bath at 65 • C for 10 min, then 4 µL buffer was added, 2 µL dNTP, 0.5 µL inhibitor, and 0.5 µL Transcriptase, in a water bath at 55 • C for 30 min, finally transferred to a water bath at 85 • C for 5 min.
Templates for quantitative real-time PCR (qRT-PCR) were prepared by collecting insects at different embryonic development stages: females laid eggs on rice seedlings after mating, and 1, 2, 3, 4, 5, 6, 7, 8, and 9 days after the eggs hatched. The insects were also collected at different developmental stages: egg, first, second, third, fourth, and fifth instar nymph, as well as short-winged females, long-winged females, short-winged males, and long-winged males. Templates for the qRT-PCR of different tissues: head, thorax, forewing, leg, mid-gut, ovary, or testis. The mid-gut is a tube with the same thickness. The mid-gut was dissected based on its relative position between the fore-gut and the malpighian tube/hind-gut. To further study the correlation between NlGro/NlGro1-L and JH/ED signaling, we also measured the expression of NlGro and NlGro1-L after downregulating hormone-related genes (NlMet, NlβFtz-f1, NlTai, NlKr-h1, NlE93, NlECR, NlUspA, NlUspB, and NlBr) by RNAi.
Primer3 was used to design primers for qRT-PCR. The primer sequences were listed in Table S1. Genes tested by qRT-PCR: NlGro, NlGro1-L, NlMet, NlβFtz-f1, NlTai, NlKr-h1, NlE93, NlECR, NlUspA, NlUspB, NlBr, and NlVg. For measuring the transcriptional response to JH/ED, NlActin was used as a reference gene. For the rest of the experiments, NlRP15 was used as a reference gene. The reference genes were selected based on a previous study [28]. Moreover, we carried out a qRT-PCR experiment to compare the stability of the reference genes (NlRP15 and NlActin), by which we selected a relatively stable reference gene for different experiments ( Figure S1). The primers of qRT-PCR were shown in Figure S2. qRT-PCR was performed using the YESEN Hieff qRT-PCR SYBR Green Master Mix kit (actin was used as the reference gene). The reaction system was SYBR Green Master Mix (1×) 10 µL, cDNA template 2 µL, upstream and downstream primers 0.4 µL each (10 µM), and up to 20 µL with DEPC water. Three biological replicates were used for each sample. ABI StepOnePlus quantitative real-time thermocycler was used. Two-step amplification was used. The first step (pre-denaturation): 95 • C, 2 min, and one cycle; the second step (PCR reaction): first 95 • C 10 s and then 60 • C, 30 s, 40 cycles. The data were analyzed with the 2 −∆∆Ct method [29].

Cloning
The template used to clone NlGro and NlGro1-L was prepared from nymphs. cDNA was synthesized using Transcriptor First Strand cDNA Synthesis Kit (Roche, Shanghai, China). Primers were designed using Primer3 (http://primer3.ut.ee/, accessed on 15 September 2017) based on the sequences downloaded from the NCBI (NlGro: 111053289; NlGro1-L: 111047810). The primer sequences were shown in Table S2. The cloning vector pMD18-T (TaKaRa) was used, and the plasmid was sent for sequencing after cloning (Sangon Biotech Co. Ltd., Shanghai, China).

JHIII and Ecdysone Treatment
Juvenile hormone III(JH III) and Hydroxyecdysone (ED) were purchased from Sigma (Sigma-Aldrich, Shanghai, China). Brown planthoppers at the fifth-instar nymph stage (0-12 h) were collected and treated with 0.2 µL JH (1 µg/µL) or 0.2 µL ED (0.05/0.1 µg/µL). The same amount of acetone was used as the control. A 5 µL micro-injection needle (Hamilton 700, Hamilton Laboratory Equipment Co., LTD, Shanghai, China) was used to topically apply 1 µL of the prepared hormone solution or acetone (control) on the back of the fifth instar nymph. At least 50 nymphs were treated for each sample. After the recovery of the treatment, the insects were cultured in a glass tube containing rice seedlings. Samples of 12 nymphs (three replicates in total) were collected and total RNA was isolated. After 6 h, 12 h, and 24 h, samples were taken, total RNA was extracted, and cDNA was synthesized for qRT-PCR.

dsRNA Synthesis and RNAi
A PCR product with a size about 500 bp was used to synthesize double-stranded RNA (dsRNA). The primers were shown in Table S3. Region of dsRNA and qRT-PCR primers was shown in Figure S2. GFP dsRNA was used as a control. dsRNA was synthesized according to the instructions of the T7 RiboMAX Express RNAi System kit (Promega, Madison, WI, USA), and frozen at −20 • C for later use.
The nymphs were injected with 0.2 µg dsRNA of the corresponding gene using the Narishige Injection System (MN-151, Narishige, Japan). The control group was injected with the same amount of GFP dsRNA. After injection, the brown planthoppers were allowed to recover for 2 h and then cultured on rice seedlings. The brown planthoppers were cultured in an incubator in groups of 10. After RNA extraction, cDNA was synthesized and qRT-PCR was used to measure the expression level of genes. The results were analyzed using the 2 −∆∆Ct relative expression method [29]. In addition, the survival rate and molting rate were calculated.

Ovary Dissection, Grading, and Oviposition
The nymphs of Nilaparvata lugens were dissected 3 days after emergence, and used for ovarian grading. The ovaries of females were dissected and graded. The grading standard can be referred to our previous report [6]. The dissected ovaries were fixed with 4% paraformaldehyde (Sangon) for 20 min. Then, we rinsed them with PBS three times, removed the fixed ovaries, placed them on a glass slide, added an appropriate amount of glycerin, and took photos with a microscope (eclipse 80i, Nikon, Japan).
The fifth instar nymphs injected with dsRNA were cultured. Males were added to mate with the females on the day of emergence. Each treatment guarantees 20 individuals. Three biological replicates were used. We recorded the time of the eggs laid and the number of eggs each day for calculating the pre-oviposition period, as well as the total number of eggs laid until the female died.

Western Blot
The fifth instar nymphs injected with dsRNA were dissected and sampled 3 days after emergence, while the sample was ground with a grinding pestle on ice and centrifuged to collect the supernatant. The protein samples were subjected to SDS-PAGE electrophoresis (Sangon), and transferred to PVDF membrane (ThermoFisher Scientific, Shanghai, China) after 150 min. We added an appropriate amount of 5% non-fat milk and blocked it slowly in a shaker. Then, it washed with 1X PBST, primary antibody anti-Vg and ACTB antibody was added (Rabbit anti-Vg antibody is a gift from Zhou Qiang of Sun Yat-Sen University, 1:10,000 dilution; Mouse anti-ACTB antibody, purchased from Sangon, 1:1000 dilution), and then incubated for 1 h. We washed it with 1X PBST and incubated it with HRPlabeled secondary antibody (Goat anti-rabbit or horse anti-mouse IgG-HRP, Cell Signalin Technology, Shanghai, China, 1:5000 dilution), and then washed it with 1X PBST.

Sequence Analysis of NlGro and NlGro1-L
NlGro and NlGro1-L were cloned and the amino acid sequences of NlGro and NlGro1-L were predicted and compared with the homologs of Gro family proteins from other species. The Gro family protein contains five conserved functional domains, among which the Q domain at the amino terminal and the WD40 domain at the carboxy terminal are highly conserved. The Q domain can not only inhibit basic transcription, but also activate transcrip-tion and act as a dimerization domain. The WD40 domain can be used for the interaction with other proteins. NlGro also contains highly conserved Q domain and WD40 domain (Figure 1). Although NlGro1-L does not contain Q domain, it has a WD40 domain (Figure 1). The phylogenetic analysis suggests that the brown planthopper is more closely related to Frankliniella occidentalis, Onthophagus taurus, Drosophila melanogaster, Bemisia tabaci, Riptortus pedestris, and Halyomorpha halys ( Figure S3). The blue dashed box represents the conserved Q domain, and the red dashed box represents the highly conserved WD40 domain.

Expression Profiles of NlGro and NlGro1-L
The expression levels of NlGro and NlGro1-L at different developmental stages and in different tissues were detected by qRT-PCR. The results showed that the expression profiles of NlGro and NlGro1-L in different embryonic stages were similar (Figure 2A,B). The expression of NlGro and NlGro1-L tends to decrease from different embryonic stages to different nymph stages (Figure 2A-D). The expression profiles of NlGro and NlGro1-L at different developmental stages are similar ( Figure 2C,D). The expression of NlGro and NlGro1-L in the long-winged adults is higher than that in the short wings ( Figure 2C,D). The expression profiles of NlGro and NlGro1-L in different tissues are also similar ( Figure 2E,F). In summary, the expressions of NlGro and NlGro1-L are relative higher in the heads, wings, and ovaries of the females. The expressions in ovary and testis are higher, indicating that NlGro and NlGro1-L may play a role in reproduction. One-way analysis of variance (Duncan's method) was used. For the letters a-h, the same letter indicates that there is no significant difference, and different letters indicate a significant difference between the two; the farther the letter is, the greater the difference between the two, p < 0.05.

Transcriptional Responses to JHIII and Ecdysone
In order to study the response of NlGro and NlGro1-L to juvenile hormone and the ecdysone signaling pathway, we used juvenile hormone and ecdysone to treat Nilaparvata lugens nymphs, and measured the changes in expression of NlGro and NlGro1-L by qRT-PCR. Furthermore, the expression of NlGro and NlGro1-L was measured after down-regulated the expression of juvenile hormone-related genes: NlMet, NlβFtz-f1, NlTai, and NlKr-h1, and ecdysone-related genes: NlE93, NlECR, NlUspA, NlUspB, and NlBr.
When the fifth instar nymphs were treated with JHIII, the expression of NlGro and NlGro1-L was significantly reduced 6 h/12 h after treatment ( Figure 3A). However, the expression of NlGro and NlGro1-L increased significantly 24 h after treatment ( Figure 3A), suggesting JHIII treatment has a significant effect on the expression of NlGro and NlGro1-L. When the fifth instar nymphs were treated with ecdysone (ED), the expression of NlGro and NlGro1-L was significantly reduced 6 h after treatment ( Figure 3B), while the expression of NlGro and NlGro1-L varied 12 h/24 h after treatment ( Figure 3B). Again, ED treatment caused a significant change in the relative expression of NlGro and NlGro1-L.

Effects of Down-Regulating Hormone-Related Genes
We used RNAi to down-regulate NlE93, NlECR, NlUspA, NlUspB, NlBr, NlMet, NlβFtz-f1, NlTai, and NlKr-h1 in the fourth instar nymphs, and collected samples 3 days after dsRNA injection. Then, the expression of NlGro and NlGro1-L was measured (actin was used as the reference gene). After down-regulating the expression of hormone-related genes in the fourth instar Nilaparvata lugens by RNAi, the changes in the expression of NlGro and NlGro1-L were similar ( Figure 3C,D). After down-regulating NlE93 and NlECR, the expression of NlGro and NlGro1-L increased. Surprisingly, after down-regulating NlE93 alone, the expression of NlGro and NlGro1-L increased significantly, up to hundreds to thousands of times ( Figure 3C,D). After down-regulating NlβFtz-f1, NlTai or NlKr-h1, the expression of NlGro and NlGro1-L decreased. After down-regulating NlKr-h1 alone, the expression of NlGro and NlGro1-L decreased to 75%. After down-regulating NlTai alone, the expression of NlGro and NlGro1-L decreased by 50%. After down-regulating NlβFtz-f1 alone, the expression of NlGro and NlGro1-L decreased to about 15% ( Figure 3C,D). In summary, the expression of NlGro and NlGro1-L was more sensitive to the down-regulation of ecdysone signaling-related genes than that of juvenile hormone-related genes.

The Emergence Was Disrupted after Down-Regulation of NlGro and NlGro1-L
In order to understand the functions of NlGro and NlGro1-L during the growth, development, and reproduction, we used dsRNA to down-regulate the expression of NlGro and NlGro1-L in the nymphs, and studied molting, survival, ovarian development, and reproduction. After the injection of NlGro dsRNA and NlGro1-L dsRNA, we found the nymphs failed to emerge as adults and died during the molting process ( Figure 4A-C). The emergence rate of the control is 93%, which decreased to about 80% after down-regulating NlGro or NlGro1-L alone. After down-regulating NlGro and NlGro1-L in combination, the emergence rate decreased to lower than 65% ( Figure 4D,G).
The expression of NlGro and NlGro1-L increased dramatically after down-regulating NlE93 ( Figure 3G). Thus, we speculated that NlE93 may interact with NlGro and NlGro1-L. When NlGro dsRNA was injected into the fourth instar nymphs, 12.87% of the fourth instar nymphs failed to molt ( Figure 4H). Moreover, the emergence rate of the fifth instar nymphs (69.75%, Figure 4H) was lower than that of the control (dsGFP, 87.13%, Figure 4H). When NlE93 was down-regulated alone, 21.73% of the fourth instar nymphs failed to molt, 39.81% of the fifth instar nymphs died, 38.46% of the fifth instar nymphs failed to molt, and no adults emerged ( Figure 4H). When NlGro and NlE93 were down-regulated together, no adults emerged ( Figure 4D-F,H).
When dsRNA was injected into the fifth instar nymphs and NlGro was down-regulated alone, the mortality rate was 9.26%, the molting failure rate was 18.75%, and 71.99% of the nymphs molted to adults ( Figure 4E,F,I). When NlE93 was down-regulated, the mortality rate was 13.1%, 19.35% of the adult failed to molt, and only 67.55% succeeded in eclosion ( Figure 4I). After down-regulating NlGro and NlE93 together, the mortality rate and the rate of molting failure decreased. The emergence rate rose to 83.89%. This is reminiscent of the dramatic increase of NlGro and NlGro1-L after the down-regulation of NlE93 (Figure 3G), suggesting a compensation effect by down-regulation of NlE93 ( Figure 4I).
In addition, we used RNAi to down-regulate NlGro or NlGro1-L in the third, fourth, and fifth instar nymphs, collected samples 3 days after dsRNA injection, and detected RNAi efficiency by qRT-PCR. We found that the expression of each gene was significantly reduced, indicating the RNAi is successful (Figures S4 and S5).

Ovarian Grading and Oviposition after Down-Regulation of NlGro and NlGro1-L
Three days after the emergence of females, the ovaries were dissected. The ovary of the females injected with dsGFP developed into grades II, III, or IV ( Figure 5). The other three treatments developed into grade I, II, III, or IV ovaries ( Figure 5). The representative images of each treatment are shown in Figure 5. After NlGro and NlGro1-L were downregulated alone or in combination, the pre-oviposition period of females was prolonged to 4 days, which is 1 day more than the control (dsGFP) ( Figure 6A). Moreover, the ovarian development was delayed ( Figure 6B). Compared with the control group, the number of ovaries that developed into grade IV was reduced, while the number of ovaries retained in grade I was increased, indicating that ovarian development was significantly delayed after NlGro or NlGro1-L were down-regulated ( Figure 6B). After down-regulating dsNlGro and dsNlGro1-L together, the ovarian development was significantly delayed, 50% of the ovaries remained at grade I, and the number of ovaries that have developed into grades III and IV was significantly reduced ( Figure 6B). Furthermore, after NlGro or NlGro1-L was down-regulated, the number of eggs laid was significantly reduced ( Figure 6C). When NlGro and NlGro1-L were down-regulated together, the number of eggs laid also reduced significantly (85) compared with that of the control (183, Figure 6C).  was used. The letters a and b indicate whether there is a statistical difference; the same letter represents no difference, while different letters represent significant differences, p < 0.05. (B) Ovarian grading after down-regulating NlGro and NlGro1-L; independent sample t test was used (Student's t test), n.s.: no significant difference, * p < 0.05, ** p < 0.01, *** p < 0.001. (C) The total number of eggs reduced after down-regulating NlGro and NlGro1-L; Duncan's multiple comparison was used. The letters a and b indicate whether there is a statistical difference; the same letter represents no difference, while different letters represent significant differences, p < 0.05. (D) Relative expression of Vg after the down-regulation of NlGro and NlGro1-L; Duncan's multiple comparison was used. The letters a and b indicate whether there is a statistical difference; the same letter represents no difference, while different letters represent significant differences, p < 0.05. (E,F) Western blot detection of Vg protein after down-regulation of NlGro and NlGro1-L: (E) Abdomen, (F) Ovary.

Gro and Gro1-L Regulate Vg Expression
The expression Vg was measured by qRT-RCR after injecting dsGFP, dsNlGro, and dsNlGro1-L (alone or in combination). The results showed that after down-regulating NlGro or NlGro1-L, the expression of Vg decreased but not significantly at a transcriptional level ( Figure 6D). When both NlGro and NlGro1-L were down-regulated, the expression of Vg decreased significantly ( Figure 6D).
To understand the change of Vg expression at a protein level, the abdomen (without ovary) and ovary were collected. Vg was detected by anti-Vg through western blot ( Figure 6E,F). The Vg expression in the abdomen was not significantly decreased compared with the control, when NlGro or NlGro1-L was down-regulated, respectively ( Figure 6E,F). However, Vg decreased significantly in the abdomen when NlGro and NlGro1-L were down-regulated together ( Figure 6E,F).

Discussion
This study found that the expression profiles of NlGro and NlGro1-L are similar, and they are expressed in all tissues at all developmental stages (Figure 2). NlGro and NlGro1-L are highly expressed in the head, wings, ovary, and testis, indicating that they might also be directly involved in the development of ovaries and testes, thereby affecting the reproduction. The high expression level indicates that NlGro and NlGro1-L play an important role.
Juvenile hormone and ecdysone interact with each other to regulate molting and reproduction [4]. JHIII can regulate changes in the expression of NlGro and NlGro1-L. After NlβFtz-f1, NlTai, and NlKr-h1 were down-regulated, the expression of NlGro and NlGro1-L was decreased ( Figure 3C,D), especially when the expression of NlβFtz-f1 was downregulated, while the expression of NlGro and NlGro1-L decreased to 15% ( Figure 3C,D). Therefore, we speculate that the expression of NlGro and NlGro1-L is related to NlβFtz-f1. βFtz-f1 is highly expressed in the molting stage of the nymphs, and it also plays a role in the formation and development of the larval cuticle [30]. βFtz-f1 regulates the formation of yolk and affects oviposition in Aedes aegypti [31]. It also regulates the reproduction in Tribolium castaneum [32]. We speculate that NlGro and NlGro1-L may affect brown planthopper molting and ovarian development through the interaction with βFtz-f1. Treatment with ecdysone (20E) or down-regulation of its receptor NlECR or key downstream transcription factor NlE93, the expression of NlGro and NlGro1-L was significantly changed (Figure 4), suggesting NlGro and NlGro1-L can participate in the growth and development at different stages. The expression of NlGro and NlGro1-L was significantly reduced after ecdysone treatment (Figure 3), suggesting the negative regulation of NlGro and NlGro1-L by ecdysone and its receptor. The expression of NlE93 increased significantly after down-regulating NlGro and NlGro1-L (Figure 3), suggesting the negative regulation of NlE93 by NlGro and NlGro1-L (Figure 7). It is worth noting that NlGro has both a highly conserved Q domain and WD40 domain, while NlGro1-L only has WD40 domain ( Figure 1). Down-regulation of NlGro and NlGro1-L together disrupted the growth and development ( Figure 5), delayed the ovarian development, and reduced the number of eggs laid (Figure 6), indicating that NlGro and NlGro1-L might have a synergistic effect. Moreover, a previous study has shown that Hairy and Groucho mediate the action of juvenile hormone receptor Methoprene-tolerant in gene repression [25]. We surmise Gro and Gro1-L mediate the interaction between ecdysone signaling and JH signaling (Figure 7). Vitellogenin (Vg) is the precursor of vitellin. After the fat body synthesizes Vg protein, which is transported to the hemocyte and is taken up by the developing oocytes through receptor-mediated endocytosis, it then undergoes a series of modifications. After modification, it is deposited as yolk protein, which provides nutrition for the development of the embryo [33]. Hormone and nutritional pathways can regulate the synthesis of Vg and the uptake by oocytes, which is required for the successful reproduction of females [34,35]. In Rhodnius prolixus, if the expression of Vg is down-regulated by RNAi, egg production would be inhibited, because the nutrient transport process in Rhodnius prolixus is compromised [36]. In the cotton boll weevil, Anthonomus grandis, when Vg expression is down-regulated, the egg production would be reduced [37]. After the fifth instar nymphs were injected with dsNlGro and dsNlGro1-L alone or in combination, both qRT-PCR and western blot showed that the expression of Vg was reduced, and it was more dramatic when two genes were down-regulated together, indicating that the delayed development of the ovary and the decrease in egg number may be a direct result of the down-regulation of Vg.
In conclusion, this study will help us to further understand the functions of NlGro and NlGro1-L, and their molecular mechanism of action, and provides a basis for the development of more direct and effective pest management solutions in the future.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/ijms23031197/s1. Author Contributions: X.L. conceived and designed the study, H.G., X.J., S.Z., Y.L. and X.L. performed experiment and analyzed data, X.L., X.J. and H.G. wrote the paper. All authors have read and agreed to the published version of the manuscript.