The role of ecdysteroid in the regulation of ovarian growth and oocyte maturation in Rhodnius prolixus, a vector of Chagas disease

Ecdysteroid is the main insect steroid hormone that has essential roles in development, molting and metamorphosis. The primary source of ecdysteroids in the larval stages of insects is the prothoracic gland (PG). Ecdysone (E) is produced in the PG and converted in target tissues to 20-hydroxyecdysone (20E), which is known as the ‘molting’ hormone (Rewitz et al., 2006a; Warren et al., 2006). Ecdysone, the precursor of 20E, is primarily synthesized from dietary cholesterol or sterols in insect PGs, since insects cannot synthesize these precursor cholesterols de novo. A set of enzymes involved in 20E production has been reported in Drosophila melanogaster and crustaceans (Niwa and Niwa, 2014; Sumiya et al., 2014). These enzymes, encoded by so-called ‘Halloween’ genes, are termed spook (spo; CYP307A1), phantom (phm; CYP306A1), disembodied (dib; CYP302A1), shadow (sad; CYP315A1) and shade (shd; CYP314A1) (Luan et al., 2013; Rewitz et al., 2006a). Most of the biosynthetic conversion steps for ecdysteroids involve these Halloween genes, which have the characteristic signature domains that place them within the cytochrome P450 (CYP450) hydroxylases family of enzymes (see Dermauw et al., 2020). However, the first step in ecdysteroid biosynthesis involves neverland (nvd), which encodes an oxygenase-like protein with a Rieske electron carrier domain and is specifically expressed in tissues that synthesize ecdysone, such as the PG. Under the action of 7,8-dehydrogenase, dietary cholesterol is first transformed to 7-dehydrocholesterol (7dC) by nvd. Several steps are then involved in the conversion of 7dC to 5β-diketol by enzymes encoded in non-molting glossy (nmg)/shroud (sro) and spo/spookier (spok), and this part of the pathway is called the black box. 5β-diketol, through sequential hydroxylations, is converted into the secreted steroid E. Then it is converted into the active 20E by shd (Fig. S1). 20E typically binds to a receptor containing the ecdysone receptor (EcR) and the ultra spiracle protein (USP), which together form a heterodimer complex (Niwa et al., 2010; Gilbert et al., 2002; Fahrbach et al., 2012). In addition, however, EcR and USP can possibly work independently since it has been shown that some EcR-mediated events do not require USP (Costantino et al., 2008; Cheng et al., 2018; Rinehart et al., 2001).

In adult insects, the PGs disappear, but ecdysteroids are still present at significant levels (Raikhel et al., 2005; Brown et al., 2009), indicating that there are different sites for ecdysteroid production and different physiological roles in adults (Hagedorn et al., 1975; Hodgetts et al., 1977). In many insect species, the adult ovaries are a main source of ecdysteroids, which play crucial roles in female reproduction and embryogenesis (Hartfelder et al., 2002; Geva et al., 2005; Verlinden et al., 2009). The role of ecdysteroids in reproduction in R. prolixus has, however, not been studied in any detail. In holometabolous insects such as Bombyx mori, Aedes aegypti, D. melanogaster and Tribolium castaneum ecdysteroids are known for their roles in controlling ovarian maturation and oogenesis (Swevers and Iatrou, 2003; Sun et al., 2002; Richard et al., 1998; Parthasarathy et al., 2010).

Rhodnius prolixus is a reduviid hemipteran that depends entirely on vertebrate blood for both its larval and adult stages. In the unfed state, R. prolixus is in a state of arrested development and can survive for several months without a blood meal (Buxton, 1930). A blood meal initiates growth and development and is also required for each cycle of egg development in adult females (Lange et al., 2022). The first report of a function for ecdysteroid in R. prolixus is an involvement in ovulation and egg laying, whereby hemolymph ecdysteroid, along with a mating factor, feed back onto the brain to initiate the release of a myotropin from medial neurosecretory cells. The myotropin increases muscle contraction of the ovaries, resulting in ovulation and egg laying (Kriger and Davey, 1984; Ruegg et al., 1981; Lange et al., 2022). In female R. prolixus, hemolymph ecdysteroid titers reach a peak in both mated and virgin females at day 5 post blood meal (PBM) and ovariectomy reduces these titers (Ruegg et al., 1981; Cardinal-Aucoin et al., 2013). However, the role of ecdysteroids in egg development has not been studied in R. prolixus. In this paper, we have identified and characterized the Halloween genes, nvd, the EcR and USP in R. prolixus and show the essential role of ecdysteroids in egg production. We hypothesized that the ovary expresses transcripts for enzymes involved in the biosynthesis of 20E and for the ecdysone receptor, and that 20E may have a direct or indirect influence on egg production in female R. prolixus. We predicted that knockdown of the ecdysone receptor or shade would cause critical defects in oocyte maturation and egg production.

Adult R. prolixus were obtained from a long-established colony at the University of Toronto, Mississauga. Insects were bred in incubators at 25°C under high humidity (∼50%). Males and females were separated during the final instar (5th instar) and at 10 days post-ecdysis (PE) into adults, were fed on defibrinated rabbit blood (Cedarlane Laboratories Inc., Burlington, ON, Canada) through an artificial feeding membrane. Since as stated earlier, unfed R. prolixus remain in a state of arrested development, 10 days PE is a stable condition from which to monitor the effects of a blood meal. Only adults that fed 2.5–3 times their initial weight (typical blood meal size for adults) were used for experimentation. Fed adult females were separated individually and paired with recently fed males for 2 days to allow for mating. Experimental insects were kept in incubators in a 12 h light:12 h dark regime at 28°C with high humidity (∼50%). To verify mating, the cubicle was examined for the ejected spermatophore. The ovarioles from fed insects were separated according to Brito et al. (2018): pre-vitellogenic ovarioles (Ov_PV) include the tropharium and immature oocytes, and vitellogenic ovarioles (Ov_V) contain follicles with vitellogenic and/or mature oocytes. Detailed diagrams showing the reproductive system and its component parts can be found in Lange et al. (2022).

VectorBase was interrogated to extract R. prolixus Halloween genes (spo, phm, dib, sad, shd), the non-Halloween gene (nvd) and the ecdysone receptor (EcR and USP) contig sequences. The amino acid sequences were deduced using the ExPASy2 tool (www.expasy.org). We used a custom tBLASTn web portal (https://blast.ncbi.nlm.nih.gov/Blast.cgi) to run BLAST to confirm the predicted sequence with each gene obtained from VectorBase. Based on the top contig hits and sequence identity of e-value and amino acid sequence identity, we confirmed the contig matching at the nucleotide level with NCBI. After confirming the extracted contig of the genes, we examined for their transcript expression in an ovary transcriptome (Leyria et al., 2020).

To identify the evolutionary relationship between R. prolixus sequences and sequences from other arthropods, phylogenetic trees were constructed using MRGAX software (https://www.megasoftware.net) and the maximum-likelihood method with the JTT matrix-based model (Jones et al., 1992). Initial trees for the heuristic search were obtained automatically by applying Neighbor-join and BioNJ algorithms to a matrix of pairwise distances estimated using the JTT model, and then selecting the topology with superior log-likelihood value. Evolutionary analyses were conducted in MEGA X (Kumar et al., 2018).

Transcript expression was quantified in a variety of tissues from unfed adult females in cold autoclaved phosphate buffered saline (PBS: 6.6 mmol l⁻¹ Na₂HPO₄/KH₂PO₄, 150 mmol l⁻¹ NaCl, pH 7.4). Trizol and chloroform were used to isolate RNA and purified RNA was quantified using a spectrophotometer DS-11+ (DeNovix Inc., Wilmington, DE, USA). Electrophoresis in a 1% agarose gel was used to monitor RNA integrity as observed by the 18S ribosomal RNA subunit (18S rRNA) band on the gel. Following RNA extraction, a DNase I (RNase-free) Kit (Thermo Fisher Scientific Inc., Canada) was used to reduce the likelihood that RNA might be contaminated with DNA. After total RNA was extracted, DNase I enzyme and DNase I buffer were added, and samples incubated at 37°C for 10 min. EDTA (0.5 mol l⁻¹) was added following the incubation period, followed by a final heat incubation of 75°C for 10 min.

Reverse transcriptase was used for cDNA synthesis from 1 μg total RNA. Random primers and 50 U MultiScribe MuLV reverse transcriptase were used in accordance with the manufacturer's instructions (High-Capacity cDNA Reverse Transcription Kit, Applied-Biosystems, by Fisher Scientific, ON, Canada). Cycling profiles for cDNA synthesis were performed at 25°C for 10 min, 37°C for 2 h and finally, 85°C for 5 min. A CFX384 Touch Real-Time PCR Detection System was used to perform qPCR (BioRad Laboratories Ltd., ON, Canada) using an advanced master mix with super green low rox reagent (Wisent Bioproducts Inc, QC, Canada) with 4 pmol forward and reverse primers (Table S1).

Diluted cDNA samples (10 times or 100 times, as appropriate), in a final reaction volume of 10 μl were used for qPCR temperature cycling of 3 min of initial denaturation at 95°C, followed by 39 cycles of 30 s at 94°C, 30 s at 60°C, and 1 min at 72°C, followed by a 10 min extension at 72°C. qPCR was performed using the CFX384 TouchTM Real-Time PCR Detection System. Quantitative validation was analyzed by the 2^^ΔΔCt method (Livak and Schmittgen, 2011). Actin and 18S rRNA were used as reference genes and have been validated in R. prolixus females under varied nutritional conditions (Majerowicz et al., 2011; Leyria et al., 2020). The dissociation curves of the cDNA products were examined to assess accuracy and a single peak was observed for each pair of primers.

Hemolymph samples were collected to measure 20E titers in unfed adult insects and at days 1–6 PBM. Hemolymph was collected using a Hamilton syringe (Hamilton Company, NV, USA) and combined with methanol at a ratio (1:3) (hemolymph:100% methanol), then stored at −20°C. Competitive ELISA (Abuhagr et al., 2014) was used to quantify the hemolymph ecdysteroid titers.

Double-stranded RNA (dsRNA) was used to downregulate the EcR/USP and shd transcripts. Using PCR, we prepared two non-overlapping fragments of target gene by conjugating the 5′-taatacgactcactatagggaga-3′ promoter from T7 RNA polymerase to the 5′ end of gene-specific primers (Table S1; Fig. S2). BLAST was used to compare the dsRNA sequences against the R. prolixus genome deposited in VectorBase. Only one high confidence hit was found for each dsRNA, i.e. genomic regions identified with 100% similarity with our dsRNA sequences, thereby confirming the specificity of each dsRNA and validating the experiments. The dsRNA was synthesized using the T7 Ribomax Express RNAi System (Promega, Madison, WI, USA). For controls, we used a dsARG (Ampicillin Resistance Gene) from the pGEM-T Easy Vector system (Promega, Madison, WI, USA).

At ecdysis, adult females were separated into four groups for knockdown of the transcripts for R. prolixus EcR, USP, Shd or the control, ARG. dsshd (1 µg µl⁻¹) and dsUSP (0.5 µg µl⁻¹) were injected on day 3 PE, and dsEcR (0.5 µg µl⁻¹) was injected on day 7 PE. Controls were injected with dsARG (0.5 µg µl⁻¹) on the equivalent days. These doses and days were found to give significant knockdown of the individual transcripts. All females were fed a blood meal on day 10 PE, and then each female was kept with two males in a cubicle at 28°C and 60% humidity and ejected spermatophore monitored to confirm successful mating. Expression of EcR, USP and shd transcripts were evaluated at 3 and 6 days PBM in the ovary (OV) and fat body (FB) (Fig. S3). Additionally, we examined the transcript levels of vitellogenin, R. prolixus Vg1 and Vg2, in the FB using RT-qPCR, as described above. Four independent biological samples were analyzed (N=4) at each time point.

Fifth instars were blood fed and kept at 25°C in an incubator. After emergence as adults, females and males were separated and kept in a 12 h light:12 h dark cycle. Adult females were injected either with dsUSP, dsshd or dsARG on day 3 PE, or with dsEcR or dsARG at 7 days PE. The insects were fed and mated on day 10 PE. Oogenesis in the ovaries was monitored, eggs laid by experimental and control females counted for 15 days, and hatching rate determined for those eggs laid.

Results are presented as the mean (mean+s.e.m.). To create the graphs, we used GraphPad Prism 7 (GraphPad Software, California, USA, www.graphpad.com). The datasets passed both normality and homoscedasticity tests. One-way ANOVA and Tukey's post hoc test for multiple group analysis was used to test for statistical differences for experiments with multiple groups and t-test was used to test for statistical differences between two groups. For each case, P<0.05 was considered statistically significant.

The genes of interest (Halloween genes, nvd, EcR and USP) were extracted using insect orthologs in the GenBank database and sequences verified using BLAST against the GenBank database. All of these genes are full length except for shd and USP, which are fragments (Table S2). All Halloween genes (spo, phm, dib, sad and shd) and non-Halloween gene (nvd) and the 20E receptor (EcR, USP) were identified in the ovary transcriptome. Phylogenetic analysis revealed that the protein sequences of Halloween genes, Nvd and EcR and USP of R. prolixus cluster with their orthologous sequences from other arthropod species (Fig. 1).

qPCR was used to quantify the transcript expression of Halloween genes, nvd, and EcR and USP from adult female tissues at 10 days PE during the unfed condition. The following tissues were examined: central nervous system (CNS), anterior midgut (AMG), Malpighian tubules (MT), hindgut (HG), FB, OV and oviduct (Ovid)+spermatheca (Sp). These transcripts showed differing levels of expression between tissues examined. mRNA levels of R. prolixus nvd, spo, phm, dib, sad and shd were significantly higher in OV compared with the other tissues (Fig. 2A–F). In addition, nvd, spo, phm, dib, sad and shd mRNA were detected in the HG and Ovid+Sp, and all were found in the CNS, except for sad. EcR and USP mRNAs were significantly higher in CNS, OV and HG compared with the other tissues. However, there was negligible expression of nvd, phm, spo, sad and shd in AMG, FB and MT. By contrast, dib, EcR and USP showed high expression in FB and low expression in AMG and MT. (Fig. 2G,H).

Based on the high transcript expression of Halloween genes, nvd and EcR in the ovary, we investigated more closely two regions of the ovary that may be involved in the synthesis of 20E, and that may respond to 20E. Expression was found for all these transcripts in the pre-vitellogenic ovariole (Ov_PV), which includes the tropharium and immature oocytes and vitellogenic ovariole (Ov_V), which are the follicles containing vitellogenic and mature oocytes. However, expression was significantly higher for all genes except nvd in the OV_PV compared with the Ov_V (Fig. 3A–H).

Transcript expression of Halloween genes, nvd and EcR was quantified in the ovaries of unfed adult females at 10 days PE, and then at various days PBM. The transcript level for nvd was elevated at day 1 PBM and that for phm was elevated at day 2 PBM. shd and sad transcripts were elevated during the first 3 days PBM, but then all these genes decreased over the next 3 days (Fig. 4A,C,E,F). dib and spo transcript levels did not show any statistical difference from the unfed levels after feeding but did decrease over the days PBM (Fig. 4B–D). In addition, the EcR transcript level increased significantly at all days examined PBM and USP transcript was also elevated at 1 and 2 days PBM; then all decreased over the next 4 days (Fig. 4G,H).

Hemolymph ecdysteroid titers were measurable on day 1 PBM, and increased dramatically on days 2–4, before plateauing at days 5 and 6 PBM (Fig. 5).

Unfed adult females that had been injected with dsRNA were blood fed 10 days PE. No effect of blood meal size was observed (Fig. S4). The effectiveness of the dsRNA treatments was tested by dissecting OV and FB at 3 days PBM (the time associated with the beginning of vitellogenesis) and 6 days PBM (the time when egg laying begins) and transcript expression of EcR, USP and shd quantified. There was a significant reduction of EcR, USP and shd transcripts in the OV and FB of knockdown insects at both time points (Fig. S3). Next, we quantified transcript expression level of Vg1 and Vg2 at 3 and 6 days PBM. A significant decrease in Vg1 and Vg2 transcript levels was observed 3 days PBM for dsEcR-, dsUSP- and dsshd-injected insects compared with dsARG-injected insects (Figs 6–8). The reduction in Vg1 and Vg2 transcript levels continued to day 6 PBM (Figs 6A–D, 7A–D and 8A–D). The ovaries of dsRNA-injected female R. prolixus were dissected and photographed at days 3 and 6 PBM. Knockdown of EcR, USP or shd led to a severe reduction in oogenesis and production of fewer mature oocytes than in the dsARG-injected control group. No oocyte resorption was observed. dsEcR-, dsUSP- or dsshd-injected insects also displayed a severe reduction in the number of growing follicles (vitellogenic follicles) compared with dsARG treatment at both time points (Figs 6E–H, 7E–H and 8E–H). In all treatments, the number of eggs laid was quantified and the results demonstrate that knockdown of EcR, USP or shd dramatically decreases the number of eggs laid per female compared with controls (dsARG) (Fig. 9A–C) and dramatically decreases hatching rate of the few eggs laid. In addition, the shape and chorion were abnormal as judged by the phenotype of the eggs (Fig. 9D–J). dsEcR and dsUSP resulted in one phenotype, with eggs that were laterally compressed (Fig. 9D–F). dsshd resulted in two phenotypes, one of which was similar to that seen for dsEcR and dsUSP, and the other in which the eggs were less laterally compressed (Fig. 9H–J). Knockdown of EcR expression resulted in a downregulation of ecdysone-responsive gene transcripts at day 3 and 6 PBM (E75, E74, Br-C, HR3 and FTZ-F1) (Fig. S5).

To investigate the effect of knockdown of the ecdysone receptor (EcR/USP) on choriogenesis, we quantified transcripts for Rp30 and Rp45 (choriogenesis genes) in OV of knockdown females at day 3 and 6 PBM. The results showed a significant reduction of Rp30 and Rp45 transcripts in insects injected with dsEcR, dsUSP or dsshd (Fig. 10).

The Halloween genes encode cytochrome P450 (CYP) enzymes, which are required for ecdysteroid biosynthesis in the insect's ovary (Peng et al., 2019; Ameku, et al., 2017). Halloween genes have been identified in many insect models such as D. melanogaster, B. mori, Acyrthosiphon pisum and Apis mellifera (Xia et al., 2004; Christiaens et al., 2010; Yamazaki et al., 2011). These genes are required for normal insect development, molting and reproduction. In this study, we identified five orthologs of the Halloween genes in the ovary of R. prolixus (spo, phm, dib, sad and shd), non-Halloween gene (nvd), as well as the genes encoding the ecdysteroid receptor, EcR and USP. The results suggest that ecdysteroids are synthesized by the ovaries in adult females of R. prolixus, from where they are released into the hemolymph following a blood meal. We also demonstrate that this pathway has a crucial role in regulating the development of eggs in R. prolixus via the ecdysteroid receptor.

In R. prolixus, the transcripts for Halloween genes, nvd, EcR and USP are expressed in different tissues in adult females, but the highest expression levels of all occur in the adult ovary. These results indicate that the ovary in adult females is a major organ that contributes to the synthesis of ecdysteroids as a hormone, but that also responds to ecdysteroids to control development of eggs. Other studies have also shown the expression of Halloween genes in adults of Manduca sexta, Plutella xylostella and Colaphellus bowringi (Rewitz et al., 2006b; Peng et al., 2019; Guo et al., 2021). This is consistent with the fact that 20E can be synthesized in adult gonads (Hagedorn et al., 1975; Lenaerts et al., 2019; Christiaens et al., 2010; Marchal et al., 2011). Moreover, in a variety of insects, including dipterans, hymenopterans and lepidopterans, the ovaries are a major source of ecdysteroids, which regulate female reproduction (Lange et al., 2022). The transcript expression levels of Halloween genes, nvd and the ecdysteroid receptor (EcR/USP) are high in both the tropharium and vitellarium. Halloween genes are also expressed in the ovarian follicle or nurse cells of D. melanogaster (Huang et al., 2008; Domanitskaya et al., 2014; Knapp and Sun, 2017; Ameku and Niwa, 2016; Ameku et al., 2017) and the four terminal hydroxylase genes are only found in the follicle cells of the developing ovary – often cited as the central site of ecdysteroid production – although phm, sad and shd are also located in the nurse cells during ovarian maturation. In Locusta migratoria and the cockroach Nauphoeta cinerea, ecdysteroid biosynthesis also occurs in follicle cells of the ovarioles (Kappler et al., 1986; Zhu et al., 1983). Interestingly, the presence of transcripts for the biosynthesis of ecdysteroids in non-reproductive tissues suggests that ecdysteroids can be synthesised in other tissues. Indeed, Bownes et al. (1984) found that female D. melanogaster have high levels of ecdysteroid in the gut and MTs, consistent with our result that the Halloween genes are highly expressed in HG and MTs. It should also be noted that dib and sad may be involved in detoxification and biocatalysis in insects (Scott, 1999; Bernhardt, 2006). As such then, some of the effects of downregulating the ecdysteroid pathway in R. prolixus could be due to the involvement of other tissues and not just the ovaries.

Halloween genes (phm, sad, shd), nvd, EcR and USP transcripts are upregulated in the first 3 days following a blood meal, suggesting that ecdysteroids are synthesized when the ecdysteroid receptor is upregulated. The transcripts encoding ecdysteroidogenesis enzymes, such as sro, spo, phm, sad and shd, are also upregulated after mating in D. melanogaster (Bellés and Piulachs, 2015; Niwa and Niwa, 2014). Wang et al. (2002) and Kapitskaya, et al. (1996) reported that EcR mRNA expression increased from 12 h PBM to 24 h PBM, and then declined at 36 h PBM in Ae. aegypti. USP transcripts are also elevated in the ovary at 18 and 24 h PBM, indicating that the ovaries can respond to 20E after a blood meal.

In R. prolixus, hemolymph ecdysteroids start to appear by day 1 PBM and reach their peak titer at days 5 and 6 PBM, matching the upregulation of the ecdysteroid synthetic pathway in ovaries during the first 3 days PBM. The hemolymph levels of ecdysteroids continue to increase when transcript levels for Halloween and nvd genes decrease after 3 days PBM, suggesting that sufficient amounts of enzymes have been synthesized for the production of ecdysteroids. Hemolymph titers of ecdysteroids measured here are consistent with Ruegg et al. (1981) who showed that ecdysteroids in the hemolymph peak 5 days after a blood meal in both virgin and mated R. prolixus. In addition, in Ae. aegypti, ecdysone hemolymph titers are slightly elevated 4 h PBM, with peak levels at 18–24 h PBM, and decline thereafter (Hagedorn et al., 1975). Ecdysteroids in R. prolixus are synthesized and released by the ovaries in sufficient quantities to account for the observed levels in the hemolymph (Cardinal-Aucoin et al., 2013). Also, in D. melanogaster 20E is secreted in vitro by ovaries (Rubenstein et al., 1982) and high-polarity products are released from ovaries (Dübendorfer and Maróy, 1986).

In this study, we used RNAi-based knockdown of EcR and USP to explore the function of ecdysteroids in reproductive processes, and included knockdown of shade, which is involved in the final step of ecdysteroid biosynthesis, converting E to 20E. Injection of dsEcR, dsUSP or dsshd into adult females led to a significant decrease in the expression levels of Rhopr-Vg1 and Rhopr-Vg2 transcripts at day 3 and 6 PBM in the fat body, as well as reducing considerably the number of developing oocytes at day 3 and 6 PBM. This suggests that the ecdysteroid signaling pathway is involved with vitellogenin synthesis in the fat body. This is consistent with Ae. aegypti, where USP and EcR regulate vitellogenesis by activation of Vg transcription by USP-EcR binding to the five regulatory regions of Vg (Martín et al., 2001). In D. melanogaster, EcR mutants show reduced accumulation of yolk, and a decline in fecundity through defects in the formation of egg chambers and nurse cells (Carney and Bender, 2000). In T. castaneum, reduction of either RXR/USP or EcR expression reduces Vg transcript levels, inhibits oocyte maturation and prevents egg laying (Xu et al., 2010). Clearly ecdysteroids play a critical role in Vg production, but only in certain insect species, such as Diptera (Dhadialla and Raikhel, 1994). In Blattodea and Hemiptera, including R. prolixus, juvenile hormone (JH) is essential for full egg production. By activating two receptors, JH regulates the production of vitellogenin in the fat body as well as its uptake by the ovary in R. prolixus (Lange et al., 2022). Our results suggest there may be crosstalk between JH and ecdysteroids in regulating vitellogenin in the fat body. In D. melanogaster, Vg synthesis in the fat body and uptake by the oocytes is controlled by 20E (Berger and Dubrovsky, 2005; Wu et al., 2021). After a blood meal in Ae. aegypti, JH promotes fat body competency for Vg synthesis, and 20E stimulates Vg expression and oocyte maturation (Shin et al., 2012; Wu et al., 2021).

In R. prolixus, knockdown of the transcripts involved in the ecdysteroid signaling pathway also reduced the number of eggs made and laid per female over 15 days, supportive of at least an effect on vitellogenesis and incorporation of vitellogenin into the growing oocytes. In addition, the abnormal shape of the small number of eggs laid in the EcR-, USP- and shd-depleted insects and the decrease in hatching of the eggs laid suggests that the eggs were unable to enter choriogenesis. Indeed, this is supported by the significant reduction of Rp30 and Rp45 transcripts in insects injected with dsEcR, dsUSP or dsshd. Impaired ovarian growth, oocyte maturation and follicle cell growth and migration resulting from knockdown of the 20E receptor has been reported in T. castaneum and impaired ovulation and oviposition in Schistocerca gregaria (Parthasarathy et al., 2010; Lenaerts et al., 2019). Knockdown of EcR altered the expression of the ecdysone responsive genes, and these results are in line with previous studies showing that ecdysteroids bind to the receptor, directly inducing early gene expression and coordinating the transcription of late genes specific to each tissue (Ashburner et al., 1974; Thummel et al., 1990). Also in D. melanogaster, chorion genes are regulated by EcR and USP (Bernardi et al., 2009) and in Blatella germanica, 20E causes precocious chorion matter deposition (Bellés et al., 1993). Moreover, EcR is also required for the expression of chorion genes and amplification in D. melanogaster (Hackney et al., 2007).

In summary, this is the first study to identify the Halloween genes in R. prolixus and to report the significance of the 20E genes for reproductive functions in oocyte development, egg laying and hatching. We identified all Halloween genes (spo, phm, dib, sad, shd) and the rieske superfamily member nvd, which are involved in ecdysteroid biosynthesis. We examined the role of 20E receptors in R. prolixus reproduction using RNAi and offer useful insights into the influence of these genes on oogenesis and choriogenesis. It may be possible to define RNAi-based pest management targets using these findings. Additional functional studies will investigate the role of ecdysone-responsive genes in the reproductive process in R. prolixus.

The authors thank Jimena Leyria for colony insect care and scientific input.