Role of Methoprene-tolerant in the regulation of oogenesis in Dipetalogaster maxima

Juvenile hormone (JH) signalling, via its receptor Methoprene-tolerant (Met), controls metamorphosis and reproduction in insects. Met belongs to a superfamily of transcription factors containing the basic Helix Loop Helix (bHLH) and Per Arnt Sim (PAS) domains. Since its discovery in 1986, Met has been characterized in several insect species. However, in spite of the importance as vectors of Chagas disease, our knowledge on the role of Met in JH signalling in Triatominae is limited. In this study, we cloned and sequenced the Dipetalogaster maxima Met transcript (DmaxMet). Molecular modelling was used to build the structure of Met and identify the JH binding site. To further understand the role of the JH receptor during oogenesis, transcript levels were evaluated in two main target organs of JH, fat body and ovary. Functional studies using Met RNAi revealed significant decreases of transcripts for vitellogenin (Vg) and lipophorin (Lp), as well as their receptors. Lp and Vg protein amounts in fat body, as well as Vg in hemolymph were also decreased, and ovarian development was impaired. Overall, these studies provide additional molecular insights on the roles of JH signalling in oogenesis in Triatominae; and therefore are relevant for the epidemiology of Chagas´ disease.

www.nature.com/scientificreports/ In this work, two Met transcripts of D. maxima (DmaxMet) were cloned and sequenced, enabling their characterization, as well as in silico receptor-ligand interaction studies by bioinformatics analyses. To gain further insight into the role of JH signalling regulating oogenesis, we measured the changes in Met transcripts in the fat body and ovaries during the gonadotrophic cycle. In addition, we evaluated the effects of reducing the expression of Met by RNA interference on female reproductive phenotypes. Our results validated the important role of Methoprene-tolerant in JH-signalling regulating oogenesis in D. maxima.

Sequence analysis and structural properties of DmaxMet. Using a Rapid Amplification cDNA
Ends approach (RACEs) we cloned and sequenced the cDNA coding for Met in D. maxima. Two isoforms were found, DmaxMet1 and DmaxMet2 (accession numbers MZ361686.1 and MZ753950.1, respectively). The DmaxMet1 open reading frame (ORF) of 2456 nucleotides encodes a 713 amino acid protein (81 kDa) with an isoelectric point of 6.68. In addition, DmaxMet2 presents an ORF of 2411 nucleotides, encoding a 676 amino acid protein (76 kDa) with an isoelectric point of 6.20 ( Supplementary Fig. S1A). The deduced DmaxMet1 and DmaxMet2 proteins presented the classic domains reported for Met, namely, a bHLH DNA-binding domain, the tandemly arranged PAS-A and PAS-B domains and the PAS-associated C-terminal motif (PAC) 6,20 (Fig. 1a and Supplementary Fig. S1B).
We employed the two R. prolixus Met isoforms sequences (RpMet1 and RpMet2) 21 to predict the molecular structures of DmaxMet1 and DmaxMet2 transcripts. D. maxima ORFs span 15 exons, with 13 of the exons showing high identity and similarity with their orthologs RpMet1 and RpMet2. While R. prolixus transcript isoforms   Fig. S2). Molecular modelling of DmaxMet was performed using the I-TASSER program 22 . Crystal structures for the PAS-B domains of Met have not been reported; consequently, our homology model was developed based on the crystal structure of the hypoxia-inducible factor 2α (HIF-2α), a member of the bHLH/PAS protein family 23 . Only the region of the full sequence containing both PAS domains was used to build the model (Fig. 1b). The TM-score (Template Modelling score) was 0.72 ± 0.11 (a TM score > 0.5 indicates correct topology) 22 .
Because the crystallized PAS region is similar in both DmaxMet isoforms, the three-dimensional protein structure obtained from DmaxMet1 was used for in silico docking of JHSB 3 . The ligand docking exposed the amino acids bordering an inferred hydrophobic JH-ligand pocket in the PAS-B domain (Fig. 1b). Alignment of PAS-B domains from DmaxMet1, Tribolium castaneum Met, Aedes aegypti Met and Drosophila melanogaster Met highlighted the conserved and mutated amino acid residues that are relevant for JH-binding identified by either modelling or mutagenesis experiments in these four species (Fig. 1c).
Changes in DmaxMet mRNA levels throughout the reproductive cycle of D. maxima. Transcript levels of the DmaxMet gene were analyzed in the fat body and ovaries at representative days during the reproductive cycle (Fig. 2a). The results showed that in both tissues, mRNA levels were low during pre-vitellogenesis, increased at the beginning of vitellogenesis, and reached maximum levels at day 4 after the blood meal. Afterward, DmaxMet transcripts sharply dropped as vitellogenesis progressed, and by day 6 after feeding, levels were low and similar to those in pre-vitellogenic and post-vitellogenic females (Fig. 2b,c).

Effect of DmaxMet gene silencing.
To better understand the role of Met in the regulation of oogenesis, we performed the knockdown of DmaxMet gene expression by RNAi in pre-vitellogenic females (Fig. 3a). Survival rates and amounts of blood ingested showed no significant differences between dsDmaxMet-treated insects and the control insects injected with a dsRNA molecule encoding the partial sequence of ampicillin resistance gene (dsARG). The efficiency of gene knockdown, evaluated by qPCR, indicated that DmaxMet transcript levels at day 2 after RNAi treatment were reduced by 84% in fat body and by 78% in ovaries (Fig. 3b,c).
Knockdown of the DmaxMet gene caused significant decreases of Vg and Lp mRNA levels in the fat body (Fig. 3d,e), as well as VgR and LpR transcripts in ovarian tissue (Fig. 3f,g). At the protein level, the contents of Vg and Lp in fat body notably decreased in DmaxMet-silenced females (Fig. 4a-c). However, when assessed in www.nature.com/scientificreports/ the hemolymph, Vg levels were severely reduced in RNAi-treated females (Fig. 4d), whereas no differences on Lp titers were observed between dsDmaxMet and dsARG injected insects (Fig. 4e). DmaxMet gene silencing also influenced the maturity of ovarian follicles, radically decreasing the development of terminal or mature follicles (Fig. 5a,b). When analyzed by immunofluorescence, the signals for Vg and Lp were notably decreased in the perioocytic space and oocyte membrane, and weakly detected in the oocytes of terminal follicles in DmaxMet-silenced females compared to controls; consistent with an impairment of lipoprotein uptake by the developing oocytes (Fig. 5c).

Discussion
In adult female Triatominae, a blood meal triggers a sequence of physiological processes controlled by endocrine factors and leading to a successful completion of oogenesis 24 . Almost a century ago, Sir Vincent B. Wigglesworth described that females of R. prolixus decapitated 24 h after feeding arrested ovarian development 3 . Later, it was reported that the CA was the source of JH, the main hormone controlling reproduction in Triatominae 25 . Since then, the effects of JH and its signalling pathway have been extensively studied. However, for many decades, attempts to identify the JH receptor were unsuccessful. Methoprene is a potent analogue of JH used as an insecticide 26 . In the 1980s it was described a D. melanogaster mutant line resistant to the lethal effects of Methoprene. The mutated gene was essential for the action of the JH analogue, and it was named Met, for Methoprene tolerant 27 . The role of Met as the intracellular receptor for JH has been validated by multiple studies in holometabolous and hemimetabolous insects 20 . A previous study performed a genomic and functional characterization of Met in R. prolixus (RpMet) 21 . Silencing of RpMet resulted in precocious metamorphosis and abnormal ovarian development 21 . Precocious development of adult features in R. prolixus nymphs were also reported upon silencing of Krüppel-homolog 1 (Kr-h1), a JH-inducible repressor of metamorphosis 28 .
We cloned and sequenced the transcripts coding for Met in D. maxima. A phylogenetic analysis showed that the DmaxMet translated nucleotide sequences are evolutionarily related to RpMet, encoding also two isoforms 21 . Both DmaxMet isoforms present the bHLH, PAS-A and PAS-B domains described in other insect species, supporting the statement that the JH receptor is highly conserved 1,7,20 .
Our modelling and ligand docking revealed amino acids bordering an inferred hydrophobic JH-ligand pocket in the PAS-B domain. As depicted in Fig. 1c, most of these amino acid residues are conserved in Met of other insect species, and have been identified as critical for JH-binding by modelling and mutagenesis experiments. In T. castaneum, mutations T254Y, V280F, V297F, T330Y, 347 M, I262F and L318F hindered JH III binding 7 . In A. aegypti, 14 residues were predicted to be important to bind JH-III: T403, H405, I411, I418, V429, F437,   12,13,31 . Circulating JHSB 3 levels during the reproductive cycle were recently reported in D. maxima females, with a sharp increase observed at day 4 after a blood meal 19 . We now established that DmaxMet mRNA expression in fat body and ovaries showed maximum levels of expression at 4 days post-feeding, when JHSB 3 hemolymph titers are high.
Insect vitellogenesis is regulated by JH and 20-hydroxyecdysone (20E). JH stimulates vitellogenesis in hemimetabolous and most holometabolous insects. 20E contols vitellogenesis in selected dipterans, lepidopterans and hymenopterans (reviewed by Wu et al., 2021) 32 . Recent reviews have covered the pleiotropic functions of JH in female insect reproduction 1,33 . Topical application of JH III on pre-vitellogenic D. maxima females stimulated the transcription of Vg and Lp genes in the fat body; while females treated with the anti-JH compound Precocene II exhibited a down-regulation in the transcription of Vg 19 . Our results using DmaxMet RNAi confirmed that the JH-Met pathway plays a critical role in the regulation of Vg expression in Triatominae 24 . Remarkably, knockdown of DmaxMet gene also down-regulated the Lp gene in the fat body, although it did not impact the Lp levels in the hemolymph. It is known that in addition to the rate of synthesis in the fat body, circulating Lp is influenced by other factors, such as its long half-life that favors the cycling among tissues, to exchange lipids as a reusable shuttle 34 .
Previous studies highlighted the role of JH controlling yolk incorporation into developing oocytes in insects; JH is involved in the transcriptional regulation of VgR in the red ant Solenopsis invicta 35 , as well as in VgR and LpR expression in Ae. aegypti 36 . In addition, it has been reported a post-transcriptional control of LpR during the reproductive cycle of the cockroach Blatella germanica 37 . In the cockroach Diploptera punctata, Met-silenced insects showed impaired vitellin storage in the terminal follicles 38 . In the oocytes of two cockroach species, vitellogenesis induced changes in the subcellular distribution of VgR and LpR, which translocated to the cortical www.nature.com/scientificreports/ zone to facilitate the receptor-ligand interaction and further lipoprotein uptake 39 . In Triatominae, knowledge of the role of JH on the expression of these receptors is limited.
In D. maxima, we previously described the action of JH on the expression of VgR and LpR genes and on their ligands Vg and Lp 19 . In the current studies we report that DmaxMet gene silencing resulted in decreased VgR and LpR transcript levels, and therefore impaired Vg and Lp uptake by oocytes, with the associated reduction in the development of terminal follicles. Terminal follicle size from dsRNA-treated insects. The results are shown as mean ± SD of three independent experiments and included samples from insects processed individually (n = 4). The statistical analyses were performed using Student's t-test (***P < 0.001). (c) Effect of dsMet and dsARG treatment on Vg and Lp uptake by developing oocytes. Positive signal for Vg (red) and Lp (green) observed in the perioocytic space and oocyte membrane (arrows) as well as in the oocytes (asterisks) of dsARG-treated females (controls) was decreased or faintly detected in dsMet-silenced insects. Bars: 100 µm. FE (follicular epithelium); TF (terminal follicle); Tp (tropharium). Similar results were obtained in three independent experiments.

Methods
Ethics statement. Housing conditions and manipulation of hens employed in the maintenance of the insect colony followed the protocol authorized by the Animal Care Committee of the Centro de Investigaciones en Bioquímica Clínica e Inmunología (CIBICI-CONICET-Universidad Nacional de Córdoba). They followed the guidelines published by the Canadian Council on Animal Care with the assurance number A5802-01 delivered by the Office of Laboratory Animal Welfare (National Institutes of Health). The animal facility at the CIB-ICI-CONICET is a dependency of the Argentine National Ministry of Science Science (http:// www. biote rios. mincyt. gob. ar). No infected insect species, human blood, or hen sacrifices were involved in the study. Details of the approved protocol were previously published 17 .
Insects. Anautogenous females were taken from a colony of D. maxima, maintained at 28 °C, 70% relative humidity, 8:16 h light/dark photoperiod. Standardized conditions of insect rearing were previously described 15 .
Briefly, females were separated from males at the last nymphal stage before blood feeding. Newly emerged females were segregated individually and placed in jars together with two fed males. Successful mating was confirmed by observation of the spermatophore. Mated females were kept in individual jars until they were able to take a blood meal (days 10-12 post-ecdysis), which resulted in a 3.0-5.5-fold increase in the body weight of the insect. Thereafter, egg laying was daily monitored, and the beginning and the end of the oviposition period were recorded. Unless otherwise stated, experiments were performed by sampling hemolymph and tissues at representative days of the reproductive cycle as previously descibed 15,17 : pre-vitellogenesis (days 4-6 post-ecdysis, unfed period); vitellogenesis (days 2, 3, 4, 6 and 12 post-blood feeding); and post-vitellogenesis (days 12-14 after the end of oviposition).
Hemolymph and tissues sampling. Hemolymph was individually collected from immobilized females with a Hamilton syringe by cutting off the legs at the level of the coxa and gently pressing the abdomen. The collected samples were placed in cold tubes containing 10 mM Na 2 EDTA, 5 mM dithiothreitol (DTT) and a cocktail of protease inhibitors (P8340, Sigma-Aldrich, St. Louis, MO, USA) as previously described 17,19 . Hemolymph samples were centrifuged at 10,000 × g for 10 min at 4 ℃ to remove hemocytes. Protein amounts were measured using the Bradford method 40 . Samples were stored at − 70 °C until use. Ovarian tissues and fat bodies from females at representative days of the reproductive cycle were dissected out in cold phosphate-buffered saline (PBS: 6.6 mM Na 2 HPO 4 /KH 2 PO 4 , 150 mM NaCl, pH 7.4), using a standard stereoscope, as previously described 17,19 .
Cloning the full-length cDNA of DmaxMet. The complete sequences of the DmaxMet transcripts were obtained by Rapid Amplification cDNA End (RACE) (SMARTer RACE 5′/3′ kit, Takara Bio USA, Inc). Briefly, for the first cDNA synthesis step, a universal mixture of primers was used to recognize specific regions of the 3′ or 5′ ends (3′ RACE-cDNA or 5′ RACE-cDNA, respectively). Synthesis of RACE-cDNAs was performed using 1 μg of ovarian total RNA. For the RACE amplification reaction, DmaxMet gene-specific primers were: a forward primer for 3′ RACE-cDNA and a reverse primer for 5′ RACE-cDNA. To achieve optimal yields, the DmaxMet primers were designed with a length of 23-28 nucleotides (nt), containing a guanidine/cytosine (GC) ratio of 50-70% and a melting temperature (Tm) > 70%. Additionally, a 15 nt sequence (GAT TAC GCC AAG CTT), necessary for the insertion of the gene of interest into the p-RACE vector, was added to the primers. Once the amplification reaction was completed, the size of the RACE product was verified by 1% agarose gel electrophoresis. For optimal quality in the sequencing, the fragments obtained by RACE were cloned on the p-RACE vector following the manufacturers' recommendations. The clones obtained were then sequenced using a 3130XL Genetic Analysers in the DNA core facility at FIU (Miami, FL, USA).
Bioinformatic analyses and modelling. The deduced amino acid sequence, molecular mass and isoelectric point prediction for the two DmaxMet isoforms were assessed using tools available on Expasy (http:// web. www. expasy. org; SIB Bioinformatics Resource Portal). The Basic Local Alignment Search Tool algorithm (BLAST, https:// blast. ncbi. nlm. nih. gov/ Blast. cgi) was employed to compare the sequence of the transcripts coding for DmaxMet isoforms with other Met sequences deposited in databases. BLAST was also used to predict the major functional domains of the receptor. Mapping of exons in DmaxMet was achieved with the Exon-Intron Graphic Maker (http:// wormw eb. org/ exoni ntron), taking as references the homologous sequences reported for R. prolixus 21 .
The molecular model of DmaxMet was built by homology modelling using I-TASSER package 22 . The HIF2α-Arnt crystal structure (PDB ID 3F1P) was chosen as an homologous template 7 . The visualization of the structures was performed by UCSF Chimera 41  www.nature.com/scientificreports/ The obtained DmaxMet three-dimensional model was used in docking simulation to analyze the interaction between DmaxMet and JHSB 3 , only the DmaxMet PAS domain were considered, since they are the only domains with crystallized templates deposited in databases. The docking calculations were carried out with DockThor version 2 42,43 .
For the evolutionary analysis of Met, the predicted amino acid sequences for DmaxMet were compared with selected Met sequences retrieved from the NCBI database. The sequences were aligned using MAFFT 44 , and twenty alternative full-length alignments were combined via GUIDANCE2 45 and used for inferring the protein phylogeny. The phylogenetic tree was constructed with PhyML 46 under Maximum Likelihood model (LG + G + I) [47][48][49] , using branch support analysis 50 . RNA extraction and reverse transcriptase quantitative real-time PCR (RT-qPCR). Ovaries and fat bodies were subjected to RNA extraction with TRIzol (Thermofisher Waltham, MA, USA), according to the manufacturer's instructions. Samples were treated with DNase to remove genomic DNA (Promega, Heidelberg, Germany) and RNA integrity was evaluated by 1% agarose gel electrophoresis. cDNA synthesis from 2 µg of total RNA was performed using the Moloney Murine Leukemia Virus reverse transcriptase protocol (M-MLV, Promega, Heidelberg, Germany). The qPCR assays were performed with an ABI Prism 7500 sequence detection system (Applied Biosystems, Foster City, CA, USA) using a Power SYBR Green PCR master mix (Thermo Fisher Scientific, Waltham, MA, USA), as previously described 19 . The 2 −ΔΔCt method was employed to evaluate of relative changes in gene expression 51 , using 18S ribosomal RNA (18S rRNA) as a reference gene 19,52 . Unless otherwise stated, all reactions were carried out by triplicate. The primer sequences of all genes assayed are displayed on Supplementary Table 1. In the case of DmaxMet, the primers were designed to target a region shared by the DmaxMet 1 and DmaxMet 2. In D. maxima, vitellogenin (Vg) is the product of the genes Vg1 and Vg2, the latter showing the highest level of expression 19 . Therefore, in the present work Vg2 was the gene quantified and named Vg for simplicity.
The amplification efficiency for each pair of primers was calculated using standard curves generated by serial dilutions of cDNA. All reactions showed an efficiency higher than 96% for the different pair of primers tested.

Double-stranded RNA (dsRNA) synthesis and knockdown of DmaxMet gene expression.
A 317 base pair Met template was used to synthesize a double-stranded RNA (dsRNA) molecule. Gene-specific primers targeting a region shared by the two DmaxMet isoforms were designed and then combined with the T7 RNA polymerase promoter sequence. The synthesis of dsRNAs was carried out using the MEGAScript RNAi kit (Thermo Fisher Scientific), according to the manufacturer's recommendations. dsRNAs were precipitated with ammonium acetate/ethanol, and then suspended in ultrapure RNase-free water to a final concentration of 5 μg/ μl. A dsRNA molecule based on the ampicillin resistance gene (dsARG) from the pGEM-T Easy Vector system was used as a control, as previously described 19,53 .
To knockdown the expression of DmaxMet, adult females at 10-12 days post-ecdysis were injected into the hemocoel with 10 µg of dsMet or dsARG (control) dissolved in 2 µl of ultrapure water using a Hamilton microsyringe. At 24 h post-injection, dsRNA-treated insects were fed with blood, and 2 days later fat bodies and ovaries were individually sampled and processed by qPCR to determine the DmaxMet mRNA levels. The effect of DmaxMet silencing on Vg and Lp gene expression in fat bodies as well as on VgR and LpR in ovarian tissue was also assessed by qPCR.
Vitellogenin and lipophorin assessments. To further evaluate the relevance of DmaxMet knockdown on the yolk protein precursors Vg and Lp metabolism, a group of females were treated with dsMet or dsARG (control). At 24 h post-injection, dsRNA-treated insects were fed with blood, and 2 days later hemolymph, fat bodies and ovaries were collected. Quantification of Vg and Lp was accomplished by indirect enzyme-linked immunosorbent assay (ELISA), using a polyclonal anti-vitellin antibody (anti-Vt, 1:4000), and a polyclonal anti-lipophorin antibody (anti-Lp, 1:5000) as previously reported 15,19,54 . Purified Vt and Lp were employed to build calibration curves 14,54 . For the assays, the hemolymph was individually collected and directly processed for ELISA. Fat bodies were weighed and then homogenized in buffer Tris-NaCl (20 mM Tris, 150 mM NaCl, pH 7.4) containing a cocktail of protease inhibitors. The homogenates were first centrifuged at 2500 × g for 10 min at 4 °C and the floating fat cake and pellet were discarded. The resulting material was centrifuged at 15,000 × g for 30 min at 4 °C and the supernatants collected and used for ELISA, as previously described 19 .
Vg and Lp were also evaluated in ovarian tissue by immunofluorescence assays, as previously described 19 . Briefly, the ovaries from dsDmaxMet or control females were dissected out and fixed with 4% paraformaldehyde in PBS, transferred to PBS/sucrose and processed for cryostat sectioning 54 . Thereafter, tissue sections (10 µm) were obtained using a Leica CM1510 cryostat (Leica Microsystems, Wetzlar, Germany) and placed onto poly-L-lysine-treated glass slides (Sigma-Aldrich, St. Louis, MO, USA). For the assays, the slides were blocked with 1% BSA, 0.1% Triton X-100, 5% fetal bovine serum in PBS during 1 h, and then, incubated sequentially as follow: anti-Vt antibody (1:4000), followed by a secondary anti-rabbit IgG antibody labeled with Alexa www.nature.com/scientificreports/ ated for normality and homogeneity of variance using the Shapiro-Wilk test, which showed that no transformations were needed. All datasets passed normality and homoscedasticity tests. Statistical analyses were performed employing Student's t-test or parametric one-way ANOVA, and using the Student-Newman-Keuls multiplecomparisons as a post-test. A P value < 0.05 was considered statistically significant.

Data availability
All data generated or analysed during this study are included in this published article and its supplementary information files. www.nature.com/scientificreports/