Molecular identification, transcript expression, and functional deorphanization of the adipokinetic hormone/corazonin-related peptide receptor in the disease vector, Aedes aegypti

The recently discovered adipokinetic hormone/corazonin-related peptide (ACP) is an insect neuropeptide structurally intermediate between corazonin (CRZ) and adipokinetic (AKH) hormones, which all demonstrate homology to the vertebrate gonadotropin-releasing hormone (GnRH). To date, the function of the ACP signaling system remains unclear. In the present study, we molecularly identified the complete open reading frame encoding the Aedes aegypti ACP receptor (ACPR), which spans nine exons and undergoes alternative splicing giving rise to three transcript variants. Only a single variant, AedaeACPR-I, yielding a deduced 577 residue protein, contains all seven transmembrane domains characteristic of rhodopsin-like G protein-coupled receptors. Functional deorphanization of AedaeACPR-I using a heterologous cell culture-based system revealed highly-selective and dose-dependent receptor activation by AedaeACP (EC50 = 10.25 nM). Analysis of the AedaeACPR-I and AedaeACP transcript levels in all post-embryonic developmental stages using quantitative RT-PCR identified enrichment of both transcripts after adult eclosion. Tissue-specific expression profiling in adult mosquitoes reveals expression of the AedaeACPR-I receptor transcript in the central nervous system, including significant enrichment within the abdominal ganglia. Further, the AedaeACP transcript is prominently detected within the brain and thoracic ganglia. Collectively, these results indicate a neuromodulator or neurotransmitter role for ACP and suggest this neuropeptide may function in regulation of post-ecdysis activities.

Scientific RepoRts | (2018) 8:2146 | DOI: 10.1038/s41598-018-20517-8 RACE-PCR reactions utilized Q5 High Fidelity DNA Polymerase in lieu of SeqAmp DNA Polymerase. Nested PCR reactions utilized gene specific forward (3′ RACE) and reverse (5′ RACE) primers and a universal primer mix (UPM) to amplify the complete cDNA encoding A. aegypti ACPR. Optimal PCR cycling parameters for subsequent amplification of ACPR were determined empirically. Specifically, for 3′ RACE this included an initial denaturation at 94 °C for 1 min, followed by 40 cycles of 30 s at 94 °C, 30 s at 68 °C, and 3 min at 72 °C to amplify PCR products using SeqAmp DNA Polymerase. For 5′ RACE, the Q5 High Fidelity DNA Polymerase was utilized with the following cycling parameters, 30 s at 98 °C, followed by 30 cycles of 5 s at 98 °C, 15 s at 65-68 °C, 1 min 10 s at 72 °C, with a final extension step of 2 min at 72 °C. Following two rounds of PCR using nested gene-specific primers, amplicons were gel extracted and cloned into the linearized pRACE vector and miniprep samples were then sent for sequencing. Finally, primers were designed at the 5′ and 3′ ends of the complete cDNA sequence (including UTRs) and at the start and stop codons of the sequence (region including only the open reading frame, excluding UTRs), and were used for subsequent PCR amplification of the receptor with Q5 High Fidelity DNA polymerase to confirm base pair accuracy.

Gene Structure and Phylogenetic Analyses. Mapping of exon-intron boundaries of A. aegypti ACPR
gene was determined using the cloned complete cDNA sequence as a query against the A. aegypti genome scaffolds database available locally on a lab computer running Geneious Pro Bioinformatics Software (Biomatters Ltd, Auckland, New Zealand). Positions of introns and exons were further confirmed using the BDGP splice site prediction server using the standardized data set of D. melanogaster genes 28 . Membrane topology of ACPR-I, II, and III were predicted using the Constrained Consensus TOPology prediction server (CCTOP) 29 . The deduced AedaeACPR-I, II and III protein sequences were aligned to the human gonadotropin-releasing hormone receptor 1 along with ACP, AKH, and CRZR receptors from other species (see Table S1) using ClustalW in MEGA 6.06 30 . Relationships between the various receptor sequences were determined through neighbour-joining 31 and maximum-likelihood phylogenetic analysis methods 32 . Bootstrap values are based on 1000 replicates.

Preparation of mammalian expression constructs. Amplicons encoding just the open reading frame
(start ATG to stop codon) were used as template for re-amplification using a modified forward primer possessing the consensus Kozak translation initiation sequence 33,34 at the 5′ end of the start codon. The resulting product was cloned into pGEM-T Easy vector and then subcloned into the mammalian expression vector, pcDNA 3.1 + (Life Technologies, Burlington, ON). Construct directionality was confirmed by Sanger sequencing and plasmid  To determine whether this difference in the resulting residue, in comparison to the A. aegypti genome database, confers any difference to the functional activity of the receptor, site directed mutagenesis was performed. Specifically, 5′ phosphorylated primers were designed (Table 1) with the forward primer possessing an adenine (position 1924) consistent with the A. aegypti genome sequence whereas our consensus sequence contained a thymine in this nucleotide position. Using these modified primers, asymmetric PCR was performed using a pGEM-T Easy plasmid construct as template to replace the Ile 472 in the cloned receptor with an Asn 472 matching the A. aegypti genome. Mutation of the coding sequence was verified by sequencing and sub-cloned into the mammalian expression plasmid, pcDNA3.1 + (as described above).
Cell culture, transfections, and bioluminescence assay. Functional activation of AedaeACPR-I was assayed using a previously established cell culture system involving a recombinant Chinese hamster ovary (CHO)-K1 cell line stably expressing aequorin 35 (Table 1) were prepared in BSA medium (10 −5 to 10 −12 M), and loaded in quadruplicates into 96-well luminescence plates (Greiner Bio-One, Germany). All peptides were commercially synthesized (Genscript, Piscataway, NJ) at a purity >90% and were prepared in dimethyl sulfoxide at a stock concentration of 1 mM. Cells were loaded into each well with an automatic injector unit and luminescence was measured for 20 seconds using a Synergy 2 Multi-Mode Microplate Reader (BioTek, Winooski, VT, USA). BSA medium alone was utilized as a negative control and 5 × 10 −5 M ATP was used as a positive control, which acts on endogenously expressed purinoceptors 36,37 . EC 50 values were calculated in GraphPad Prism 7.02 (GraphPad Software, San Diego, USA) from dose-dependent curves from four independent transfections.
Tissue dissections, RNA extraction, and cDNA synthesis. Lightly CO 2 -immobilized four-day old adult male (n = 30) and female (n = 20) A. aegypti were submerged in DPBS, and the following body segments and/or tissues were dissected and isolated: head, midgut, Malpighian tubules, hindgut, ovaries, testes, accessory reproductive tissues, and carcass (remaining fat body, musculature, and cuticle). Tissues were lysed in RNA lysis buffer containing 1% 2-mercaptoethanol. Whole adult RNA was obtained from submerging several males and females in RNA lysis buffer containing 1% 2-mercaptoethanol and using a sterile plastic pestle to disrupt the tissue. To measure the developmental expression profile for AedaeACPR, first to fourth instar larvae, pupae, as well as one-and four-day old adult mosquitoes were collected and submerged in RNA lysis buffer and flash frozen in liquid nitrogen. Total RNA was isolated from whole animal and individual adult tissues samples mentioned above using the PureLink TM RNA mini kit following manufacturer protocol with an on-column DNase treatment to remove genomic DNA (Invitrogen, Burlington, ON). Purified total RNA samples were quantified with a Take3 micro-volume plate and measured on a Synergy Multi-Mode Microplate Reader (BioTek, Winooski, VT, USA). To assess ACP and ACPR transcript levels, cDNA was synthesized from 20 ng total RNA using the iScript ™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad, Mississauga, ON) following manufacturers protocol, including a ten-fold dilution of cDNA following synthesis.
Quantitative PCR. ACP and ACPR transcript abundance was quantified on a StepOnePlus ™ Real Time PCR system (Applied Biosystems, Carlsband, CA) using PowerUP ™ SYBR ® Green Master Mix (Applied Biosystems, Carlsband, CA). Cycling conditions were as follows: 1) UDG activation 50 °C for 2 min, 2) 95 °C for 2 min, and 3) 40 cycles of (i) 95 °C for 15 seconds and (ii) 60 °C for 1 minute. Gene-specific primers designed over different exons were used to amplify ACPR, with the forward primer designed over exon 5 (nucleotides 1458-1476) to ensure specificity for ACPR-I, and the reverse primer over exon 6 (nucleotides 1585-1603). Gene specific primers amplifying AedaeACP were designed over multiple exons (Table 1; forward: nucleotides 89-112, reverse: nucleotides 403-421) based on a previously published mRNA sequence (Genbank Accession Number: FN391984) 20 . Relative expression levels were determined using the ΔΔC T method and were normalized to the geometric mean of rp49, rpL8, and rps18 reference genes, which were previously characterized and determined as optimal endogenous controls 38 . The AedaeACPR spatial expression profile was determined using 7-9 biological replicates, all of which included three technical replicates per reaction and a no-template negative control. The AedaeACPR developmental expression is an average of 3-5 biological replicates that each included duplicate technical replicates for each target gene and a no-template negative control. The AedaeACP spatial expression profile consisted of 3-4 biological replicates and the developmental expression is an average of 3-5 biological replicates. Specificity of primers for target mRNA were assessed by conducting no reverse-transcriptase controls, analysis of dissociation curves, and Sanger sequencing of amplicons. Data were analyzed using a one-way ANOVA with Dunnett's multiple comparisons test where p < 0.05 was considered significant.
Data availability. The datasets generated and analysed during the current study are available from the corresponding author on reasonable request.

Results and Discussion
We have identified the complete cDNA sequence encoding the A. aegypti adipokinetic hormone/corazoninrelated peptide (ACP) receptor (Fig. 1a). Following initial cloning and sequence analysis, three transcript variants were identified, AedaeACPR-I, ACPR-II, and ACPR-III (Fig. 1b). AedaeACPR-I is 2596 bp (GenBank Accession number: MF461644), which includes a 1734 bp open reading frame (ORF) encoding a 577 residue receptor (Fig. 1a). The cloned 5′ untranslated region (UTR) is 509 bp in length and the 3′UTR is 346 bp and contains a predicted polyadenylation sequence (nucleotide position 2547-2552). AedaeACPRs II and III are transcript variants of 2442 bp (GenBank: MF461645) and 2240 bp (GenBank: MF461646) in length, which yield deduced proteins comprised of 328 and 243 amino acids, respectively. Only AedaeACPR-I has the seven expected hydrophobic transmembrane (TM) domains characteristic of GPCRs, whereas AedaeACPR-II has only five TM domains and AedaeACPR-III has only three TM domains (see Fig. S1). Similarly, previous studies in R. prolixus 25 (Fig. 1b).
Alignment of A. aegypti ACPR-I, with selected receptors from A. gambiae, T. castaneum, R. prolixus, and B. mori, reveals conservation of the ACP receptor across insect species (Fig. 2). Specifically, AedaeACPR-I shares 59.4% sequence identity with the A. gambiae ACP receptor, 42.4% identity with the R. prolixus ACPR-C, 41.5% identity with the T. castaneum ACPR and 33.8% identity with the B. mori ACP receptor. Overall, there is a high degree of conservation over the seven predicted TM domains, particularly over TM regions one, two, three, five and seven. Strong sequence identity is also observed in the first and second intracellular loops, as well as the first extracellular loop. All of the receptor sequences, except for BommoACPR, which harbors an Asp in place of Asn, possess the conserved NPXXY motif in the seventh TM domain characteristic of rhodopsin-like (family A) GPCRs 39,40 . Another conserved motif found in rhodopsin-like GPCRs is the E/DRY motif adjacent to the second intracellular loop 40 . The Arg of the E/DRY motif and a negatively charged residue on TMVI of the GPCR undergo ionic interactions, known as the ionic lock, which stabilizes the inactive state of the receptor 41 . In particular, the ACP receptors possess a DRF motif in the silkworm B. mori, and DRC motifs are found in the mosquitoes A. gambiae and A. aegypti, in place of the characteristic DRY motif found in hemipteran R. prolixus and coleopteran T. castaneum, which have more conserved features of rhodopsin-like GPCRs 40,42 .
Phylogenetic analysis using the neighbor-joining and maximum-likelihood methods (not shown) yielded trees with highly similar and well supported topologies (Fig. 3). All the ACP receptors analyzed are positioned within a single clade that is a sister group to the clade comprised of the AKH receptors. Together, the AKH and ACP receptor clades form a monophyletic group which is a sister group to the clade comprised of CRZ receptors. The AedaeACPR-I identified herein clusters closely with the other insect ACPRs that have previously been identified and functionally characterized confirming that the receptor isolated in this study is an ortholog of other insect ACP receptors 17,18,22,25 . Predicted ACPRs from other mosquito species including, Anopheles darlingi, Culex pipiens, and the Asian tiger mosquito Aedes albopictus also cluster closely to the A. gambiae and A. aegypti ACP receptors.
A heterologous receptor functional assay involving CHO-K1 cells was used to validate the cloned receptor as a bona fide ACP receptor. Indeed, the AedaeACPR-I was dose-dependently activated by AedaeACP (EC 50 = 1.025 × 10 −8 M) (Fig. 4a), confirming the proposed identity of the receptor based on phylogenetic analysis. Kinetic analysis of receptor activation demonstrated maximal luminescence response was evident over the first five seconds following application of the ACP peptide, indicative of an immediate and transient elevation of intracellular calcium levels elicited through activation of the ACP receptor (Fig. 4b). Our results also confirm the specificity of the ACP receptor for the ACP peptide alone (see Table 2), since no detectable luminescence indicative of receptor activation was observed in response to the closely related peptides, AedaeAKH and AedaeCRZ, or other tested peptides, specifically AedaeCAPA-1 and pyrokinin-1 (AedaePK1), which share no structural similarity to AedaeACP. Our findings in this study are consistent with past reports, since similar binding specificity of ACP receptors has been observed previously in T. castaneum, A. gambiae, and R. prolixus with EC 50 values reported in the low nanomolar range 17,25 . Additionally, previous research in the aforementioned insects have also observed that the AKH receptors are not activated by ACP or corazonin peptides, and similarly, the corazonin receptors are not activated by ACP or AKH peptides 17,25,43,44 . Thus, consistent with these previous observations, we determined that although these neuropeptide systems are structurally and evolutionarily related, they are indeed independent of one another and do not exhibit any cross talk in A. aegpyti. Notably, however, previous studies in B. mori have revealed that high concentrations of Bommo-ACP (previously referred to as AKH3) resulted in the activation of Bommo-AKHR whereas sensitivity to its natural AKH ligand was approximately 100-fold higher 23 . Similarly, high doses of the AKH peptides in B. mori, Bommo-AKH1 and Bommo-AKH2, were also found to activate putative B. mori ACPRs (A28 and A29), albeit at significantly higher concentrations than ACP 22 .
In comparison to the A. aegypti genome, a number of single nucleotide polymorphisms (SNP) were observed across the entire cDNA sequence. Of those occurring within the open reading frame, only one SNP (nucleotide 1924, found within the seventh exon, which corresponds to the C-terminus of the receptor) results in a different amino acid at residue Ile 472 , compared to the Asn 472 predicted by the A. aegypti genome. Modification of the isoleucine residue (Ile 472 ) obtained in our cDNA to the genome consistent asparagine residue (Asn 472 ) resulted in no change to receptor activation by its endogenous ACP ligand, as determined by equal luminescent response by both the cloned AedaeACPR-I and the mutated AedaeACPR-I-N472I (Fig. S2). No luminescence signals were Next, utilizing RT-qPCR, we investigated the molecular expression of the ACP signaling system during development and in individual tissues of adult A. aegypti. Although we identified three transcript variants, only the ACPR-I transcript yields a complete receptor protein which we functionally deorphanized, and so expression profiles were determined only for this transcript variant (see methods). Developmental expression profiling revealed enrichment of both ACPR-I (Fig. 5a) and ACP (Fig. 5b) transcripts following the transition from pupal to adult stages. In particular, one-day and four-day old male A. aegypti had the highest levels of ACP and ACPR-I transcript abundance. Similar findings for the ACP receptor were observed in R. prolixus 17,25 ; however, in contrast, ACPR transcript levels in T. castaenum were highest in late embryonic and early larval stages and decreased thereafter as the beetle progressed in development 17,25 . In A. aegypti, the observed enrichment of the ACP and ACPR-I transcripts could be indicative of a post-eclosion function for the ACP system. Spatial transcript expression profiles of A. aegypti ACPR aimed to reveal potential functional roles for ACP and its receptor by determining prospective target tissues. The ACP receptor (ACPR-I) was found to be significantly enriched in the abdominal ganglia of both male (Fig. 5c) and female (Fig. 5e) A. aegypti when compared relative to expression in the whole adult (males, p = 0.0025 and for females, p = 0.0016). ACPR-I transcript was also observed in the carcass, accessory reproductive tissue, testes, head and thoracic ganglia of adult male mosquitoes, although not significantly enriched as was found in the abdominal ganglia (Fig. 5c). Similarly, ACPR-I transcript was also found in the head, thoracic ganglia, hindgut, and accessory reproductive tissues of adult female A. aegypti (Fig. 5e). Enrichment of the ACP receptor transcript in the nervous system is not surprising, since expression of the transcripts encoding the ACP receptor and peptide have been described in the nervous system of other insects. Specifically, in fifth instar and adult R. prolixus, ACPR transcript was found to be enriched in the  25 . ACPR expression in T. castaneum was revealed to be greatest in the brain in comparison to the torso (i.e. body minus the head) 17 . AedaeACP transcript was detected in the central nervous system, where it was enriched in the brain and thoracic ganglia of male (Fig. 5d) and significantly enriched in the head (p = 0.0127) and thoracic ganglia (p = 0.004) of female mosquitoes (Fig. 5f). Consistent with our quantified ACP tissue-specific expression profile, ACP transcripts in A. aegypti and A. gambiae were detected solely in head and thorax body segments of adult mosquitoes 20,21 .
Expression of A. aegypti ACPR-I within the central nervous system suggests that ACP may be functioning as a neuromodulator and/or a neurotransmitter. This possibility is supported by the extensive varicose immunoreactive staining of ACP in the central nervous system of T. castaneum first instar larvae, where a neurosecretory role was suggested 17 . Specifically, immunoreactivity (IR) was observed in three to four neurons in the brain, the central brain neuropil with projections from the brain descending into the suboesophageal ganglion (SOG), thoracic ganglia, and abdominal ganglia, with no projections observed exiting the nervous system 17 . Immunocytochemistry using an antiserum against D. melanogaster AKH, which recognizes both AKH and ACP in A. gambiae and A. aegypti 20,21 , revealed immunoreactivity throughout the mosquito nervous system. In both A. aegypti and A. gambiae, AKH-like immunoreactivity was observed in two pairs of lateral neurosecretory cells in the brain, but was explained by the authors of this study to represent ACP-producing neurons, since AKH synthesis and storage is restricted to the corpus cardiacum 20,21 . Within the fused thoracic ganglia of adult AedaeAKH A. aegypti and A. gambiae, AKH-like immunoreactivity was observed within at least one to a few cells within the ventral region of the pro-and mesothoracic ganglia 20,21 . Thoracic extracts were negative for ACP-like activity in A. gambiae 20,21 , however, expression of the AKH transcript in adult male A. aegypti is absent in the head and thorax region 20,21 ; thus, it is unclear if the cells detected previously within the thoracic ganglia are ACP-or AKH-producing neurons in mosquitoes although our data indicate that the A. aegypti ACP transcript is prominently expressed in both the brain and thoracic regions of the nervous system. In R. prolixus, ACP-like immunoreactivity is observed in two bilaterally paired cell bodies at the anterior portion of the protocerebrum near the optic lobes, and one bilateral pair of cell bodies medially positioned in the protocerebrum 10 . Considering  the extensive immunohistochemical distribution of ACP throughout the nervous system of insects 16,19,20 , the prominent expression of ACP transcript in the brain and thoracic ganglia along with the significant enrichment of ACPR-I within the abdominal ganglia of adult A. aegypti strongly supports that ACP may be functioning centrally within the nervous system. In L. migratoria ACP was identified in the storage lobe of the CC, in contrast to the glandular lobe where AKH is found, suggesting synthesis of this neuropeptide within neurosecretory cells of the brain 18 . Furthermore, it was previously suggested that given ACPR was found in the CC/CA complex in R. prolixus, ACP may be involved in the regulation of other hormones in a manner similar to its distant vertebrate homolog, GnRH 25 . We also determined A. aegypti ACPR-I expression is not restricted to nervous tissue since transcript expression was detected in other tissues/organs including the female hindgut (Fig. 5e) and male carcass (Fig. 5c). ACPR expression in the hindgut, the primary site of reabsorption of ions and metabolites 45 , was unexpected since neither AKH nor CRZ have been found to regulate hydromineral balance. Thus, this possible function for ACP on the hindgut will require further investigation. Detection of the ACPR transcript in the carcass, which includes the body wall musculature and fat body, suggests that ACP and AKH may share a lipid mobilizing function. However, this possibility is unlikely since ACP was shown to have no influence on lipid or carbohydrate metabolism in female A. gambiae nor did it influence energy stores in male insects of L. migratoria or P. americana 46 . Interestingly, contrary to our predictions, both spatial and temporal expression profiles reveal greater expression of ACP and ACPR in adult males compared to females, which is consistent with an earlier microarray analysis in A. gambiae that determined higher ACP expression in adult males, compared to adult female and last instar larvae 47 . There is no clear explanation for such a sex-specific difference in ACP and ACPR transcript expression, however similar to our findings, male D. melanogaster express greater levels of the AKH receptor than their female counterparts 8 . ACPR-I transcript expression was also observed in the accessory reproductive tissues of both male and female A. aegypti. Expression of ACPR in reproductive tissue has also been observed in R. prolixus 25 . Perhaps, in addition to structural homology between ACPR and GnRHR, there is a yet undiscovered functional conservation between these two signaling systems. Furthermore, in Gryllus bimaculatus, pharmacologically elevating AKH titre through injections resulted in a significantly lower egg production 48 . In Caenorhabditis elegans, knockdown of the AKH-GnRH peptide and GnRH receptor resulted in reduced progeny in early worms 49 . Also, in Glossina morsitans, knockdown of the AKHR transcript resulted in higher levels of whole body lipids and, in pregnant flies, the inability to utilize lipid reserves resulted in delayed larval development and thus reproductive disruption 50 . Whether these effects on reproduction in these organisms are a direct result of signalling involving the AKH or GnRH receptor or AKH peptide remains unclear. Additionally, analysis of seminal fluid protein content of A. albopictus revealed the AKH peptide as one of the proteins transferred from males to female mosquitoes during mating 51 . Recently, CRZR transcript expression in R. prolixus was also observed in male and female reproductive tissues, which suggests some potentially overlapping reproductive target tissues in insects 44 .
Currently, no definitive function for ACP has been determined and functional studies in other insects have revealed that ACP does not influence energy mobilization and so does not duplicate the actions of AKH 10,46 . Additionally, ACP failed to increase heart-beat frequency, suggesting that the physiological actions of ACP does not mirror the most established function of CRZ 10 . Further studies are necessary to unravel the function of the ACP system, which could include methods aimed at knockdown of the ACP peptide or receptor.