BNGR-A25L and -A27 are two functional G protein–coupled receptors for CAPA periviscerokinin neuropeptides in the silkworm Bombyx mori

CAPA peptides, such as periviscerokinin (PVK), are insect neuropeptides involved in many signaling pathways controlling, for example, metabolism, behavior, and reproduction. They are present in a large number of insects and, together with their cognate receptors, are important for research into approaches for improving insect control. However, the CAPA receptors in the silkworm (Bombyx mori) insect model are unknown. Here, we cloned cDNAs of two putative CAPA peptide receptor genes, BNGR-A27 and -A25, from the brain of B. mori larvae. We found that the predicted BNGR-A27 ORF encodes 450 amino acids and that one BNGR-A25 splice variant encodes a full-length isoform (BNGR-A25L) of 418 amino acid residues and another a short isoform (BNGR-A25S) of 341 amino acids with a truncated C-terminal tail. Functional assays indicated that both BNGR-A25L and -A27 are activated by the PVK neuropeptides Bom-CAPA-PVK-1 and -PVK-2, leading to a significant increase in cAMP-response element–controlled luciferase activity and Ca2+ mobilization in a Gq inhibitor–sensitive manner. In contrast, BNGR-A25S was not significantly activated in response to the PVK peptides. Moreover, Bom-CAPA-PVK-1 directly bound to BNGR-A25L and -A27, but not BNGR-A25S. Of note, CAPA-PVK–mediated ERK1/2 phosphorylation and receptor internalization confirmed that BNGR-A25L and -A27 are two canonical receptors for Bombyx CAPA-PVKs. However, BNGR-A25S alone is a nonfunctional receptor but serves as a dominant-negative protein for BNGR-A25L. These results provide evidence that BNGR-A25L and -A27 are two functional Gq-coupled receptors for Bombyx CAPA-PVKs, enabling the further elucidation of the endocrinological roles of Bom-CAPA-PVKs and their receptors in insect biology.

In the silkworm B. mori, two splice variants of the capability (capa) gene, capa-a and capa-b, have been identified in silico. The capa-a variant encodes CAPA-PVK-1, CAPA-PVK-2, and CAPA-PK, whereas capa-b encodes identical CAPA-PVK-1 and -2 and a C-terminally extended form of CAPA-PK, showing many characteristics similar to those of the Manduca and Drosophila capa genes (30). Peptidomic analysis has shown that CAPA-PVK-1 has been detected in silkworm larvae and moths, whereas CAPA-PVK-2 has been found in silkworm eggs and larvae (31). Two orphan receptor genes, BNGR-A25 and -A27, have been identified in silico as putative CAPA-PVK receptors in the Bombyx genome (17,32). However, there are still no biochemical and pharmacological data to confirm the bioinformatics analysis results. The CAPA signaling is considered as important in the regulation of a wide range of physiological processes. Pairing of orphan receptors with the endogenous peptides is an important step to fully understand the neuropeptide-mediated downstream signaling cascades. In addition, the silkworm is an economically important insect, being a primary producer of silk, and meanwhile, as a model of Lepidoptera, the silkworm could serve as a good system for investigation of signaling and physiological functions of the CAPA peptide/receptor system. Thus, identification of the cognate receptors for Bombyx CAPA-PVKs by using a reverse physiological approach will pave the way to elucidate the possible roles in the regulation of insect fluid homeostasis, cardioacceleratory effect, developmental processes, feeding, mating, and behavior (4,(33)(34)(35).
In the present study, we have cloned the cDNAs encoding the putative CAPA peptide receptors, BNGR-A27 and two splice variants of BNGR-A25, from the brain of the silkworm B. mori and heterogeneously expressed them in human embryonic kidney 293 (HEK293) cells and Spodoptera frugiperda 21 (Sf 21) cells. Further functional and pharmacological characterization clarified the signaling pathways mediated by these putative CAPA-PVK receptors.

Cloning and expression of the putative Bombyx CAPA receptors BNGR-A25S, -A25L, and -A27
The Bombyx putative neuropeptide receptor genes BNGR-A25 and -A27 have been categorized into the same group of the CAPA-like receptors by genomic and phylogenetic analyses (17,32). The cDNA sequences encoding the full-length BNGR-A27 (GenBank TM accession number AB330448) and two isoforms of BNGR-A25 (GenBank TM accession number AB330446 and the sequence of BNGR-A25L assembled by analysis of the RNA-seq data (SRR4425250 and SRR4425251)) were obtained by RT-PCR using total RNA isolated from brain tissue of silkworm larvae. The full sequence of the predicted ORF of BNGR-A27 encodes a protein of 450 amino acids. We isolated two isoforms of BNGR-A25; the long isoform (BNGR-A25L) contains 418 amino acid residues with the intact C-terminal tail, whereas the short isoform (BNGR-A25S) consists of 341 amino acid residues with a truncated Cterminal tail. The 3Ј-RACE method was performed to obtain a 3Ј-RACE PCR fragment matching the truncated variant BNGR-A25S (supplemental Figs. S2 and S3). Bioinformatics analysis of the sequences revealed that the Bombyx BNGR-A25 genomic gene is composed of seven exons and six introns, and the C-terminally truncated BNGR-A25S is generated by an intron retention event, leading to the introduction of a premature stop codon (TAA) that truncates the protein at residue 341 (supplemental Fig. S4). Protein sequence alignment between BNGR-A25S, -A25L, and -A27 and CAPA-like receptors from D. melanogaster, A. gambiae, and P. xuthus was performed (Fig. 1A) Next, to verify correct expression and localization, BNGR-A25S, BNGR-A25L, and BNGR-A27 with an N-terminal FLAG or a C-terminal enhanced green fluorescent protein (EGFP) were constructed and expressed in both human embryonic kidney 293 (HEK293) cells and Spodoptera frugiperda 21 (Sf 21) cells. Western blot analysis indicated that BNGR-A25S, -A25L, and -A27 had similar total expression levels (supplemental Fig.  S1A). Confocal microscopy revealed that BNGR-A25L and -A27 were predominantly expressed and localized to the cell surface with some intracellular accumulation in the absence of the ligand in both HEK293 and Sf 21 cells, whereas BNGR-A25S was expressed but was predominantly distributed in the cytoplasm (Fig. 1B). The ELISA analysis confirmed that both BNGR-A25L and -A27 exhibited significant cell-surface expression, whereas BNGR-A25S had almost no cell-surface expression in HEK293 cells and significantly decreased cell-surface expression in Sf21 cells compared with BNGR-A25L ( Fig. 1C and supplemental Fig.  S1D). Further colocalization experiments demonstrated that extensive BNGR-A25S was observed within the endoplasmic reticulum, but obvious BNGR-A25S was retained in the Golgi compartments (supplemental Fig. S1, B and C).

BNGR-A25L and -A27 are specifically activated by Bom-CAPA-PVK-1 and Bom-CAPA-PVK-2
The cyclic AMP-response element (CRE)-driven reporter gene-based functional assay was initially developed for determining the functional activity of G s -and G i -coupled GPCRs. However, previous studies showed that G q -coupled receptors could induce CRE-driven reporter gene transcription through a G q -dependent PKC/CREB cascade (36,37). Our recent study (38) demonstrated that G q -coupled Bombyx diapause hormone

Identification of Bombyx CAPA periviscerokinin receptors
receptor is activated to induce a significant increase in CREdriven luciferase activity in a G q inhibitor-sensitive manner. Therefore, we employed this CRE-driven luciferase assay combined with specific inhibitors to evaluate the signaling activities of putative CAPA receptors. HEK293 cells stably expressing FLAG-tagged BNGR-A25S, -A25L, and -A27, respectively, and pCRE-luc were established to determine the intracellular levels of cAMP. Upon stimulation with different concentrations of Bom-CAPA-PVK-1, -PVK-2, and -PK, BNGR-A25L and -A27 were activated, inducing a significant increase in luciferase activity in HEK293-A25L and HEK293-A27 cells by PVK-1 and PVK-2, whereas BNGR-A25S exhibited no response upon treatment with PVK-1, PVK-2, or PK (Fig. 2, A-C). Further investigation demonstrated that both BNGR-A25L and -A27 exhibited no response upon stimulation by other Bombyx neuropeptides, including adipokinetic hormone (AKH), corazonin, tachykinin; vice versa, Bom-CAPA-PVK-1 and -PVK-2 failed to activate other Bombyx neuropeptide receptors, such as adipokinetic hormone receptor (AKHR), corazonin receptor, and tachykinin receptor (Fig. 2, E and F). However, a direct cAMP measurement assay by ELISA illustrated that PVK-1 and PVK-2 failed to induce significant cAMP production in HEK293-A25L or HEK293-A27 cells (Fig. 2D). These findings suggest that G q but not G s protein is likely to be involved in the BNGR-A25L/ A27-mediated signaling (38 -40).
To further assess the functional activity of BNGR-A25S, -A25L, and -A27, HEK293 cells and Sf 21 cells expressing BNGR-A25S, -A25L, and -A27 were used to determine the intracellular Ca 2ϩ mobilization in response to agonists. PVK-1 and PVK-2 elicited a rapid increase of intracellular Ca 2ϩ in a dose-dependent manner with EC 50 values of 16.03 and 4.09 nM (Fig. 3, A, B, and D), respectively, in both HEK293-A25L and Sf 21-A25L cells, but not in HEK293-A25S and Sf 21-A25S cells. Similarly, HEK293-A27 and Sf 21-A27 cells also could be stimulated by PVK-1 and PVK-2 and elicited a rapid increase of intracellular Ca 2ϩ in a dose-dependent manner with EC 50 values of 7.29 and 3.69 nM (Fig. 3, C and E), respectively. In contrast, Bom-CAPA-PK could not induce Ca 2ϩ mobilization in both BNGR-A25L-and BNGR-A27-expressing cells. Taken together, our results suggest that BNGR-A25L and -A27 were specifically activated by Bom-CAPA-PVK-1 and -PVK-2.

Bombyx CAPA receptors activate ERK1/2 signaling and undergo a rapid internalization upon agonist stimulation
The phosphorylation of ERK1/2 has long been used to measure the functional outcome of GPCR activation (1,41). We therefore assessed the ERK1/2 activation by BNGR-A27. As shown in Fig. 4 (A and B), CAPA-PVK-1 and -PVK-2 but not CAPA-PK elicited transient activation of ERK1/2 in both HEK293-A25L/Sf 21-A25L cells and HEK293-A27/Sf 21-A27 cells. In addition, upon agonist binding and activation, GPCRs undergo internalization from the cell surface to the cytoplasm, which is a well-characterized phenomenon (42). To visualize the internalization of BNGR-A25L and -A27, a plasmid vector to express them fused with EGFP at the C-terminal end was constructed. Observation with confocal microscopy revealed that upon activation by PVK-1 or PVK-2, BNGR-A25L and -A27 underwent a rapid and dramatic redistribution from the cell surface to the cytoplasm (Fig. 5A). This was also confirmed by quantitative ELISA data (Fig. 5B). Moreover, in an arrestin recruitment assay, using fusion expression of arrestins with EGFP at the C terminus, PVK-1 induced a significant translocation of ␤-arrestin2 and Bmkurtz, a novel nonvisual arrestin from Bombyx, from the cytoplasm into the plasma membrane, whereas only a small amount of ␤-arrestin1 was recruited to the plasma membrane in response to PVK-1 (Fig. 5C). Collectively, the current results suggest that both BNGR-A25L and -A27 are functional receptors specific for B. mori neuropeptides CAPA-PVK-1 and -PVK-2. We therefore designated them as Bombyx CAPA receptors (Bom-CAPA-PVK-Rs).

Activation of Bombyx CAPA receptor via a G q -dependent pathway
Bom-CAPA-PVK-R was activated to elicit intracellular Ca 2ϩ mobilization in both HEK293-A25L/HEK293-A27 cells and Sf 21-A25L/Sf 21-A27 cells in a dose-dependent manner. Additional functional assays with specific inhibitors were performed to assess the detailed Bom-CAPA-PVK-mediated signaling pathways. Pretreatment with the G q inhibitor UBO-QIC (43), PLC inhibitor U73122, extracellular calcium chelator EGTA, and intracellular calcium chelator BAPTA-AM (44) led to a significant reduction in agonist-mediated intracellular Ca 2ϩ . Spaces have been introduced to optimize alignment. The seven transmembranes ␣-helices are indicated by TM I to TM VII. The black area with white lettering indicates non-homologous residues; the white area with black lettering indicates identical residues; the gray area indicates the most frequent residues; the red area indicates strongly similar residues; and the blue area indicates weakly similar residues. B, HEK293 cells and Sf21 cells expressing BNGR-A25S-EGFP, -A25L-EGFP, and -A27-EGFP were examined by confocal microscopy. C, ELISA analysis expression of BNGR-A25S, -A25L, and -A27. HEK293 cells and Sf21 cells were transiently transfected with FLAG-tagged BNGR-A25S (light gray column), FLAG-tagged BNGR-A25L (dark gray column), FLAG-tagged BNGR-A27 (black column), or empty vector (white column) and subsequently detected by the M2 anti-FLAG monoclonal antibody via an ELISA assay. Error bars, S.E. for three independent experiments. One-way ANOVA with Tukey's post hoc test revealed differences from basal (control) (*, p Ͻ 0.05; ***, p Ͻ 0.001; ns, not significant). All experiments were repeated independently at least three times with similar results.

Identification of Bombyx CAPA periviscerokinin receptors
mobilization. By contrast, L-type calcium channel blocker nifedipine and G ␤␥ inhibitor gallein (45) showed no effect on intracellular Ca 2ϩ mobilization induced by PVK-1 (Fig. 7A). Moreover, the G q inhibitor UBO-QIC (43) and G i inhibitor pertussis toxin (PTX) were also used to characterize the signaling pathways involved in the activation of ERK1/2 or CRE-driven luciferase expression in HEK293-A25L and HEK293-A27 cells. Bombyx CAPA receptor-induced ERK1/2 phosphorylation and luciferase activity were significantly blocked by the G q inhibitor UBO-QIC but could not be blocked by the G i inhibitor PTX (Fig. 7, B and C). In addition, pretreatment with cholera toxin (CTX), an activator of G s protein, caused a significant increase in CRE-driven luciferase activity in both vehicle-and PK-treated HEK293 cells, and stimulation with PVK-1 and PVK-2 induced further increase in lucifer- F, CRE-driven luciferase activities in AKHR/pCRE-Luc-, corazonin receptor (CrzR)/pCRE-Luc-, and tachykinin receptor (TKR)/pCRE-Luc-transfected HEK293 cells were measured in response to different B. mori neuropeptides. Error bars, S.E. for three independent experiments. One-way ANOVA with Tukey's post hoc test revealed differences from vehicle (control) (***, p Ͻ 0.001). All experiments were repeated independently at least three times with similar results.

Identification of Bombyx CAPA periviscerokinin receptors
ase activity (Fig. 7C). Taken together, the current results provide strong evidence that Bombyx CAPA receptor is specifically activated by Bom-CAPA-PVK-1 and -PVK-2, via a G q -dependent pathway, leading to the PLC-Ca 2ϩ -PKC signaling cascade.

Isoform BNGR-A25S is not a functional CAPA receptor but acts as a dominant negative for BNGR-A25L
Alternative splicing-generated GPCR variants result in precisely regulated differences in expression and trafficking, ligand-binding properties, signaling pathways, and G proteincoupling efficiency (46). As shown in Figs. 1B, 2A, 3A, and 5D, when expressed alone, BNGR-A25S is certainly not a functional CAPA receptor. We therefore assess the possible effects of BNGR-A25S on expression and signaling when co-expressed with BNGR-A25L and BNGR-A27. Results derived from confocal microscopy and quantitative ELISAs showed that BNGR-A25L and -A27 are predominantly localized at the plasma membrane; in contrast, BNGR-A25S exhibits preferential retention in intracellular com-

Identification of Bombyx CAPA periviscerokinin receptors
partments (Fig. 1, B and C). However, co-expression of BNGR-A25S resulted in intense cytoplasmic accumulation of BNGR-A25L, but not BNGR-A27 (Fig. 8, A and B). Additionally, when co-expressed with BNGR-A25S, BNGR-A25L, but not BNGR-A27, exhibited a significant reduction in ERK phosphorylation and CRE-driven luciferase activity in response to the stimulation of Bom-CAPA-PVKs (Fig. 8, C and D). Taken together, these results suggest that BNGR-A25S is more likely to play a dominant-negative role in the regulation of the cell-surface expression and signaling of canonical BNGR-A25L.

Discussion
The CAPA signaling system is considered a potential target for medical importance and crop pest control. Therefore, iden-  A and B, PVK-1-and PVK-2-induced activation of ERK1/2. ERK1/2 phosphorylation of HEK293 cells or Sf21 cells transiently transfected with BNGR-A25L or BNGR-A27 and treated with 100 nM PVK-1, PVK-2, and PK for 5 min was assessed by Western blotting. The pERK was quantified by densitometry and expressed as a percentage of the total ERK. Error bars, S.E. for three independent experiments. One-way ANOVA with Tukey's post hoc test revealed differences from basal (control) (**, p Ͻ 0.01; ns, not significant). All experiments were repeated independently at least three times with similar results. A, HEK293 cells transiently transfected with A25L-EGFP or A27-EGFP were stimulated with 100 nM PVK-1, PVK-2, and PK for 30 min and then examined by confocal microscopy. B, HEK293 cells transfected with FLAG-A25L or FLAG-A27 were stimulated with 100 nM PVK-1, PVK-2, and PK for 30 min, and the internalization of BNGR-A25L or -A27 was determined by ELISA. C, PVK-1 mediated recruitment of ␤-arrestin1, ␤-arrestin2, and BmKurtz in BNGR-A25L-or BNGR-A27-expressing HEK293 cells. HEK293 cells co-transfected with FLAG-A25L or FLAG-A27 and ␤-arrestin1-EGFP, ␤-arrestin2-EGFP, or Bombyx Kurtz-EGFP were stimulated with PVK-1 (100 nM) for 5 min and then examined by confocal microscopy. The squares in red are amplified toward the right of each corresponding image. Error bars, S.E. for three independent experiments. One-way ANOVA with Tukey's post hoc test revealed differences from basal (control) **, p Ͻ 0.01; ***, p Ͻ 0.001; ns, not significant). All experiments were repeated independently at least three times with similar results.

Identification of Bombyx CAPA periviscerokinin receptors
tification of its receptors is essentially required for elucidation of the signaling mechanism(s) involved in the regulation of fluid homeostasis and even for further development of selective and environmentally friendly pest-control insecticides. Since the identification of the first CAPA receptor from D. melanogaster (9,10), CAPA receptors have been subsequently found in representatives of the major orders, including Hymenoptera, Coleoptera, and Lepidoptera (47). CAPA receptors have also been identified in important vectors of medical and veterinary pathogens, such as different mosquito species, kissing bug, and tick (12-14, 48, 49), but not in A. aegypti (47). Genomic data mining and phylogenetic analysis suggested the presence of two paralogs of CAPA receptors in the genomes of M. sexta and the honey bee A. mellifera (16,18). Our previous study indicated that in the B. mori genome, of 46 putative neuropeptide receptor genes, two orphan GPCR genes, BNGR-A25 and -A27, have been identified as putative CAPA peptide receptors (32). In this study, we have cloned three paralogs of putative CAPA receptors, BNGR-A25S, -A25L, and -A27, from the silkworm B. mori. Using both HEK293 and Sf21 cells, after transfection with expression constructs, BNGR-A25L and -A27 exhibited high-level expression and correct localization on the cell surface. BNGR-A25L and -A27 were specifically activated in intracellular Ca 2ϩ mobilization and ERK1/2 phosphorylation by Bombyx CAPA-PVK-1 and CAPA-PVK-2, but not by Bombyx CAPA-PK and other neuropeptides, including Bombyx corazonin and tachykinin. Moreover, the synthetic Bom-CAPA-PVK-1 peptide N-terminally tagged with a 5-FAM-based binding assay demonstrated that Bom-CAPA-PVK-1 directly bound to BNGR-A25L and -A27. In addition, our data further showed that Bom-CAPA-PVK-R underwent a rapid internalization from the cell surface to the cytoplasm via a ␤-arrestindependent pathway following exposure to agonists. Ligand-induced receptor internalization has been recognized as a key element to regulate the strength and duration of receptor-mediated cell signaling and to directly reflect the activation of the receptor (51). Collectively, our data provide biochemical and pharmacological evidence to confirm that Bom-CAPA-PVK-1 and -PVK-2 are the canonical ligands for BNGR-A25L and -A27. Based on phylogenetic analysis and our current results, therefore, we suggest that BNGR-A25L and -A27 are paired with neuropeptides Bom-CAPA-PVK-1 and -PVK-2 and are better named Bom-CAPA-PVK-R1 and Bom-CAPA-PVK-R2, respectively.
Previous investigation of the myomodulatory activity of CAPA-PVKs using the cockroach hyperneural muscle bioassay showed that CAPA-PVK-induced muscle contractions were essentially dependent on the Ca 2ϩ influx through both voltagedependent nifedipine-sensitive and non-voltage (receptor)operated Ca 2ϩ channels (27). In the fruit fly, the Dme-CAPA-PVKs and Mas-CAPA-PVK-2 have been shown to activate tubule fluid secretion by increasing intracellular Ca 2ϩ concentration in the principal cells through both IP3 receptor-mediated Ca 2ϩ release from the endoplasmic reticulum and L-type voltage-dependent Ca 2ϩ channels (24 -26). In the current study, our data demonstrated that upon agonist stimulation, Bom-CAPA-PVK-R1 and -R2 evoked a rapid and transient rise of intracellular Ca 2ϩ in a dose-dependent manner in both HEK293 and Sf 21 cells (Fig. 3). Data derived from experiments with different inhibitors showed that G q , PLC, IP3 receptor, and Ca 2ϩ channels located on the plasma membrane are involved in Bom-CAPA-PVK-R-mediated Ca 2ϩ mobilization. However, the G ␤␥ subunit and the L-type Ca 2ϩ channel are not involved in the Bom-CAPA-PVK-R-triggered Ca 2ϩ signaling cascade (Fig. 7A). Previous studies indicated that CAPA-PVKs activated Ca 2ϩ influx through nifedipine-sensitive L-type Ca 2ϩ channels in the plasma membrane (24,25,52). A recent study using a genetically encoded aequorin-based luminescent Ca 2ϩ reporter targeted to only tubule principal cells demonstrated that CAPA peptides trigger intracellular Ca 2ϩ mobilization in a biphasic manner: a fast primary response followed by a slow secondary response (53). Taken together, our results suggest that Bom-CAPA-PVK-R1 and -R2 are G q -coupled receptors, upon binding with agonist; Bom-CAPA-PVK-R1 and -R2 evoke intracellular Ca 2ϩ mobilization through the IP3 receptor in the endoplasmic reticulum and Ca 2ϩ channels in the plasma membrane. However, there need to be more experiments to clarify which Ca 2ϩ channels in the plasma membrane are involved in the Bom-CAPA-PVK-R-induced Ca 2ϩ influx.
Recent studies using emerging genetic sequencing technologies have demonstrated that a single gene could generate different transcriptional isoforms because of a process known as alternative splicing, allowing the generation of several structurally and functionally distinct protein isoforms (54 -56). Of 353 GPCRs detected from human airway smooth muscle, 192 GPCRs have, on average, five different expressed receptor isoforms due to splicing events (57). There are four B. mori pheromone biosynthesis-activating neuropeptide receptor variants and up to 25 alternative splicing proteins of the mouse Oprm1 gene (58). Alternative splicing also could act as a major mechanism that modulates gene expression and function of GPCRs in different species (46, 59 -61). The C-terminally truncated splice variants of the -opioid receptor and human somatostatin receptor subtype 5 (SST5) have been found to be functionally active (62,63). However, the majority of truncated GPCR splice variants serve as dominant-negative mutants in the regulation of the cellsurface expression and signaling of canonical receptors (64). In the current study, we identified BNGR-A25S as a splice variant with a truncated C-terminal tail generated by an intron retention event. Further investigation indicated that BNGR-A25S is not a functional receptor but is likely to act as a dominant-negative mutant

Identification of Bombyx CAPA periviscerokinin receptors
for BNGR-A25S. However, additional studies are required to address the physiological roles of this splice variant BNGR-A25S in different tissues and different development stages.
In conclusion, we have paired Bom-CAPA-PVK-1 and -PVK-2 as specific canonical ligands for the Bombyx orphan receptors BNGR-A25L and -A27. Upon direct interaction with

Identification of Bombyx CAPA periviscerokinin receptors
neuropeptide ligands, the Bom-CAPA-PVK receptors couple to G q protein, triggering intracellular Ca 2ϩ mobilization through the IP3 receptor in the endoplasmic reticulum and Ca 2ϩ channels in the plasma membrane. Exposure to agonists also elicits ERK1/2 phosphorylation and receptor internalization. In addition, the Bombyx orphan receptor as a splice variant with a truncated C-terminal tail, BNGR-A25S, exhibits no functional activity when expressing alone but shows dominant-negative effects on cell-surface expression and signaling of co-expressed BNGR-A25L in response to Bom-CAPA-PVKs (Fig. 8). Our present study provides the first in-depth insights into Bom-CAPA-PVK-Rs-mediated signaling and will be helpful for further elucidation of the roles of Bom-CAPA-PVK-Rs in the regulation of fundamental physiological processes.

Molecular cloning and plasmid construction
Total RNA was isolated from the brain of B. mori larvae using the RNAiso Plus reagent (Takara BIO, Kusatsu, Japan), following the manufacturer's instructions. The cDNA was synthesized using a PrimeScript first strand cDNA synthesis kit (Takara BIO) according to the manufacturer's instructions. To amplify the full-length sequences encoding BNGR-A25S, -A25L, and -A27, six pairs of primers for each receptor were designed based on the sequences of GenBank TM accession numbers AB330446 and AB330448 and the sequence of BNGR-A25L assembled by analysis of the RNA-seq data (SRR4425250 and SRR4425251). As shown in Table 1 and supplemental Table S2, all primers were used in this paper. To express in insect cells, corresponding PCR products were cloned into pBmIE1-FLAG and pBmEGFP-N1, which replaced the corresponding sites of pCMV-FLAG and pEGFP-N1, as described previously (65). EGFP was fused at the C-terminal end of receptors.

Cell culture and transfection
The human embryonic kidney cell line 293 (HEK293) was maintained in DMEM supplemented with 10% FBS, 100 units/ml penicillin, 100 g/ml streptomycin, and 4 mM L-gluta-

Identification of Bombyx CAPA periviscerokinin receptors
mine (Invitrogen) at 37°C in a humidified incubator containing 5% CO 2 . The Sf 21 cell line was maintained in TC100 insect medium supplemented with 10% FBS, 100 units/ml penicillin, and 100 g/ml streptomycin at 28°C. The BNGR-A25S, -A25L, and -A27 cDNA plasmid constructs were transfected into HEK293 and Sf 21 cells using Lipofectamine 2000 and X-tremeGENE HP, respectively, according to the manufacturer's instructions. Two days after transfection, stably expressing cells were selected by the addition of 800 mg/liter G418.

cAMP accumulation measurement
The pCRE-Luc was performed to investigate the intracellular levels of cAMP as a reporter gene system that contains the cAMP-response elements and firefly luciferase coding region as described before (65). Cells transiently or stably co-expressing BNGR-A25S, -A25L, or -A27 with pCRE-Luc were grown to 90 -95% confluence in 96-well plates and stimulated with different concentrations of Bom-CAPA-PVK-1, -PVK-2, or -PK in DMEM without FBS for 4 -6 h at 37°C. Luciferase activity was detected by a firefly luciferase assay kit (KenReal, Shanghai, China). First, each well of cells was lysed by 40 l of lysis buffer for 15 min, and then 30 l of the whole-cell lysate were mixed with 15 l of detection buffer for 10 min. Finally, the CREluciferase activity was measured using a 96-well TopCount NXT scintillation and luminescence counter (PerkinElmer Life Sciences). The direct cAMP concentration was assessed using a commercially available cAMP detection kit (R&D Systems, Minneapolis, MN).

Intracellular calcium measurement
The fluorescent Ca 2ϩ indicator Fura-2/AM was performed to detect the intracellular calcium flux, as described previously (39). Briefly, the stably or transiently expressing BNGR-A25S, -A25L, or -A27 HEK293 and Sf 21 cells were washed twice with PBS and suspended at 5 ϫ 10 6 cells/ml in Hanks' balanced salt solution for HEK293 cells or Hepes-buffered medium for Sf 21 cells. The cells were then loaded with 3 M Fura-2/AM for 30 min. Cells were washed twice in Hanks' solution or Hepes-buffered medium. Then cells were stimulated with the indicated concentrations of Bom-CAPA-PVK-1, -PVK-2, and -PK. Finally, intracellular calcium flux was measured for 60 s by the ratio of excitation wavelengths at 340 and 380 nm in a fluorescence spectrometer (Infinite 200 PRO, Tecan, Männedorf, Switzerland). All experiments of measuring Ca 2ϩ mobilization were repeated independently at least 3-5 times.

Measurement of cell-surface receptor
ELISA was performed for quantification of receptors on the cell surface as described before (66). First, cells were transfected with various target GPCRs. 48 h after transfection, cells were stimulated with or without corresponding ligands for the indicated time. Next, cells were fixed with 4% formaldehyde for 10 min at room temperature and blocked for 60 min with 1% bovine serum albumin in TBS (20 mM Tris, 150 mM NaCl, pH 7.5). Then the cells were incubated for 2 h with a 1:2000 dilution of the mouse anti-FLAG M2 monoclonal antibody (Sigma) and washed three times with TBS followed by incubation with HRPconjugated goat anti-mouse (1:2000 in 1% BSA/TBS) for 60 min. 150 l of HRP substrate (Sigma) were added to each well and incubated at 37°C for 20 -30 min. The reactions were stopped by adding 100 l 1% SDS, and the samples were measured at 405 nm using a Bio-Rad microplate reader.
added to the cells incubated on ice for 60 -90 min. The cells were washed three times in FACS buffer and suspended in FACS buffer with 1% paraformaldehyde for 15 min. The binding ability of PVK-1 with BNGR-A25S, -A25L, or -A27 was determined by measuring the fluorescence intensity of 5-FAM and presented as a percentage of total binding.

Localization and translocation assay by confocal microscopy
For the expression and translocation analysis of receptor, HEK293 cells and Sf 21 cells transiently or stably expressing BNGRs-EGFP were seeded into glass coverslip 6-well plates (39). After treatment with various ligands for the indicated time or concentration, the cells were washed with PBS and fixed with 4% paraformaldehyde in PBS for 10 min at room temperature. The cells were mounted in 50% glycerol and visualized by fluorescence microscopy using a Zeiss LSM510 laser-scanning confocal microscope attached to a Zeiss Axiovert 200 microscope and linked to a LSM5 computer system. Excitation was performed at 488 nm, and the fluorescence detection used a 505-530-nm bandpass filter.

Immunoblot analysis
To examine the phosphorylated form of ERK, HEK293 cells or Sf 21 cells expressing BNGR-A25S, -A25L, or -A27 were incubated for the indicated times with different concentrations of Bom-CAPA-PVK-1, -PVK-2, and -PK. Subsequently, the cells were lysed by lysis buffer (Beyotime) containing protease inhibitor (Roche Applied Science) at 4°C for 30 min on a rocker and then scraped. Afterward, proteins were electrophoresed on a 10% SDS-polyacrylamide gel and then transferred to PVDF membranes. The membranes were blocked with 5% skim milk and then probed with rabbit monoclonal anti-phospho-ERK1/2 antibody (1:2000) (Cell Signaling Technology), followed by detection using HRP-conjugated goat anti-rabbit IgG (Beyotime). The membranes were stripped and reprobed using a polyclonal anti-␤-actin/anti-␤-tubulin (1:5000) (Beyotime) or anti-ERK1/2 antibody (1:2000) (Cell Signaling Technology) as a control for protein loading. Immunoreactive bands were detected with an enhanced chemiluminescent substrate (Cell Signaling Technology), and the membranes were then scanned using a Tanon 5200 chemiluminescent imaging system (Tanon, Shanghai, China). The intensities of bands were quantified by the Bio-Rad Quantity One imaging system.