Inhibitory Role of the Mud Crab Short Neuropeptide F in Vitellogenesis and Oocyte Maturation via Autocrine/Paracrine Signaling

Neuropeptides, in addition to their classical role in the nervous system, act on intraovarian factors to regulate reproductive functions in vertebrates. However, this function of neuropeptides has not been characterized in crustaceans. Short neuropeptide F (sNPF), a highly conserved invertebrate neuropeptide, has been reported to be involved in feeding, metabolism, and in differentiation processes including reproduction. Although sNPF and its receptor (sNPFR) have been detected in the ovary in different species, ovarian colocalization of sNPF/sNPFR has not been investigated. In this study, we identified Scylla paramamosain (mud crab) sNPF (Sp-sNPF) as an endogenous ligand for the S. paramamosain orphan G protein-coupled receptor NPY2R in mammalian cell line HEK293T. We designated this receptor as Sp-sNPFR. RNA in situ hybridization in pre-vitellogenic ovary and reverse transcription-PCR on isolated denuded oocytes and follicle layers showed that Sp-sNPF was exclusively localized to the follicle cells, whereas Sp-sNPFR was detected in both follicle cells and oocytes. We also found that Sp-sNPF partly suppressed spontaneous maturation of denuded oocytes and caused intracellular cAMP accumulation and Ca2+ mobilization. Moreover, injection of synthetic Sp-sNPF peptides inhibited the expression of vitellogenin and vitellogenin receptor genes in vivo. These combined results suggest for the first time that Sp-sNPF may have inhibitory functions in vitellogenesis and oocyte maturation possibly via the autocrine/paracrine pathway.


INTRODUCTION
Neuropeptides are widespread in the nervous system of animals (1). Neuropeptides are secreted from neural cells and act as neuromodulators or neurotransmitters in the nervous system (2). Furthermore, neuropeptides are secreted from neuroendocrine cells, transported through blood to peripheral organs, and function as neurohormones (3). They also play a pivotal role in regulating reproduction (4). One such example is that of gonadotropin-releasing hormone (GnRH), a key decapeptide hormone involved in the regulation of reproduction which is secreted by the hypothalamus and transported via the hypophysial portal system to stimulate the production of luteinizing hormone (LH) and follicle-stimulating hormone (FSH), which further act on the ovary (5,6). In addition to its involvement in this classical hypothalamic-pituitary-gonadal (HPG) reproduction axis in vertebrates, GnRH also plays a role as a neuroendocrine factor by directly binding to its receptor (GnRHR) in the ovary (7,8). Mostly, neuropeptides are released by the nervous system and regulate reproduction directly or indirectly (8,9). However, neuropeptides can also act as intraovarian regulatory factors. Intraovarian GnRH/GnRHR system has been suggested to play an autocrine/paracrine role in gonadal steroidogenesis, cell proliferation, and apoptosis (8,(10)(11)(12)(13). In addition to GnRH/GnRHR, two key upstream regulation neuropeptide systems, gonadotrophin-inhibitory hormone (GnIH)/GnIH receptor (GnIHR) and kisspeptin/kisspeptin receptor, in the vertebrate reproduction axis also exist in ovary (14)(15)(16)(17)(18)(19)(20)(21)(22)(23). The regulation of intraovarian neuropeptides has been well studied in vertebrates (7,8,18,24,25). However, a detailed knowledge about the intraovarian neuropeptides acting on ovary in invertebrates has not been reported.
Short neuropeptide F (sNPF), a homolog of vertebrate neuropeptide Y (NPY), is a C-terminally conserved neuropeptide in invertebrates. sNPF belongs to FMRFa-like peptides (FLPs) with the RFamide conserved motif (26) and has been widely studied in insects, especially in Drosophila (27). The first sNPF isolated from the midgut of American cockroach, Periplaneta americana, was postulated to play a role in digestion (28). Studies have mainly focused on the role of sNPF in regulation of feeding and metabolism (29)(30)(31)(32)(33)(34). However, a role for sNPF in reproduction has also been identified. For instance Led-NPF-1, sNPF peptide from Leptinotarsa decemlineata (Colorado potato beetle) and Scg-NPF, a Locusta migratoria sNPF peptide, were shown to accelerate egg development in L. migratoria (34)(35)(36)(37). However, their regulatory role in signaling is not known. Drosophila sNPF and its receptor (sNPFR) have been considered as upstream regulators of insulin signaling pathway (30). In Drosophila, sNPF and sNPFR1 proteins were detected in the larval brain where sNPFR1 was localized to the brain insulin producing cells (IPCs) and sNPFnergic neurons were found adjacent to these IPCs (30,38). sNPF peptides stimulate insulin-like peptides (ILPs) synthesis in IPCs and affect insulin signaling (30). In insects, insulin signaling pathway regulates developmental and differentiation functions such as reproduction (39,40). Juvenile hormone (JH) and 20-hydroxyecdysone (20E) were found to act as the insect gonadotropic hormones and to be regulated by insulin signaling by a complex physiological regulatory network (39,40). sNPF peptides can directly regulate the production of JH. In the silkworm, Bombyx mori, sNPF peptides localize to the corpora cardiaca and inhibit JH biosynthesis in the corpora allata (41,42). Moreover, sNPF was speculated as a neurohormone directly functioning on the ovary of red imported fire ant, Solenopsis invicta Buren as the S. invicta sNPF receptor (Si-sNPFR) protein was detected in oocytes by immunolocalization (43). These findings suggest that sNPF may serve as neuro-regulator in reproduction. Whether or not sNPF directly stimulates ovary is not known.
However, the reproductive function of sNPF in crustaceans is not known. S. paramamosain sNPF (Sp-sNPF) transcript was identified to encode a 126 amino acids (aa) precursor peptide with a 25 aa signal peptide followed by three mature peptides, each 9-12 aa in length and each containing a XPXRLRFamide (X represent variable residues) conserved motif (44). Interestingly, in our previous study, we found that Sp-sNPF transcript was highly expressed in the ovary in addition to the nervous system (44). We hypothesized that sNPF may act as intraovarian regulatory factor similar to vertebrate neuropeptides. In this study, we investigated the potential intraovarian autocrine/paracrine role of Sp-sNPF by receptor identification, ligand-receptor localization, signal transduction, and by biological relevance.

Molecular Cloning and Plasmid Construction
Total RNA from the crab ovary was isolated with the Trizol Reagent (Invitrogen) according to the manufacturer's instructions. The cDNA was synthetized using the RevertAid First Strand cDNA Synthesis Kit (Thermo) as per manufacturer's instructions. To clone the entire coding region of Sp-sNPFR gene, primers ( Table 1) were designed based on the sequence in the transcriptome databases of brain (SRA accession: SRP068003) and muscle (SRA accession: SRP111448) from S. paramamosain. PCR product was sequenced prior to inserting into the HindIII and BamHI sites of the pEGFP-N1 vector. The construct was sequenced to validate the sequence and orientation.

Primers and Probes
Sequences

Confocal Microscopy
For detection of sNPFR expression on the cell surface, HEK293T transiently transfected with Sp-sNPFR/pEGFP-N1 plasmid for 24 h were stained with the cell membrane probe DiI (Beyotime) at 37 • C for 10 min and fixed with 4% paraformaldehyde (PFA) for 15 min. Cells were further incubated with a nuclear dye DAPI (Invitrogen) for 10 min. Stained cells were mounted in 50% glycerol and imaged using a two-photon laser confocal fluorescence microscopy (LSM 780 NLO, Zeiss).

Activation of Signaling
HEK293T transiently co-transfected with Sp-sNPFR/pEGFP-N1 and dual-luciferase reporter system vectors were grown to 90% confluency in a 24-well plate. Cells were first treated with synthetic Sp-sNPF peptide mixture and the mixture containing 9 synthetic FLPs (GL Biochem,

In Situ Hybridization
Crab ovaries in a pre-vitellogenic stage were dissected, fixed in 4% PFA, and prepared for in situ hybridization as described (53). Primers, listed in Table 1, spanning 381 and 250 bp nucleotides of Sp-sNPF and Sp-sNPFR cDNA, respectively, were used for the preparation digoxigenin-labeled cRNA riboprobes using a DIG-RNA Labeling Kit (Roche). Hybridization was performed at 57 • C for 16 h according to the manufacturer's instructions.

Isolation and Incubation of Crab Ovary
Ovaries (late vitellogenic stage) with fully grown (FG) oocytes were removed from female crabs and placed in 10 cm culture dish containing calcium-free crab saline (440 mM NaCl, 11.3 mM KCl, 26 mM MgCl 2 , 10 mM Hepes, 10 mM Glucose, pH 7.4 adjusted with NaOH) (55). The follicle layers were carefully separated with fine forceps under stereoscope (Leica). To determine the expression of Sp-sNPF and Sp-sNPFR gene, total RNA from intact ovary, denuded oocytes, and follicle layers were extracted with the Trizol Reagent (Invitrogen). cDNA was synthesized using TransScript II One-Step gDNA Removal and cDNA Synthesis SuperMix kit (TransGen Biotech) according to the manufacturer's protocols. The cDNA templates were used for PCR amplification using Sp-sNPF and Sp-sNPFR primers ( Table 1). PCR products were analyzed by 1.5% agarose gel electrophoresis.

Intracellular Calcium Measurement
The denuded oocytes were washed thrice after they were separated from the ovaries with FG oocytes and were loaded with 1 µM Fluo-4 AM (Molecular Probes) for 40 min in 25 • C. Oocytes were washed thrice, resuspended in calcium-free crab saline, and incubated for 30 min at 25 • C. Calcium flux was measured in oocytes stimulated with Sp-sNPF peptides using excitation at 486 and emission at 526 nm in a fluorescence spectrometer (Infinite 200Pro, Tecan).

Effect of Sp-sNPFs on Oocyte Maturation
To obtain spontaneously matured oocytes, ovaries with the oocytes in germinal vesicle (GV) stage were placed in a glass dish with crab serum (the status of GV stage oocytes was maintained better in crab serum than that in crab saline, data not shown). After the follicle layers were removed, the denuded oocytes were isolated from ovary and incubated in crab serum containing 5 µM Sp-sNPF peptides. The observation of oocyte germinal vesicle breakdown (GVBD) was carried out as described (56).
The GVBD was counted in intact ovary as well as in the denuded oocytes following incubation with in Sp-sNPF peptides for 0, 5, 15, 30, and 60 min.

Measurement of the Accumulation of cAMP
The measurement of cAMP accumulation and the calculation of GVBD rate were performed at the same time. Oocytes were lysed and the concentration of cAMP was determined using a competitive binding technique based on enzymelinked immunosorbent assay (ELISA) (cAMP EIA kit, Cayman) according to the manufacturer's protocols. In Vivo Effect of Sp-sNPF Peptides on Sp-Vg and Sp-VgR Expression Female crabs in pre-vitellogenic stage (weighing 45-56 g) were reared in the laboratory and were fed a shrimp diet. The initial control crabs were sacrificed on first day of the injection assay. Ten microliter of 1 mM Sp-sNPF peptides (final concentration in hemolymph about 5 µM) were injected into the arthrodial membrane at the base of last walking leg every 2 days while concurrent control crabs received the same volume of carrier.

Data Analysis
All values are expressed as mean ± SD. Multiple sequence alignment of sNPFR sequences was performed with ClustalX and embellished by LaTEX TexShade (57). Data of ligand-receptor assays were analyzed using nonlinear curve fitting (GraphPad Prism 6.0) to obtain activation curves and EC 50 values. The mRNA levels of Sp-Vg and Sp-VgR gene were normalized against the internal control β-actin, and the relative expression was calculated using the 2 − Ct method. Statistical significance was determined using one-way ANOVA followed by Duncan's multiple range tests. All statistical analyses were performed using IBM SPSS Statistics 20.0. Pictures were drawn by GraphPad Prism 6.0 and modified with Adobe Photoshop CS5.

Cloning and Expression of Sp-sNPFR in HEK239T Cells
Two partial sequences annotated as neuropeptide Y receptor type 2 (NPY2R) were mined from two transcriptome databases of S. paramamosain. The predicted Sp-sNPFR gene sequence (GenBank accession number: MH382826) encoding 458 amino acids was obtained when the two partial sequences were assembled. The protein sequence showed 60.1% identity to the mosquito, Aedes aegypti sNPF receptor, 61.1% identity to the African malaria mosquito, Anopheles gambiae sNPF receptor, 58.7% identity to B. mori sNPF receptor, 56.4% identity to S. invicta sNPF receptor, 61.9% identity to the tsetse fly, Glossina morsitans morsitans sNPF receptor, 52.4% identity to the fruit fly, Drosophila melanogaster sNPF receptor, and 54.4% identity to the Desert Locust, Schistocerca gregaria sNPF receptor (Figure 1). Enhanced green fluorescent protein (EGFP) fused to the C-terminus of Sp-sNPFR was transiently expressed in HEK293T. Confocal microscopy showed that Sp-sNPFR-EGFP mainly localized to the cell membrane as demonstrated (Figure 2).

Activation of Sp-sNPFR by Sp-sNPFs
We found that in cells transfected with Sp-sNPFR and dualluciferase reporter, Sp-sNPFR was exclusively activated by Sp-sNPF peptides mixture, while all other FLPs with the RFamide C-terminus did not activate Sp-sNPFR at 1 µM concentration ( Figure 3A). Furthermore, Sp-sNPFR was found to be activated by all the three Sp-sNPF peptides in a dose-dependent manner. The EC 50 values were found to be 1.2, 3.8, and 8.5 µM for Sp-sNPF1, Sp-sNPF2, and Sp-sNPF3, respectively ( Figure 3B).

Localization of Sp-sNPF and Sp-sNPFR mRNA in the Ovary
In the mud crab, ovary consists of oocytes and follicle layers surrounding the oocytes (Figure 4). Histological study showed that oocytes were larger in size than the follicle cells at previtellogenic stage (Figures 5A,B). In situ hybridization showed that the positive signals of Sp-sNPF mRNA were located on the follicle cells ( Figure 5C) and Sp-sNPFR was detected in both oocytes and follicle layer cells (Figure 5E). No signal was detected when the sense riboprobes of Sp-sNPF ( Figure 5D) and Sp-sNPFR ( Figure 5F) were used. Results of RT-PCR matched the findings of in situ hybridization (Figure 6).

Oocytes Were Directly Activated by Sp-sNPFs
Based on our finding that Sp-sNPFR mRNA was detected in denuded oocytes (Figure 6), we employed Ca 2+ mobilization technology to investigate whether oocytes were activated by Sp-sNPFs. As shown in Figure 7, oocytes exhibited a dosedependent response in the intracellular Ca 2+ mobilization when challenged with different concentrations of Sp-sNPF peptides.

Effect of Sp-sNPFs on Oocyte Maturation
To explore the potential role of Sp-sNPF in oocyte maturation, denuded oocytes (GV stage) which were separated from the ovaries were chosen. As indicated in Figure 8A, denuded oocytes underwent GVBD fast and over 90% oocytes exhibited GVBD after 15 min in the control group. However, the time of GVBD ratio was delayed to 30 min in oocytes treated with 5 µM Sp-sNPF peptides ( Figure 8A). Moreover, the GVBD ratio of oocytes was decreased in oocytes incubated with Sp-sNPF1, Sp-sNPF2, and Sp-sNPF3 for 5 min by 18, 33, and 21%, respectively. GVBD ratio of oocytes was decreased by 14, 12, and 14%, respectively, when Sp-sNPF1, Sp-sNPF2, and Sp-sNPF3 were used for incubation at 15 min ( Figure 8A). We also observed a higher level of cAMP in oocytes incubated with 5 µM Sp-sNPF peptides ( Figure 8B). This was in contrast to control oocytes where the cAMP level was reduced during the GVBD process ( Figure 8B).

Effect of Sp-sNPFs on Expression of Sp-Vg and Sp-VgR in vivo
Our previous study demonstrated that the expression of Sp-sNPF decreased continuously during vitellogenesis (44). In this study, we investigated the potential role of sNPF in vitellogenesis by analyzing the effect of injected synthetic Sp-sNPF peptides on the expression of Sp-Vg and Sp-VgR in hepatopancreas and ovaries derived from pre-vitellogenic females. We found that injection of Sp-sNPF1, Sp-sNPF2, and Sp-sNPF3 significantly inhibited Sp-Vg expression in hepatopancreas by 78, 59, and 83%, respectively. The expression level of Sp-VgR in ovaries was decreased by 74%, 56%, and 68%, in the presence of Sp-sNPF1, Sp-sNPF2, and Sp-sNPF3 respectively (Figure 9).
Although a large number of sNPFs have been characterized in different crustacean species (44)(45)(46)(47)(48)(49)(50), no sNPFR has been identified in crustaceans. In the present study, analyses of sequence similarity in transcriptome database revealed a candidate sNPFR annotated as a NPY-like receptor (NPY2R) in S. paramamosain. In addition, the ligand-receptor activation of signaling demonstrated that the candidate sNPFR was activated by three synthetic S. paramamosain sNPF peptides, Sp-sNPF1, Sp-sNPF2, and Sp-sNPF3, in a dose-dependent manner, with EC 50 of 1.2, 3.8, and 8.5 µM respectively and which resulted in cAMP accumulation. All other tested peptides with the RFamide C-terminus did not elicit a cAMP response. Therefore, we designated the candidate sNPFR as the S. paramamosain sNPF receptor (Sp-sNPFR). To our knowledge, this receptor is the first crustacean sNPF receptor that has been characterized in this study.
To date, several insect sNPF receptors have been cloned and characterized (33,41,(59)(60)(61)(62)(63)(64)(65)(66)(67)(68). Among these receptors, the sNPFRs from D. melanogaster (67,68)   G. m. morsitans (64), and the oriental fruit fly, Bactrocera dorsalis (Hendel) (65) have been deorphanized by ligandreceptor activation of signaling in a heterologous cell expression system, with EC 50 values in the nanomolar range. Compared to the high sensitivity of insect sNPFRs, Sp-sNPFR in our study had a relatively higher half-maximal effective concentration (EC 50 ) in the HEK293T expression system. Likewise, the intracellular Ca 2+ concentration of S. paramamosain oocytes were mobilized by relatively high concentrations of synthetic Sp-sNPF peptides in vitro. Nonetheless, our findings are similar to a study which showed that a short neuropeptide F-related receptor from the mollusk, Crassostrea gigas exhibited EC 50 value in the micromolar range, similar to our results (69).
Studies in insect sNPFRs revealed that Si-sNPFR is localized in the ovary (43) suggesting that sNPF may directly act on ovary. As S. invicta sNPF was detected in the nervous system, authors speculated that sNPF may act as a neurohormone on the ovaries (43). In our previous study (44), we found a higher level of Sp-sNPF transcript in the ovary of S. paramamosain suggesting that Sp-sNPF may function as an autocrine/paracrine factor within the ovary.
Our results showed that Sp-sNPF mRNA was exclusively localized in the follicle cells, whereas Sp-sNPFR was present in both follicle cells and oocytes. These results implied that Sp-sNPF may bind Sp-sNPFR via an autocrine/paracrine way. The localization of Sp-sNPF and Sp-sNPFR is consistent with the intraovarian expression pattern of S. paramamosain bone morphogenetic protein (Sp-BMP) and its receptors (54). BMP belongs to the transforming growth factor β (TGF-β) superfamily, classical ovarian local autocrines/paracrines factors (70,71). In vertebrates, BMPs and their receptors have a different localization, and are speculated to have an autocrine/paracrine regulatory mechanism in the ovary (72)(73)(74)(75). The distribution of Sp-sNPF and Sp-sNPFR suggested that there may be a similar autocrine/paracrine regulatory mechanism mediated by Sp-sNPF in the ovary for follicle cells to signal the follicle cells/oocytes. Finally, the finding that the Sp-sNPF peptides elicited a dramatic increase in intracellular Ca 2+ is consistent with its proposed role as an intraovarian regulatory factor through an autocrine/paracrine regulatory mechanism.
When FG oocytes were separated from the follicle cells, they underwent spontaneous maturation with almost all oocytes exhibiting GVBD in 15 min. Meanwhile, the cAMP level in the oocytes decreased significantly. In contrast, when Sp-sNPF peptides were added to the culture medium, GVBD rate of oocytes was significantly reduced, while the cAMP levels were partly reversed. This phenomenon is in agreement with the previous reports in zebrafish where removal of the follicle layers significantly increased in spontaneous maturation of the oocytes (76,77). These findings suggest that surrounding follicle layer cells may play a similar role in both vertebrates and invertebrates in inhibiting oocyte maturation. The secretory factors that may be responsible for this inhibitory effect of follicle cells on oocyte maturation have not been fully explored. Estradiol (E 2 ) and Pituitary Adenylate Cyclase Activating Polypeptide (PACAP) were identified as two such factors where E 2 partly inhibited the spontaneous maturation of the denuded oocytes via GPER, a membrane receptor located in the oocytes (76), and PACAP had a similar inhibitory effect on oocyte maturation (77). Sp-sNPF may be one such factor that suppresses oocyte maturation by binding to Sp-sNPFR to promote cAMP accumulation. However, like E 2 and PACAP, the inhibition of Sp-sNPF was incomplete suggesting that other factors could be involved in this regulatory mechanism.
To provide further clues to elucidate the importance of Sp-sNPF in S. paramamosain ovary, we performed in vivo experiments. In contrast to the previous reports in L. migratoria (35,36), injection of Sp-sNPF peptides significant suppressed the expression of Sp-VgR in the pre-vitellogenic stage ovary in our study. VgR, located in the oocytes, has been regarded as a key receptor in the ovary which mediates the release of vitellogenin from hepatopancreas in crabs (78). Considering the correlation of Vg/VgR system, we also examined the expression level of Sp-Vg in hepatopancreas when the crabs were injected with Sp-sNPF FIGURE 9 | Inhibitory effects of Sp-sNPF peptides on the expression of Sp-Vg and Sp-VgR in vivo. Hepatopancreas and ovaries were sampled at 10 days post-injection of Sp-sNPF peptides for detecting the expression levels of Sp-Vg and Sp-VgR. The relative expression levels were normalized to β-actin. All data are presented as mean ± SD (n = 4) from a representative experiment. *P < 0.05, **P < 0.01 vs. concurrent control.
peptides. In agreement with the expression pattern of Sp-VgR, Sp-sNPF peptides exhibited a similar inhibitory effect on Sp-Vg mRNA expression. The results suggest that Sp-sNPF may be an inhibitor of vitellogenesis.
In summary, this study shows the S. paramamosain orphan receptor NPY2R is a receptor for the neuropeptide Sp-sNPF (Sp-sNPFR). To our knowledge, this is the first sNPFR reported in crustaceans. Furthermore, the tissue and cellular localization of Sp-sNPF/Sp-sNPFR and the signal transduction are consistent with its proposed function as an intraovarian regulatory factor inhibiting vitellogenesis and oocyte maturation through an autocrine/paracrine regulatory mechanism. This research provides a new insight into the sNPF functions in reproduction.