PACAP-38 and PACAP(6–38) Degranulate Rat Meningeal Mast Cells via the Orphan MrgB3-Receptor

Infusion of pituitary adenylate cyclase activating peptide-38 (PACAP-38) provokes migraine attacks in migraineurs and headache in non-migraineurs. Adverse events like long-lasting flushing and heat sensation can be terminated with oral antihistamine treatment, indicating the involvement of mast cell activation after PACAP-infusion. Degranulation of rat peritoneal mast cells was provoked by several isoforms of PACAP via previously unknown receptor pharmacology. The effect might thus be mediated either via specific splice variants of the PAC1-receptor or via an unknown receptor for PACAP-38. In the present study, we characterize degranulation of rat meningeal mast cells in response to PACAP-receptor ligands. Furthermore, we investigate if PACAP-38-induced mast cell degranulation is mediated via PAC1-receptor splice variants and/or via the orphan Mas-related G-protein coupled member B3 (MrgB3)-receptor. To address this, the pharmacological effect of different PACAP isoforms on meningeal mast cell degranulation was investigated in the hemisected skull model after toluidine blue staining followed by microscopic quantification. Presence of mRNA encoding PAC1-receptor splice variants and the MrgB3-receptor in rat mast cells was investigated by Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) analysis. The effect of PACAP isoforms on PAC1- and MrgB3-receptor-expressing Xenopus laevis oocytes were performed by two-electrode voltage-clamp (TEVC) electrophysiology. PACAP-38 is a more potent mast cell degranulating agent than Pituitary Adenylate Cyclase Activating Peptide-27 (PACAP-27) in the meninges. Presence of mRNA encoding the PAC1-receptor and its different splice variants could not be detected in peritoneal mast cells by RT-PCR, whereas the orphan MrgB3-receptor, recently suggested to be a mediator of basic secretagogues-induced mast cell degranulation, was widely present. In PAC1-receptor-expressing Xenopus laevis oocytes both PACAP-38, PACAP-27 and the specific PAC1-receptor agonist maxadilan were equipotent, however, only PACAP-38 showed a significant degranulatory effect on mast cells. We confirmed Pituitary Adenylate Cyclase Activating Peptide(6–38) [PACAP(6–38)] to be a PAC1-receptor antagonist, and we demonstrated that it is a potent mast cell degranulator and have an agonistic effect on MrgB3-receptors expressed in oocytes. The present study provides evidence that PACAP-induced mast cell degranulation in rat is mediated through a putative new PACAP-receptor with the order of potency being: PACAP-38 = PACAP(6–38) > > PACAP-27 = maxadilan. The results suggest that the observed responses are mediated via the orphan MrgB3-receptor.

Infusion of pituitary adenylate cyclase activating peptide-38 (PACAP-38) provokes migraine attacks in migraineurs and headache in non-migraineurs. Adverse events like long-lasting flushing and heat sensation can be terminated with oral antihistamine treatment, indicating the involvement of mast cell activation after PACAP-infusion. Degranulation of rat peritoneal mast cells was provoked by several isoforms of PACAP via previously unknown receptor pharmacology. The effect might thus be mediated either via specific splice variants of the PAC 1 -receptor or via an unknown receptor for PACAP-38. In the present study, we characterize degranulation of rat meningeal mast cells in response to PACAP-receptor ligands. Furthermore, we investigate if PACAP-38-induced mast cell degranulation is mediated via PAC 1 -receptor splice variants and/or via the orphan Mas-related G-protein coupled member B3 (MrgB 3 )-receptor. To address this, the pharmacological effect of different PACAP isoforms on meningeal mast cell degranulation was investigated in the hemisected skull model after toluidine blue staining followed by microscopic quantification. Presence of mRNA encoding PAC 1receptor splice variants and the MrgB 3 -receptor in rat mast cells was investigated by Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) analysis. The effect of PACAP isoforms on PAC 1 -and MrgB 3 -receptor-expressing Xenopus laevis oocytes were performed by two-electrode voltage-clamp (TEVC) electrophysiology. PACAP-38 is a more potent mast cell degranulating agent than Pituitary Adenylate Cyclase Activating  in the meninges. Presence of mRNA encoding the PAC 1receptor and its different splice variants could not be detected in peritoneal mast cells by RT-PCR, whereas the orphan MrgB 3 -receptor, recently suggested to be a mediator of basic secretagogues-induced mast cell degranulation, was widely present. In PAC 1 -receptor-expressing Xenopus laevis oocytes both PACAP-38, PACAP-27 and the specific PAC 1 -receptor agonist maxadilan were equipotent, however, only PACAP-38

INTRODUCTION
Pituitary adenylate cyclase-activating peptide-38 (PACAP-38) is a 38-amino acid neuropeptide located in both sensory and parasympathetic perivascular nerve fibers (Moller et al., 1993;Mulder et al., 1994). A C-terminal truncated 27-amino acid  version is endogenously present as well but is less abundant (Miyata et al., 1990;Arimura et al., 1991;Ogi et al., 1993). A 20-min intravenous infusion of PACAP-38 provokes migraine attacks in migraine patients as well as headache in non-migraineurs (Schytz et al., 2009). At present, three PACAP-receptors have been identified: PAC 1 , VPAC 1 and VPAC 2 . The neurotransmitter vasoactive intestinal peptide (VIP) shares high amino acid sequence homology with PACAP and its affinity to VPAC 1 and VPAC 2 equals that of PACAP (Spengler et al., 1993;Pantaloni et al., 1996) whereas binding to the PAC 1 -receptor is 1,000 times lower (Miyata et al., 1989(Miyata et al., , 1990Harmar et al., 1998). Interestingly, VIP only induces a mild headache and no migraine-like attacks in migraineurs (Rahmann et al., 2008), which leads to the suggestion that PACAP and the PAC 1 -receptor are key targets for future migraine treatment.
Infusion of PACAP-38 caused not only migraine attacks but also heat sensation and long-lasting flushing (Schytz et al., 2009). This is in line with PACAP-38 being a mast cell degranulator and mast cells have been suggested to play a role in migraine pathogenesis (Moskowitz, 1993;Levy et al., 2006Levy et al., , 2007. Degranulation of mast cells can be induced either by an allergen-IgE-dependent mechanism or via an IgE-independent mechanism. The latter mechanism can be activated by a group of molecules known as basic secretagogues. These molecules only share one physicochemical nature, their cationic property (Ferry et al., 2002). Several of these molecules are endogenous peptides and high concentrations are required for initiation of mast cell degranulation, an effect that involves pertussis toxin-sensitive G-proteins coupled to phospholipase C (PLC) activation (Ferry et al., 2002).
Inspired by clinical findings, we have previously characterized the degranulating effect of various PACAP-analogues on isolated rat peritoneal mast cells. Based on the expectation that degranulation is mediated through the PAC 1 -receptor, we found an unpredicted order of potency . In peritoneal mast cells, the PAC 1 -receptor antagonist Pituitary Adenylate Cyclase Activating Peptide(6-38) [PACAP(6-38)] caused mast cell degranulation that was as potent as PACAP-38 (Robberecht et al., 1992;Baun et al., 2012). Furthermore, the PAC 1 -receptor agonist maxadilan was ineffective .
Several PAC 1 -receptor splice variants have been cloned and characterized by ligand binding and signal transduction (Spengler et al., 1993;Pantaloni et al., 1996;Pisegna and Wank, 1996;Dautzenberg et al., 1999;Lutz et al., 2006). , Tatemoto et al. (2006 found the Mas-related G-protein coupled receptor member X 2 (MrgX 2 ) to be present in human mast cells. Mast cell degranulation induced by basic secretagogues appeared in the same concentrations as responses found in MrgX 2expressing cells. The rat counterpart of MrgX 2 was found to be the Mas-related G-protein coupled receptor member B 3 (MrgB 3 ; Tatemoto et al., 2006). In the present study, we hypothesized that PACAP mediated degranulation by rat peritoneal and dural mast cells were either caused by a splice variant of the PAC 1 -receptor or via MrgB 3 -receptors.

Animals
A total of 115 male Sprague-Dawley rats weighing 320-440 g (Taconic Europe, Ejby, Denmark) were used in this study. The rats were group-housed under a 12-h light/dark cycle and allowed ad libitum access to a standard rodent diet and water. All rats were euthanized by inhalation of a CO 2 /O 2mixture followed by CO 2 asphyxiation. Experimental procedures were approved by the National Danish Animal Experiments Inspectorate (License number 2014-15-0201-00256) and carried out in accordance with Danish legislation.

Mast Cell Degranulation in Hemi-skull Preparations
Mast cell degranulation was performed as previously described (Pedersen et al., 2015). In brief, skulls were cut mid-sagittal and the brain halves were removed, leaving the dura mater undisturbed. This was followed by immediate addition of either 350 µl 0.1, 1 or 10 µM PACAP-38 (custom synthesis by Caslo Laboratory ApS, Lyngby, Denmark) or vehicle (saline) in phosphate buffered saline (PBS). After 30 s incubation, the reactions were terminated, and the skulls were fixated in 4% paraformaldehyde in phosphate buffered saline (PBS; Glostrup Hospital Pharmacy, Denmark). The dura mater was dissected from the skull, whole mounted on slides, and mast cells were visualized by staining with 0.1% acidified toluidine blue (Sigma Aldrich, Germany). Tissues were dehydrated in graded alcohols prior to cover slip mounting. The level of mast cell degranulation was evaluated by 400× magnification (Nikon Eclipse Ni microscope) by a researcher blinded to the treatment and was counted in 10 consecutive fields along the stem part of the middle meningeal artery. Mast cells were considered degranulated if an extensive dispersion of more than 10 extruded granules were localized outside the cell or if an extensive loss of staining gave the cell a ''ghostly'' look.

Peritoneal Mast Cell Isolation and RNA-Extraction
Peritoneal mast cells were harvested from three rats by injecting 20 ml oxygenated buffer (137 mM NaCl, 2.7 mM KCl, 1 mM MgCl 2 , 0.5 nM CaCl 2 , 0.4 mM NaH 2 PO 4 , 10 mM HEPES and 5.6 mM glucose, pH 7.6) in the peritoneal cavity of asphyxiated rats. The cavity was then gently massaged and subsequently opened by midline incision for the lavage to be removed by pipetting. Cells were washed three times by sedimentation at 13 • C by a 7 min centrifugation at 400 g. The pellet was re-suspended in 5 ml oxygenated buffer and layered on top of a BSA-Percoll density gradient (81%) containing 162 µl 35% bovine serum albumin (BSA), 8.1 ml Percoll (GE Healthcare, Buckinghamsure, UK), 580 µl distilled water and 1.16 ml salt solution (1.54 M NaCl, 27 mM KCl, 3.8 mM CaCl 2 ). Cell types were separated by centrifugation at 225 g for 25 min at 13 • C. The density gradient was discarded, and the pellet was again washed three times. The purity of mast cells was determined by histological characterization of the percentwise mast cell fraction. Only samples with a purity >90% were used for further analysis. Peritoneal mast cell RNA was extracted using the Isolation of Small and Large RNA Kit (Macherey-Nagel, Germany) in combination with TRIzol (Qiagen) according to manufacturer's recommendations.

Reverse Transcriptase-Polymerase Chain Reaction
cDNA was synthesized from 500 ng peritoneal mast cell RNA using the iScript cDNA Synthesis kit (BioRad) according to instructions. PAC 1 -receptor splice variants were identified in peritoneal mast cells and spinal cord using HotStarTaq DNA polymerase (QIAGEN) with 10 µM primer. The MrgB 3 -receptor was only tested in mast cell RNA. Primers against the Adcyap1r1 gene (NM_001270582.1, encoding the PAC 1 -receptor), RGD1560730 (XM_006229262.3, encoding the MrgB 3 -receptor), and β-actin were designed in Primer3 (Broad Institute) and tested for specificity by BLAST alignment tool (NCBI) and ordered from DNA Technology, Aarhus, Denmark ( Table 1). PAC 1 -primers were specifically designed to span the extracellular N-terminal (exon 3 to exon 8), exon 14 (known as ''hip'') and exon 15 (known as ''hop''). β-actin was included as a positive control. The amplification protocol was as follows: initial heat activation at 95 • C for 15 min, followed by 45 cycles with denaturation at 95 • C for 1 min, annealing at variable temperatures depends on the primer set (50 • C for β-actin and N-terminal, 56 • C for MrgB 3 and 66 • C for Hip + Hop) for 1 min, and extension at 72 • C for 1 min; final extension at 72 • C for 10 min. Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) products were visualized by agarose gel electrophoresis.

In vitro Transcription
DNA of the coding region of Rattus Norvegicus Adcyap1r1 variant 5 (NM_001270582.1) encoding the null-splice variant (short N-terminal and neither exon 14 or 15) of the PAC 1 -receptor (Vector: EX-Rn10199-M03) was ordered from GeneCopoeia. DNA encoding the RGD1560730 gene (XM_006229262.3, GenScript), was cloned into the pXOOM vector as previously described (Jespersen et al., 2002). DNA was purified using Plasmid DNA Purification NucleoBond Xtra Midi-kit (Macherey-Nagel). The inserts were fully sequenced (MWG Operon) to confirm the expected sequence (data not shown). Extracted plasmids were linearized down-stream the poly(A) segment using the XhoI restriction enzyme (New England Biolabs, Ipswich, MA, USA). RNA was in vitro transcribed by synthetization from the T7 RNA polymerase promoter using the mMessenger mMachine kit (Ambion) according to manufacturer's protocol. Messenger RNA was purified using the MEGAclear kit (Ambion). Transcribed RNA integrity was assessed by agarose gel electrophoresis.

Statistical Analysis
Concentration-response curves for both oocytes and dural mast cells were analyzed for overall effects of PACAP-38 and -27 by two-way analyses of variance (ANOVA) followed by Sidak's test for multiple pairwise comparisons. Effects of PAC 1 -receptor ligands on mast cell degranulation or on receptor-expressing oocytes were analyzed with a one-way ANOVA followed by Tukey's multiple comparisons test. The effect of PACAP-38 and PACAP(6-38) on MrgB 3 -expressing oocytes were analyzed by an unpaired two-tailed t-test. Differences between groups were considered significant when p < 0.05 and data are presented as mean with standard error of the mean (±SEM). GraphPad Prism 7 (GraphPad Prism Software, San Diego, CA, USA) was used for statistical analysis.

Effects of PACAP-38 and PACAP-27 on Dura Mast Cell Degranulation
PACAP-provoked mast cell degranulation was characterized by stimulating the dura mater with PACAP-38, PACAP-27 or saline in concentrations ranging from 0.1 to 10 µM (n = 5-6; Table 2). PACAP-38 stimulation resulted in an eight-fold increase and highly significant mast cell degranulation at the 10 µM concentration (p < 0.0001). However, even at the highest tested concentration, PACAP-27 did not induce mast cell degranulation that was significantly different from saline treatment.

Splice Variants of PAC 1 -Receptor in Rat Peritoneal Mast Cells
Because mast cell degranulation induced by increasing concentrations of the different PACAP-isoforms did not follow the known order of potencies for PAC 1 -, VPAC 1 -or VPAC 2receptors, we investigated the presence of PAC 1 -receptor mRNA and possible splice variants by using RT-PCR (Figure 1). As a positive control, we included rat spinal cord tissue in which the PAC 1 -receptor previously has been localized (Dickinson et al., 1999). We performed RT-PCR analysis targeted to several areas involved in splice variation of the PAC 1 -receptor (N-terminal, exon 14 and exon 15) and found the PAC 1 -receptor to be absent in mast cells and present in spinal cord (Figure 1). The lack of PAC 1 -receptor mRNA expression in rat peritoneal mast cells indicates mast cell degranulation to be mediated via a non-PAC 1 -receptor.

mRNA Expression of MrgB 3 -Receptor in Rat Peritoneal Mast Cells
The PAC 1 -receptor was not expressed in rat peritoneal mast cells. Thus, we investigated a possible expression of the MrgB 3receptor as previously shown (Tatemoto et al., 2006). Using primers directed towards MrgB 3 -receptor mRNA, we found it to be expressed in rat peritoneal mast cells (Figure 2). Therefore, we decided to study the effect of PACAP isoforms on MrgB 3 -receptors and to compare the pharmacological profile on the PAC 1 -receptor using TEVC in the Xenopus laevis oocyte expression system.
In MrgB 3 -receptor-expressing oocytes, PACAP-38, but not PACAP-27, induced a rapid concentration-depended inward current in the concentration range of 1-10 µM (n = 3-24). The maximum current induced by PACAP-38 was −1.87 ± 5 µA (n = 12) at 3 µM ( Figure 3B). The maximum response to PACAP-27 was found at 10 µM resulting in a current of −0.14 ± 0.4 µA (n = 10), which was not significantly different from baseline. These rapid responses were not seen in un-injected oocytes. In concentrations between 0.1 and 3 µM no effect was observed in un-injected oocytes, indicating that Xenopus laevis oocytes do not endogenously express PAC 1 -or MrgB 3 -receptors. However, concentrations at 10 µM PACAP-38 (15 out of 19 oocytes) but not PACAP-27 occasionally caused a delayed long-lasting response in un-injected oocytes, which was distinct from the above described fast responses. Thus, 10 µM PACAP-38 was not included in the experiments. Taken together, expression of the PAC 1 -receptor and the MrgB 3receptor in Xenopus laevis oocytes shows that PAC 1 is activated by PACAP-27 as well as PACAP-38, whereas the MrgB 3 -receptor is activated by PACAP-38 only. Thus, the activation profile FIGURE 3 | (A) Two-electrode voltage clamp performed on Xenopus laevis oocytes expressing the PAC 1 -receptor (n = 8-19, n control = 6-16) showed that increasing concentrations (0.01-10 µM) of PACAP-38 and PACAP-27 cause a significant increase in membrane current. The inward currents are measured at a holding potential of −70 mV. In (B) the Xenopus laevis oocytes is expressing the MrgB 3 -receptor (n = 4-24, n control = 3-13). In these oocytes only PACAP-38 induces a significant change in current (Vm = −70 mV). (C) Concentration-response curves of PACAP-38 and PACAP-27 on rat meningeal mast cell degranulation after 30 s stimulation at concentrations ranging from 0.1 to 10 µM. It was found that a significant mast cell degranulation only was observed after stimulation with PACAP-38. Values are given as percentage of degranulated mast cells of total number of counted mast cells (n = 5-6). * Represent p < 0.05, * * * p < 0.001, and * * * * p < 0.0001 [two-way analysis of variance (ANOVA), Sidak's multiple comparisons test]. All values are given as mean ± standard error of the mean (SEM). found for MrgB 3 resembles the effects of ligands observed for mast cell degranulation ( Figure 3C). Taken together, mast cell degranulation induced by PACAP-38 and PACAP-27 in the meninges resembled mostly the current changes evoked in MrgB 3 -receptor transfected oocytes.

DISCUSSION
PACAP-38, but not the related peptide VIP, induces migraine headache in migraineurs suggesting the specific PACAPreceptor, PAC 1 , as a potential target for migraine treatment (Rahmann et al., 2008;Schytz et al., 2009). Furthermore, all participants in the clinical provocation studies experienced long-lasting flushing, especially on the face and trunk, which could be terminated by antihistamine treatment, suggesting the involvement of mast cell degranulation (Schytz et al., 2009). In a previous series of experiments performed on rat peritoneal mast cells, we found that PACAP-38, but not PACAP-27 and VIP, caused degranulation. In addition, we showed that the selective PAC 1 -receptor agonist maxadilan had no effect on mast cell degranulation. It was also found that Only PACAP-38 caused a change in current. All measurements were done at a holding potential of −70 mV. In (A). * * * represent p < 0.001 as compared to control (A). In (C) ns, p = 0.2879 (Mann-Whitney non-parametric t-test). * * Represent p < 0.01 as compared to PACAP-38 (one-way ANOVA, Tukey's multiple comparisons test). Values are given as mean ± SEM. the selective PAC 1 -receptor antagonist, PACAP(6-38), induced a pronounced mast cell degranulation . Based on these observations, we suggested that the PACAPprovoked meningeal mast cell degranulation is mediated through another receptor than the PAC 1 -receptor . Our results suggest that PACAP-induced degranulation of rat peritoneal and meningeal mast cells is mediated via the orphan MrgB 3 -receptor.

Pharmacology of PACAP on Mast Cell Degranulation
The weak degranulating effect of VIP and PACAP-27 compared to the strong degranulation effect of PACAP-38 in rat mast cells is inconsistent with the previously reported equipotent profiles of PACAP-38, PACAP-27, and VIP on VPAC 1 -and VPAC 2 -receptors (Harmar et al., 2012). In migraineurs, VIP does not provoke migraine headache, which suggests VPAC 1 -and VPAC 2 -receptors to be of minor importance in comparison to the PAC 1 -receptor (Rahmann et al., 2008). Furthermore, the PAC 1 -receptor antagonist PACAP(6-38) has not been shown to have an affinity to VPAC 1 -and VPAC 2 -receptors (Harmar et al., 2012). Taken together, this leads us to rule out the possible involvement of VPAC 1 -and VPAC 2 -receptors in PACAP-mediated mast cell degranulation.
FIGURE 7 | A correlation of 0.967 is obtained between the mean amount of degranulation induced by five PACAP isoforms , maxadilan and M65), n = 5-11 in meningeal mast cells and their net charge at neutral pH. Calculations are performed using the Bachem Peptide Calculator (http://www.bachem.com/service-support/peptidecalculator/). domain and the third intracellular loop account for fine tuning of ligand affinity and signal transduction through adenylate cyclase or PLC activation (Deutsch and Sun, 1992;Spengler et al., 1993). The presence of a 21-amino acid domain in the extracellular N-terminal domain (PAC 1 -full, short) impairs PACAP-27 binding (Pantaloni et al., 1996). Hip and hop (exon 14 and 15, respectively) insertions into the third intracellular loop are suggested to modulate G-protein coupling and favor PACAP-38 induced PLC activation via Gq-proteins as compared to PACAP-27 (Spengler et al., 1993;Blechman and Levkowitz, 2013). This could explain the difference found in PACAP-38 and PACAP-27 provoked mast cell degranulation. We, therefore, designed primers directed towards the N-terminal part and towards exon 14 and 15. However, using the RT-PCR analysis we were unable to identify PAC 1 -receptor expression in the mast cell transcriptome. To further confirm the validity of the primers, we made parallel RT-PCR experiments on mRNA from spinal cord showing the presence of both the N-terminal part and the hip-hop variants of the PAC 1receptor. Based on these findings, we suggest PACAP-38provoked mast cell degranulation to act via a target distinct from PAC 1 -receptors.

PACAP as a Basic Secretagogue
Degranulation of mast cells induced via the IgE-independent pathway is mediated by a variety of compounds collectively designated as basic secretagogues. This is a mechanism highly conserved among mammals and birds, which appoints it to be ancient and fundamental (Halpern and Wood, 1950;Taneike et al., 1988). In general, basic secretagogues are positively charged, although hydrophobic structured compounds causing rapid mast cell degranulation [within ∼10-20 s through PLC stimulation, which is sensitive to Gi-protein inhibition, e.g., pertussis toxin (Ferry et al., 2002;Tatemoto et al., 2006)]. PACAP-38 induced an almost total mast cell degranulation within the first 10-20 s after application. The degranulation was impaired by the PLC-inhibitor U-73122, whereas adenylyl cyclase inhibitor SQ22536 was ineffective , indicating PLC activation as the responsible transduction pathway for PACAP-induced mast cell degranulation.
Interestingly, basic secretagogues seem to activate connective type mast cells independent of their putative receptor but only when applied in high concentrations (Ferry et al., 2002). Plotting the net charge of PACAP related molecules at neutral pH towards the level of degranulation induced by these PACAP analogues at 10 µM (Figure 7), we found a linear relationship with an R 2 value close to 1. Thus, a high net charge of the molecules correlates with a high mast cell degranulating effect of the PACAP analogous tested, and several factors indicate that PACAP may act as a basic secretagogue to cause mast cell degranulation despite the absence of PAC 1 -receptors.

Expression of MrgB 3 -Receptors
The exact mechanism of basic secretagogue-mediated mast cell degranulation remains unclear. The ability of basic secretagogues to act as direct activators of purified G-proteins as well as numerous failed attempts to identify an endogenously expressed receptor could suggest a receptor-independent mechanism of action (Mousli et al., 1990(Mousli et al., , 1994Seebeck et al., 1998). However, this would require the ability of peptides, despite their positive charge, to diffuse across the membrane in order to reach intracellular G-proteins, and as this has not been shown, the hypothesis seems unlikely (Tatemoto et al., 2006). The identification of a basic secretagogue receptor has for a long time been sought but without success (Ferry et al., 2002). In 2006, Tatemoto et al. (2006) studied members of the Mrg family. These are G-protein coupled receptors and expressed in a subset of nociceptive sensory neurons, thus making them interesting targets (Dong et al., 2001;Lembo et al., 2002). Expression of MrgX 2 -receptors in humans and MrgB 3 -receptors in rats were shown to be present in connective type mast cells and with affinity to various peptides like PACAP(6-27), mast cell depleting peptide and 9,10 ]-substance P could link these receptors to PACAP-mediated mast cell degranulation (Tatemoto et al., 2006). In a recent study, MrgX 2 -receptor (human) and MrgB 2 -receptor (which is the mouse orthologue of the human MrgX 2 -receptor and the rat MrgB 3 -receptor) was convincingly demonstrated to be mast cell-specific and responsible for inflammatory activation by basic secretagogues (McNeil et al., 2015). By RT-PCR, we found the MrgB 3 -receptor to be present in rat peritoneal mast cells and decided to study the effect of selected PACAP analogues on Xenopus laevis oocytes expressing MrgB 3 -receptors and to compare the responses to the effects obtained by the same PACAP analogues on PAC 1receptor-expressing oocytes.

Experiments on MrgB 3 -Receptor and PAC 1 -Receptor-Expressing Oocytes as Compared to Mast Cell Degranulation
In rat MrgB 3 -receptor-expressing oocytes, we found that PACAP-38, but not PACAP-27 and maxadilan induces currents. The effect mediated by the different PACAP agonists in the oocytes had the same characteristics as those previously found to induce a significant degranulation of rat meningeal mast cells. The findings were also in line with our previous results from peritoneal mast cells . Importantly, the concentrations required for effects were in the 1-10 µM range. In PAC 1 -receptor-expressing oocytes, PACAP-38, PACAP-27 and maxadilan were equipotent and responses were found to be significant at 10 times lower concentrations.
In line with our results from peritoneal mast cells , we found the PAC 1 -receptor antagonist PACAP(6-38) both to be a potent degranulator of rat meningeal mast cells and to induce a significant current in MrgB 3 -receptorexpressing oocytes. Contrarily, PACAP(6-38) showed the predicted antagonistic effect on PACAP-38-induced currents in PAC 1 -receptor-expressing oocytes (Harmar et al., 2012). Taken together, our studies suggest mast cell degranulation to be mediated via MrgB 3 -receptors and not via the PAC 1receptor. However, the conclusion is limited by the fact that currently no selective antibodies or antagonists directed towards the MrgB 3 -receptor are available to provide the final pharmacological evidence.
Interestingly, a similar finding of PAC 1 -receptor pharmacology and functional observations in the rat trigeminovascular system was reported. In these studies, PACAP-38, but neither VIP, PACAP-27 nor maxadilan, mediated the release of the sensory vasodilator peptide, calcitonin gene-related peptide (CGRP) from the trigeminal nucleus caudalis. Furthermore, the response to PACAP-38 seemed not to be mediated via PAC 1 -receptors due to lack of inhibition by the PAC 1 -receptor antagonist M65 (Jansen-Olesen et al., 2014). Future studies will have to rule out if MrgB 3 -receptors are involved in PACAP-38induced CGRP release in trigeminal nucleus caudalis.

CONCLUSION
In the present study, we found PACAP agonists and antagonists to have the same pharmacological effect in meningeal mast cells as previously found in peritoneal mast cells. By RT-PCR, we showed that there was no PAC 1 -receptor transcription in peritoneal mast cells thus excluding the possibility of the degranulating effect to be mediated via PAC 1 -receptor splice variants. However, we found mRNA encoding the rat MrgB 3receptor to be expressed in mast cells. This receptor was previously suggested to mediate mast cell degranulation in rat after application of basic secretagogues. This finding led us to investigate the effect of different PACAP analogues on Xenopus laevis oocytes expressing either PAC 1 -or MrgB 3 -receptors. The expressed MrgB 3 -receptor but not the PAC 1 -receptor share the same order of potency for PACAP analogues as found in rat peritoneal and meningeal mast cells. Thus, we hereby suggest the MrgB 3 -receptor to be a mediator for PACAP-induced mast cell degranulation.

DATA AVAILABILITY
All datasets generated for this study are included in the manuscript.

AUTHOR CONTRIBUTIONS
IJ-O, SP and DK designed the research. KC and FH contributed with valuable intellectual input for improvement of the study. SP and SC performed the experiments. IJ-O, SP, CK, DK and SC performed data analysis. IJ-O and SP drafted the manuscript. JO, DK, KC, FH and SC read and corrected the manuscript. JO, IJ-O, SP and DK received financial support for the study. All authors read and approved the final manuscript.

FUNDING
This study was financially supported by the Lundbeck Foundation (R77-A6952), Candy's Foundation, The Danish Research Council (11-107831), the Novo Nordisk Foundation, Frimodt-Heineke Foundation, Else and Mogens Wedell-Wedellsborg Foundation, and the A.P. Moller Foundation for the Advancement of Medical Science.