PaOctβ2R: Identification and Functional Characterization of an Octopamine Receptor Activating Adenylyl Cyclase Activity in the American Cockroach Periplaneta americana

Biogenic amines constitute an important group of neuroactive substances that control and modulate various neural circuits. These small organic compounds engage members of the guanine nucleotide-binding protein coupled receptor (GPCR) superfamily to evoke specific cellular responses. In addition to dopamine- and 5-hydroxytryptamine (serotonin) receptors, arthropods express receptors that are activated exclusively by tyramine and octopamine. These phenolamines functionally substitute the noradrenergic system of vertebrates Octopamine receptors that are the focus of this study are classified as either α- or β-adrenergic-like. Knowledge on these receptors is scarce for the American cockroach (Periplaneta americana). So far, only an α–adrenergic-like octopamine receptor that primarily causes Ca2+ release from intracellular stores has been studied from the cockroach (PaOctα1R). Here we succeeded in cloning a gene from cockroach brain tissue that encodes a β-adrenergic-like receptor and leads to cAMP production upon activation. Notably, the receptor is 100-fold more selective for octopamine than for tyramine. A series of synthetic antagonists selectively block receptor activity with epinastine being the most potent. Bioinformatics allowed us to identify a total of 19 receptor sequences that build the framework of the biogenic amine receptor clade in the American cockroach. Phylogenetic analyses using these sequences and receptor sequences from model organisms showed that the newly cloned gene is an β2-adrenergic-like octopamine receptor. The functional characterization of PaOctβ2R and the bioinformatics data uncovered that the monoaminergic receptor family in the hemimetabolic P. americana is similarly complex as in holometabolic model insects like Drosophila melanogaster and the honeybee, Apis mellifera. Thus, investigating these receptors in detail may contribute to a better understanding of monoaminergic signaling in insect behavior and physiology.


Introduction
Biogenic amines play a pivotal role in the regulation of physiology and behavior of most animals. In insects and other protostomes, tyramine and octopamine functionally substitute the noradrenergic/adrenergic system present in most vertebrates. The two phenolamines can act as neurotransmitters, neuromodulators and neurohormones as well (for recent reviews, see: [1][2][3][4]). Besides the vinegar fly Drosophila melanogaster and the honeybee (Apis mellifera) as well-established model organisms for investigating the roles of biogenic amines on behavioral plasticity and social behavior [5][6][7][8][9], the cockroach (Periplaneta americana) is well suited as a model to examining the contribution of biogenic amines on physiology and neurobiology as well [10][11][12].
Here we applied a molecular cloning and an in-silico approach to gain deeper insight into the monoaminergic receptor clade of P. americana. We succeeded in cloning a β-adrenergic-like octopamine receptor gene expressed in P. americana brain tissue. When constitutively expressed in a cell line, PaOctβ2R stimulated adenylyl cyclase activity. Notably, octopamine was two orders of magnitude more potent than tyramine in terms of activating the receptor. For both agonists the rank order of synthetic antagonists was similar, with epinastine being the most efficacious inhibitor. In addition to this functional characterization, we examined transcriptome data for the presence of monoaminergic receptors. Together with hitherto cloned receptor genes, a total of four dopamine, four serotonin, five octopamine, three tyramine receptors, and two putative biogenic amine receptors with unknown ligand specificity were identified. Thus, the monoaminergic receptor clade in P. americana is similarly complex as in other arthropods. We are convinced that these findings will significantly improve further physiological, neurobiological as well as pharmacological studies that rely on specific signaling of the monoaminergic system to modulate cockroach behavior.

Molecular and Structural Properties of the PaOctβ2R
Based on currently known biogenic amine receptors from A. mellifera and D. melanogaster, degenerated primers were designed. We succeeded in cloning the complete cDNA encoding a PaOctβ2R by PCR on single stranded cDNA synthesized on mRNA isolated from adult cockroach brain tissue. The cDNA contains an open reading frame (ORF) of 1371 bp.
The deduced amino acid sequence consists of 457 residues with a calculated molecular weight of 52.1 kDa and a pI of 8.73. The hydrophobicity profile according to Kyte and Doolittle [32] and prediction of transmembrane helices using TMHMM Server v.2.0 [33] suggest seven transmembrane (TM) domains ( Figure 1A,B), a cognate feature of GPCRs.
The TM segments are flanked by an extracellular N-terminus of 69 residues and an intracellular C-terminus of 67 residues. We submitted the PaOctβ2R sequence to Phyre2 [34] and obtained a three-dimensional model of the receptor that supported the typical membrane arrangement of a GPCR ( Figure 1C).  [32] using a window size of 19 amino acids. Peaks with scores greater than 1.6 (dashed line) indicate possible transmembrane regions. (B) Prediction of transmembrane domains with TMHMM server v. 2.0 [33]. Putative transmembrane domains are indicated in red. Extracellular regions are shown with a purple line, intracellular regions with a blue line. (C) Color-coded (rainbow) 3D model of the receptor as predicted by Phyre2 [34]. The extracellular N-terminus (N) and the intracellular C-terminus (C) are labeled.
The primary sequence of the PaOctβ2R contains several putative sites for posttranslational modification (Supplementary Figure S1). Two potential N-glycosylation sites (N-X-(S/T)) are present in the extracellular N-terminus: N 21 ST and N 45 LT. Nine consensus sites for phosphorylation by protein kinase A and eight consensus sites for phosphorylation by protein kinase C are found in the cytoplasmic domains of the receptor protein ( Figure S1). Cysteine residues (C 404 , C 407 , and C 410 ) in the C-terminal loop might serve as palmitoylation sites. Insertion of these fatty acids into the plasma membrane additionally stabilizes the protein and might facilitate interaction with the corresponding G-protein.
In addition to these sites, several characteristic sequence motifs of GPCRs were identified in the primary structure of PaOctβ2R. The D 163 RY motif (D 3.49 R 3.50 Y 3.51 ; labeling according to [35]) is located at the cytoplasmic end of TM3 ( Figure S1). In TM7, the residues N 383 PLIY (N 7.49 P 7.50 L 7.51 I 7.52 Y 7.53 ) constitute part of the hydrophobic interaction site with the phenyl moiety of the biogenic amine. Furthermore, D 146 (D 3.32 ) in TM3 and S 235/239 (S 5.42/5.46 ) in TM5 ( Figure S1) most likely interact with the protonated amino group and hydroxyl group of the biogenic amine, respectively, as has been experimentally demonstrated for several members of biogenic amine receptor family [36,37].

Tissue Distribution of PaOctβ2R mRNA
The expression pattern of PaOctβ2R mRNA in various tissues of P. americana was investigated by RT-PCR with specific primers. The level of expression was generally very low. After the initial RT-PCR reaction, the PaOctβ2R transcript was only detected in the brain sample. After reamplification, the PaOctβ2R transcript could also be detected in samples of the leg muscle, the salivary gland, and the midgut (Figure 3). Receptor mRNA expression could not be reliably detected in samples from the Malpighian tubules. To ensure that the fragments were not amplified from genomic DNA, samples were treated with DNase I. When tissue samples were treated additionally with an RNase cocktail prior to initial RT-PCR, no PCR product could be amplified (data not shown).  [40] by using the core amino-acid sequences of TM 1-4, TM 5, TM 6, and TM 7. The evolutionary history was inferred by using the Maximum Likelihood method based on the Poisson correction model. The tree with the highest log likelihood (−28600.17) is shown. The percentage of trees in which the associated taxa clustered together is shown next to the branches. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. The analysis involved 93 amino acid sequences. Human rhodopsin (HsRHOD) was used to root the tree. Receptor subclasses are given on the right. Abbreviations of species in alphabetical order are: Am Apis mellifera, Dm Drosophila melanogaster, Hs Homo sapiens, Pa Periplaneta americana, Pc Priapulus caudatus, Pd Platynereis dumerilii, Sk Saccoglossus kowalevskii. Protostomian species names are highlighted in red, whereas deuterostomian species names are given in blue. Accession numbers and annotations of all sequences used in the phylogenetic analysis can be found in Supplementary Table S1. Detection of PCR products amplified on total RNA isolated from brain, salivary glands, Malpighian tubules, midgut, and leg muscle. Amplification failed when samples were digested with an RNAse cocktail prior to RT-PCR (data not shown). The lower panel shows RT-PCR products amplified with actin-specific primers (Accession No. AY116670) as a control.
Additional candidates for biogenic amine receptors, including two additional βadrenergic-like octopamine receptors and two α-adrenergic-like octopamine receptors, were identified by screening the transcriptome data of various cockroach tissues. Table S2 summarizes the distribution of PaOctβ2R and all other monoaminergic receptors of the cockroach in the available transcriptomes.

Expression of PaOctβ2R-HA in flpTM Cells
To unravel the ligand specificity and intracellular signaling pathway activated by PaOctβ2R and to determine its pharmacological properties, flpTM cells were stably transfected with a construct C-terminally modified with an haemagglutinin A tag (PaOctβ2R-HA). Independent cell lines were obtained and examined by immuno-fluorescence staining for homogeneity ( Figure 4) and Western blotting ( Figure S2). Immunological staining of cells was performed with an anti-HA antibody detecting the tagged receptor ( Figure 4, green). Since the cells also constitutively express a cyclic nucleotide-gated (CNG) channel, this protein was labeled with a specific antibody raised against the intracellular C-terminus ( Figure 4, red). Cell nuclei were stained with TOPRO-3 (blue). The CNG channel but not the receptor protein was identified in flpTM cells that served as negative control ( Figure 4A,B). In contrast, both proteins were stained in flpTM + PaOctβ2R-HA expressing cells ( Figure 4E,F). Labeling was detected in the plasma membrane and in intracellular compartments. To assess whether the receptor protein was post-translationally glycosylated, membrane proteins isolated from the receptor-expressing and parental (flpTM) cell lines were both treated with or without PNGase F and then separated by SDS-PAGE. Incubation of the Western blot membrane with anti-HA antibodies labeled two bands of~68 and 60 kDa ( Figure S2A, lane 1) in the sample that was not treated with PNGase F. Treating the proteins with PNGase F resulted in a complete loss of the 68 kDa band and an increase in the amount of the 60 kDa band ( Figure S2A, lane 2). The apparent molecular weight corresponds well to the calculated value of PaOctβ2R-HA (53.4 kDa). No specific bands were labeled in lanes containing proteins from flpTM cells ( Figure S2A, lanes 3 + 4). The blot was subsequently developed with an anti-CNG channel antibody ( Figure S2B, lanes 1-4). In each lane, a single band was detected. Treatment with PNGase F resulted in a reduction of the apparent molecular weight of the channel protein ( Figure S2B, lanes 2 + 4). Based on these results it is reasonable to assume that PaOctβ2R-HA is substantially glycosylated, at least in the heterologous expression system.

Ligand Specificity of the PaOctβ2R-HA Receptor
The β-adrenergic-like octopamine receptors from, e.g., D. melanogaster [24,25] and A. mellifera [26] have been shown to stimulate adenylyl cyclase activity and thereby raise [cAMP] i upon activation. These insect receptors are preferentially activated by octopamine and to a certain extent by tyramine as well. In a first series of experiments, we thus examined the ability of different biogenic amines, i.e., octopamine, tyramine, dopamine, histamine, and serotonin, to evoke PaOctβ2R-HA-dependent cAMP production in receptor-expressing as well as in the parental (flpTM) cell line. Increasing [cAMP] i cause opening of the CNG channels and influx of extracellular Ca 2+ which was detected by the Ca 2+sensitive dye FLUO-4. No stimulation of adenylyl cyclase activity was recorded in flpTM cells when incubated with 0.1 or 1 µM concentrations of biogenic amines ( Figure 5A). In contrast, PaOctβ2R-HA-expressing cells showed increasing [cAMP] i after application of both octopamine and tyramine. Neither dopamine, histamine nor serotonin caused an increase in [cAMP] i ( Figure 5B). In a next step, concentration series of octopamine and tyramine ranging from 10 −12 M to 10 −7 M for octopamine and 10 −11 M to 10 −6 M for tyramine were applied to the cell lines. The fluorescence signals were normalized to values obtained in flpTM + PaOctβ2R-HA cells with the highest ligand concentration (=100%) and used to calculate concentration response curves for the receptor ( Figure 6; Table 1). The concentration-response curves for both phenolamines were sigmoid and saturated at an octopamine concentration of ≥10 −9 M and a tyramine concentration of ≥10 −7 M, respectively ( Figure 6). The ligand concentration leading to half maximal activation of PaOctβ2R-HA (EC 50 ) was 2.97 × 10 −10 M octopamine and 3.68 × 10 −8 M tyramine, respectively (for mean values of all experiments, see Table 1). In non-transfected flpTM cells, no change in the fluorescence signal was observed upon application of either octopamine or tyramine. Accordingly, all subsequent measurements with antagonists (see Section 2.6) were carried out on an octopamine or tyramine background causing~75% of the maximal response.
The pharmacological experiments conducted so far suggested that PaOctβ2R-HA has a clear (~100 fold) preference for octopamine over tyramine and can therefore be considered a functional β-adrenergic-like octopamine receptor.
Yohimbine, 5-MT, and 8-OH-DPAT did not impair octopamine-or tyramine-stimulated PaOctβ2R-HA activity.    Based on these data the same rank order of antagonist potency was uncovered for octopamine-and tyramine-stimulated PaOctβ2R-HA. The receptor was efficiently blocked by epinastine, mianserin, and phentolamine whereas yohimbine, 5-MT, and 8-OH-DPAT were non effective at all.

Discussion
There is ongoing interest in understanding the physiological and behavioral roles of octopaminergic signaling in insects (for recent reviews, see: [2][3][4][41][42][43]). An important step in meeting this challenge is to determine the molecular and functional-pharmacological properties of octopamine receptor subtypes. Here we describe the functional characterization of PaOctβ2R, the second octopamine receptor subtype of the cockroach, P. americana, an established model insect in neurobiological, physiological, and toxicological studies (for reviews, see: [10,[44][45][46]). The primary structure of PaOctβ2R phylogenetically clusters with protostomian β-adrenergic-like octopamine receptors. Activation of PaOctβ2R by the phenolamines octopamine and tyramine led to a substantial increase in cAMP synthesis. Further candidates for biogenic amine receptors, including two additional β-adrenergic octopamine receptors, were identified by screening the transcriptome data of various cockroach tissues.

Structural Properties of the PaOctβ2R Protein and Phylogenetic Classification
Applying several in-silico analyses confirmed that PaOctβ2R is a member of the class A (rhodopsin-like) GPCR family. This assessment is supported by the presence of cognate amino acid residues and motifs within the TM segments in PaOctβ2R, e.g., N 383 PLIY in TM7 or the D 163 RY motif at the C-terminal end of TM3.
Most class A (rhodopsin-like) GPCRs are activated by ligands docking to specific residues in the binding pocket of the receptor near the extracellular side. Functionally important amino acid residues present in β-adrenergic-like octopamine receptors are well conserved in the PaOctβ2R sequence. These are an aspartic acid residue (D 146 ) in TM3 and two of four closely grouped serine residues found in TM5 (S 235, 239 ) (see: Supplementary Figure S1). Octopamine appears to bind via its amine group and its hydroxyl group to the aspartic acid and one of the serine residues of the receptor, respectively [47]. In addition, phenylalanine and/or tryptophan residues in TM6 and TM7 (see: Supplementary Figure S1) might contribute to π-π interaction with delocalized electrons in octopamine and stabilize the receptor ligand interaction.
The coupling of GPCRs to specific G proteins is brought about by amino-acid residues in close vicinity to the plasma membrane of the 2nd and 3rd intracellular loops and of the cytoplasmic C-terminus of the receptor [48,49]. Various insect β-adrenergic-like octopamine receptors possess strikingly similar amino-acid sequences throughout their 2nd cytoplasmic loops and in the vicinity of TM5 and TM6 within their 3rd cytoplasmic loops, regions largely determining the specificity of receptor/G s -protein coupling [50].
Our phylogenetic analysis including all major insect biogenic amine GPCR families resulted in a well-resolved phylogram ( Figure 2). Protostomian β-adrenergic-like octopamine receptors seem to be closely related to deuterostomian β-adrenergic receptors and D 1 -like dopamine receptors (Figure 2), emphasizing the idea of "ligand-hopping" during evolution of aminergic GPCRs [38]. When new receptors evolved by gene duplication, they eventually needed new ligands for activation. Because of structural constrains, the only way to obtain "new" aminergic ligands was to repurpose already existing biogenic amines from other systems. The frequent ligand exchanges during evolution of aminergic GPCRs strongly contrasts with the situation observed for neuropeptide and protein hormone GPCRs, where generally co-evolution between receptors and their ligands takes place [38,51,52].

Posttranslational Modification of PaOctβ2R
Posttranslational modifications in intracellular loops of PaOctβ2R, like phosphorylation, may affect the signaling properties of the protein. Cysteine residues in the C-terminus of different biogenic amine receptors were found to undergo posttranslational palmitoylation [53]. This modification generates a fourth intracellular loop that participates in receptor-G protein binding. PaOctβ2R has three such cysteine residues in its C-terminal end. In addition, the primary structure of PaOctβ2R harbors two potential sites for N-linked gly-cosylation ( Figure S1) and we were able to show that at least the heterologously expressed receptor protein is present to a considerable extent in glycosylated form ( Figure S2).

Octopamine Receptors as Molecular Targets of Insecticides
As mentioned above, octopamine signaling in insects is highly complex. Octopamine receptors characterized so far can be classified as either α 1 -adrenergic-like which preferentially couple to G q proteins and induce intracellular Ca 2+ mobilization, α 2 -adrenergiclike which preferentially couple to G i proteins and inhibit adenylyl cyclase activity, or β-adrenergic-like which preferentially couple to G s proteins and activate adenylyl cyclases [2,3,19,22,23]. Octopamine receptors, especially those that lead to an increase in [cAMP] i , are also potential molecular targets of selectively acting insecticides, e.g., from the class of formamides and plant essential oils (for reviews, see: [61][62][63]). As early as in the middle of the 1980s, the formamidines demethylchlordimeform (DMCD), BTS-27271, and amitraz were shown to mimic the action of octopamine in elevating adenylyl cyclase activity in the nervous tissue of P. americana [64,65]. In D. melanogaster, however, DMCD and chlordimeform inhibit octopamine-stimulated adenylyl cyclase activity, whereas amitraz activates the enzyme [66]. Dihydrooxadiazine insecticides were reported to elevate cAMP levels in homogenates of the P. americana nervous system most likely by stimulation of octopamine receptors [67]. The neurotoxic activity of plant essential oils and their purified constituents against various insect species is also likely due to their interactions with octopamine and/or tyramine receptors [68][69][70]. Recently, Kita and colleagues [71] systematically performed pharmacological studies using heterologously expressed octopamine receptors of the silkworm Bombyx mori to identify the molecular target of the formamidine amitraz. The β-adrenergic-like octopamine receptor BmOAR2 [37] turned out to be more sensitive to amitraz and its metabolite DPMF than the α1-adrenergic-like octopamine receptor BmOA2 [71,72]. Moreover, DPMF was confirmed to be a more potent octopamine receptor agonist than amitraz itself, indicating that amitraz undergoes metabolic activation. Site-directed mutagenesis studies clearly showed that amitraz and DPMF act as orthosteric antagonists [73]. These groundbreaking studies provided evidence that target-site insensitivity significantly contributes to the widespread amitraz resistance [73]. In veterinary medicine, amitraz is also used to control ectoparasitic mites, e.g., the cattle tick Rhipicephalus microplus [74,75] and Varroa destructor, a mite that attacks and feeds on hon-eybees [76,77]. A non-synonymous SNP in a β-adrenergic-like octopamine receptor of amitraz-resistant R. microplus was proposed to be the cause of amitraz resistance [78]. In an elegant study, Guo and colleagues [79] found that amitraz and DPMF activated various V. destructor octopamine receptors. The mite VdOctβ2R was more sensitive to amitraz and its metabolite than the honeybee AmOctβ2R [79]. Furthermore, replacement of three beespecific residues with their counterparts in the mite receptor increased amitraz sensitivity of the bee AmOctβ2R receptor, suggesting that these three residues are responsible for the resistance of honeybees to amitraz [79]. Last but not least, behavioral studies using D. melanogaster null mutants of octopamine receptors identified the DmOctβ2R as the sole target of amitraz in the fly [79].
The American cockroach P. americana is considered to be an insect pest of significant public health importance. They can passively transport pathogenic bacteria, such as Salmonella, on their body surfaces, particularly in areas like hospitals [80]. In addition, house dust containing cockroach feces and body parts can trigger allergic reactions and asthma in certain individuals [81,82]. For these reasons, cockroach populations are controlled using insecticides [83]. With the characterization of the PaOctβ2R described here and in particular the production of a cell line that stably expresses this receptor, we now have an efficient tool to screen for potential and highly selective insecticides. Further insecticide targets can possibly be found among the receptors that we have identified in the transcriptomes. . Finally, a PCR was performed on single-stranded P. americana-brain cDNA to amplify the entire coding region of PaOctβ2R by using gene-specific primers annealing in the 5 -and 3 -untranslated regions (SF: 5 -CGTCGTGAACCTCTGACATC-3 ; SR: 5 -CCTCAAGTCGGGTAACTGTC-3 ). The nucleotide sequence of PaOctβ2R has been submitted to the GenBank database (accession#: OL739164). N-glycosylation sites were predicted by NetNGlyc 1.0 Server (http://www.cbs. dtu.dk/services/NetNGlyc/ (accessed on 10 September 2021)). Putative phosphorylation sites were predicted by NetPhos 3.1 Server (http://www.cbs.dtu.dk/services/NetPhos/ (accessed on 10 September 2021); [85]).

Transcriptome Sequencing and De novo Assembly of Nucleotide Sequences
Various tissues from adult P. americana (CNS, frontal ganglion, corpora cardiaca, heart, leg muscle, midgut) were prepared as described in [86]. Subsequent RNA isolation, cDNA preparation and Illumina Next Generation Sequencing were carried out by BGI (Beijing Genomics Institute, New Territories, Hong Kong, China). Each library was sequenced for 100 bp paired ends, the resulting RAW reads were filtered by removing adapter sequences, contamination, and low-quality reads. De novo assembly of RNA-Seq data was conducted by using Trinity (v2.2.0) [87]. The local BLAST tool of BioEdit v. 7.0.5.3 [40] was used to search for orthologous sequences of cockroach monoaminergic receptors, with receptor sequences of A. mellifera and D. melanogaster as queries [38].

Multiple Sequence Alignment and Phylogenetic Analysis
For phylogenetic analysis, we included amino acid sequences of biogenic amine receptors of various protostomian and deuterostomian species. Sequences were obtained from NCBI databases (NCBI, Bethesda, MD, USA). Multiple sequence alignments were performed with BioEdit and amino acid sequences were trimmed to regions encoding TM 1-4, TM 5, TM 6, and TM 7. Afterwards, evolutionary analyses were conducted in MEGA7 [88]. The evolutionary history was inferred by using the Maximum Likelihood method based on the Poisson correction model. Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using a JTT model, and then selecting the topology with superior log likelihood value. The analysis involved 93 amino acid sequences. There was a total of 238 positions in the final dataset. The human rhodopsin sequence (HsRHOD) formed the outgroup.
Sequence identity and similarity of β 2 -adrenergic-like octopamine receptors between P. americana, Cryptotermes secundus, D. melanogaster, and A. mellifera were determined by using the BLOSUM62 similarity matrix.

RT-PCR Amplification of PaOctβ2R Fragments
RT-PCR experiments to determine the tissue distribution of PaOctβ2R were performed as described earlier [27,28]. Briefly, receptor-specific fragments were amplified from 100 ng total RNA isolated from brain, salivary glands, midgut, Malpighian tubules, and leg muscle of adult male cockroaches. Sense and antisense primers were 5 -CTCAACCGCTTCCATCCT CC-3 (RT-F) and 5 -GCTTCTCTTGCCTGTTCGCC-3 (RT-R), respectively. Amplification resulted in fragments of the expected length of 344 bp. cDNA was synthesized for 30 min at 50 • C followed by a denaturation step at 94 • C for 2 min. Amplification was performed for 35 cycles at 94 • C for 40 s, 66 • C for 40 s, and 72 • C for 30 s, followed by a final extension at 72 • C for 10 min. For reamplification, 2 µL of the PCR product of the initial RT-PCR were used as a template. RT-PCR was also performed with actin-specific primers (Accession No. AY116670) as an internal control (ActinF: 5 -CGAGTAGCTCCTGAAGAGC-3 ; ActinR: 5 -GGCCTCTGGACAACGGAACC-3 ; fragment length: 488 bp).

Construction of PaOctβ2R-HA Expression Vector
An expression-ready construct of PaOctβ2R in pcDNA3.1(+) vector was generated by PCR. Specifically, receptor encoding cDNA was modified in a PCR with primers PaOctβ2Rexpr-F (5 -TTTAAGCTTCCACCATGGCGTCCAATCCCGATATC-3 ) and PaOctβ2R-expr-R (5 -TTTGAATTCCAGACTGCTGCCGAACTCGC-3 ). In front of the start codon, a HindIII restriction site (AAGCTT) and a Kozak consensus motif (CCACC; [89]) were inserted. The stop codon was replaced by an EcoRI recognition sequence (GAATTC). We reused the pcAm5-ht1A-HA construct [90] and exchanged the Am5-ht1A cDNA for the Paoctβ2R cDNA by ligation into the HindIII and EcoRI sites. The resulting construct (pcPaoctβ2R-HA) is extended in frame at the 3 end with a sequence encoding the hemagglutinin A (HA) tag (YPYDVPDYA) which allowed us to monitor receptor protein expression using a specific anti-HA antibody (Roche Applied Science/Sigma-Aldrich/Merck, Mannheim, Germany). The correct insertion was confirmed by DNA sequencing.

Functional Expression of the PaOctβ2R-HA Receptor
For PaOctβ2R-HA expression and functional analysis, we used a human embryonic kidney (HEK293; flpIn cells; Invitrogen/ThermoFisher Scientific; #750-07)-based cell line that had been transfected with a gene encoding a variant of the A2-subunit of the olfactory cyclic nucleotide-gated ion channel (CNG; [91]; flpTM cells, provided by Sibion biosciences, Jülich, Germany). Cells were transfected with 10 µg of the pcPaOctβ2R-HA construct by a modified calcium phosphate method [92] following a previously established protocol [93]. Transfected cells were selected in the presence of the antibiotics G418 (1 mg/mL) and hygromycin (100 µg/mL).

Functional Analysis of the PaOctβ2R-HA Receptor
A stably transfected cell line was established to examine PaOctβ2R-HA receptor activity by Ca 2+ imaging. Control measurements were performed in the parental (flpTM) cell line. Changes in [cAMP] i were registered indirectly via co-expressed CNG channels that are opened by cAMP and cause an influx of extracellular Ca 2+ [18,26,91]. Changes in [Ca 2+ ] i were monitored with the Ca 2+ -sensitive fluorescent dye Fluo-4. Cells were grown in 96-well dishes to a density of approximately 2 × 10 4 cells per well and were loaded at room temperature with Fluo-4 AM as described previously [26]. After 90 min, the loading solution was substituted for dye-free extracellular solution (ECS; 120 mM NaCl, 5 mM KCl, 2 mM MgCl 2 , 2 mM CaCl 2 , 10 mM HEPES, and 10 mM glucose, pH 7.4 (NaOH)) containing 100 µM 3-isobutyl-1-methylxanthine (IBMX; Sigma-Aldrich/Merck, Darmstadt, Germany) as an inhibitor of cell-endogenous phosphodiesterases. The plate was transferred into a fluorescence reader (FLUOstar Omega, BMG Labtech, Ortenberg, Germany) to monitor Fluo-4 fluorescence. The excitation wavelength was 485 nm. Fluorescence emission was detected at 520 nm. Concentration series of various biogenic amines and synthetic receptor ligands were added once Fluo-4 fluorescence had reached a stable value in each well. The changes in Fluo-4 fluorescence were recorded automatically. Concentration-response curves were established from at least three independent experiments. Data points were derived from four-fold or eightfold determination. Data were analyzed and displayed using Prism 5.04 software (GraphPad, San Diego, CA, USA).