Experimental animals

For this study, 5-month-old (adult) L. stagnalis specimens were randomly obtained from our laboratory-bred stocks (Balaton Limnological Research Institute). Snails were maintained in large holding tanks (density: 100 individuals/tank) containing 10 L oxygenated artificial snail water with low copper content at 20 °C (± 1 °C) on a 12 h light:12 h dark regime. Specimens were fed on lettuce ad libitum three times a week.

Nucleotide sequencing and bioinformatic analysis

The whole CNS was dissected from the snails (n = 10) and homogenized using a TissueLyser LT (QIAGEN) in TRI reagent (#93,289, Merck). RNA was isolated with Direct-zol™ RNA MiniPrep (#R2050, Zymo Research) following manufacturer’s instructions. The RNA was quantified by a Qubit 3.0 device using BR RNA Kit (#Q10211, Thermo Fisher Scientific) and the quality was checked on Agilent Bioanalyzer 2100 using RNA 6000 Nano Kit (#5067 − 1511, Agilent).

Nanopore- and Illumina-sequencing were used for identification of any homolog in the neuronal transcriptome of L. stagnalis to the chordate and vertebrate PACAP prepropeptides, active peptides, and their receptors. In the case of Nanopore-seq, the library was prepared using the cDNA-PCR Kit (#SQK-PCS108, Oxford Nanopore Technologies) according to the manufacturer’s description. The sample was sequenced on a MinION device with R9.4.1 flowcells (#FLO-MIN106, Oxford Nanopore Technologies). Real-time base calling was performed using Guppy v3.2.2 software. Adapters were trimmed with Porechop v9.0 (Wick et al. 2018) and sequences with internal adapters (chimera reads) were also split with Porechop. Filtering on quality and read length was performed with Nanofilt (De Coster et al. 2018). Read assembly, manual correction, and consensus sequence extraction were made using the CLC Genomics Workbench v12.0.3 software de novo pipeline (QIAGEN). In the case of Illumina-seq, the library was prepared using NEBNext Ultra II Directional RNA Library Prep Kit for Illumina (#E7765L, New England Biolabs). Briefly, mRNA was isolated from 400 ng total RNA using NEBNext Poly(A) mRNA Magnetic Isolation Module (#E7490L, New England Biolabs). Thereafter, the mRNA was fragmented, end prepped, and adapter-ligated. Finally, the library was amplified according to the manufacturer’s instructions. The quality of the libraries was checked on 4200 TapeSation System using D1000 Screen Tape (#5067–5582, Agilent), the quantity was measured on Qubit 3.0. Illumina sequencing was performed on the Illumina NovaSeq6000 instrument with 2 × 151 run configuration. Read quality was visualized using FastQC (Andrews 2010), and then analyzed using rCorrector (Song and Florea 2015), which employs a k-mer based method to correct random sequencing errors in Illumina RNA-seq reads. Reads were assembled using the Trinity de novo assembly program (Grabherr et al. 2011).

The publication of the neuronal transcriptome assembly and transcriptomic data (e.g., BUSCO, GO, and KEGG analyses) is in preparation - since the neuronal transcriptome is not the main topic of this manuscript and was only partly used to this study, the assembly will be published and publicly available in another paper. Following the homolog searching approach of a paper by Cardoso and his co-workers (Cardoso et al. 2020), we used a broad group of relevant sequences (mammal, bird, amphibian, fish [agnathan, teleost, non-teleost]), and cephalochordate) as search queries (Supplementary Table 3; Supplementary Table 4). As an additional investigation, homolog searchers for other members of the Secretin superfamily were also carried out (Supplementary Table 5). As a supplementary investigation, we also made the homolog searching in (1) the public L. stagnalis CNS transcriptome shotgun assembly (sequence read archive: #DRR002012; (Sadamoto et al. 2012) and (2) the preliminary L. stagnalis reference genome data to which we have access as part of the genome consortium (genome publication in preparation).

Sequence analysis, phylogenetics, and expression of cluster B receptors

The conserved domain search was performed with the NCBI CDD/SPARCLE (Lu et al. 2020; Marchler-Bauer et al. 2017) to check if the key regions were present in the deduced Cluster B protein sequences. For 3D structure prediction, the computational modelling was made with AlphaFold2, a protein structure predicting algorithm (Jumper et al. 2021). The amino acid sequences were run through the GoogleColab platform (Bisong 2019) followed by a further examination with PyMOL (PyMOL 2020) and Swiss-PdbViewer 4.1.0 (Guex and Peitsch 1997).

The multiple sequence alignment used to generate the maximum likelihood tree consisted of the two newly identified L. stagnalis sequences and further 28 amino acid sequences obtained from the previous phylogenetic analysis of Cardoso and his co-workers (Cardoso et al. 2020). The alignment was performed using the ClustalW function with BLOSUM62 substitution matrix in Molecular Evolutionary Genetics Analysis v7 software30 (Kumar et al. 2016). The fitting model was LG with gamma-distributed rates. Bootstrapping support for the tree was conducted with 1000 bootstrap replicates, the bootstrap values (%) are indicated at each branch point. All positions containing gaps and missing data were eliminated. There was a total of 236 positions in the final dataset.

The CNS and heart were dissected from the animals (n = 5). Total RNA from the pooled tissues was isolated as presented above. To ensure that the samples are free from genomic DNA contamination, an additional DNase treatment was performed after the RNA preparation using TURBO DNA-free™ Kit (#AM1907, Thermo Fisher Scientific) following manufacturer’s instructions. The RevertAid H Minus First Strand cDNA Synthesis Kit (#K1631, Thermo Fisher Scientific) was used for reverse transcription (RT), applying random hexamer primers and 300 ng total RNA. PCR primers were designed for the identified L. stagnalis Cluster B sequences with the SnapGene® Viewer software (GSL Biotech, Chicago, IL, version 4.1.7). The applied primer set for the Cluster B sequence 1 was as follows: 5′ – CTA TTC CTG GCC TTC GTC CTA CG– 3′ (forward) and 5′ – AGA GGA CAC GGA CGA TGT TGA T – 3′ (reverse). The applied primer set for the Cluster B sequence 2 was as follows: 5′ – GAA CAG GTG GCT CGA ATA ACG AC– 3′ (forward) and 5′ – ATG GTG TTT CTT GAG CAG TGC AAC – 3′ (reverse) (Integrated DNA Technologies, Belgium). The PCR reaction was performed in 10 µL reaction volume at 95 °C for 4 min followed by 35 cycles of 95 °C for 30 s, 56 °C for 30 s and 72 °C for 15 s by a T1 Thermocycler PCR device (Biometra®, Germany). PCR of actin using RNA samples that had not been reverse transcribed was also performed to check for genomic DNA contamination (i.e. no RT control). PCR product was checked by agarose gel-electrophoresis using 2% gel (#16500-100; Thermo Fisher Scientific) and GeneRuler™ 100 bp Plus DNA ladder (#SM0328, Thermo Fisher Scientific).

PACAP-38 peptide synthesis

The synthesis of PACAP-38 peptide was performed using a solid-phase procedure with Fmoc-chemistry. Peptide chains were elongated on a Tentagel S-Ram resin (0.23 mmol/g) and the synthesis was performed using a CEM Liberty Blue machine. The peptide was detached from the resin by using a mixture containing 90% TFA, 4% water, 2% dithiothreitol, 2% triisopropylsilane, and 2% p-cresol. The resulted crude peptide was purified by reverse-phase HPLC using a Phenomenex Luna C18 (250 × 21.2 mm, 100 Å, 10 μm) column. The appropriate fractions were pooled and lyophilized (purity > 98%) and analyzed using liquid chromatography-electrospray ionization-mass spectrometry.

Immunohistochemistry

The heart was dissected from individual snails (n = 5) and pinned out on a Sylgard-coated dish containing 4% paraformaldehyde in 0.1 M phosphate buffer (pH = 7.4) overnight at 4 °C. After washing with phosphate-buffered saline (PBS; 137 mM NaCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, pH = 7.4), fixed tissues were cryoprotected in 20% glucose solution for 4 h at room temperature and embedded into Cryomatrix (#6,769,006, Thermo Fisher Scientific). Series of 12–14 μm-thick cryostat sections were mounted onto slides (#J3800AMNZ, Thermo Fisher Scientific). Next, the samples were incubated with PBS containing 0.25% TritonX-100 and 0.25% bovine serum albumin (BSA) (PBS-TX-BSA) for 1 h at 4 °C.

The sections were incubated with a sheep polyclonal anti-human PACAP-38 antibody (#ab35342, Abcam) diluted 1:1000 in PBS-TX-BSA or a rabbit anti-PAC1 receptor antibody (from Showa University, Tokyo, Japan; antigen: a synthetic peptide corresponding to the C-terminal intracellular domain conjugated to hemocyanin from giant keyhole limpet [KLH]) diluted 1:2000 in PBS-TX-BSA for 24 h at 4 °C. After the incubation, the samples were washed with PBS for 2 × 15 min. The PACAP-38-sections were further incubated with a rabbit anti-sheep IgG secondary antibody conjugated with horseradish peroxidase (HRP, #ab97130, Abcam) diluted 1:1000 in PBS-TX-BSA overnight at room temperature. The immunoreaction was visualized by adding 0.05% 3,3-diaminobenzidine (DAB) as chromogen and 0.01% H2O2 as substrate. The development was monitored under a stereomicroscope and stopped by changing the developing solution for 0.1 M Tris–HCl buffer. The PAC1 receptor-sections were further incubated with a donkey anti-rabbit IgG secondary antibody conjugated with NorthernLights™ NL557 (#NL004, R&D System) diluted 1:200 in PBS-TX-BSA overnight at room temperature. Finally, the sections were washed in PBS and mounted into mounting medium (DAKO, Glostrup, Denmark). The stained tissues were analyzed using a light and fluorescent microscope. Because the primary antibodies were generated against an antigen conjugated with KLH, to abolish cross-reaction to hemocyanin in tissues of L. stagnalis, the antibodies were preadsorbed overnight with KLH (#H7017, Merck). The specificity of the antibodies was tested applying method control and preadsorption test. In the case of method control, BSA was used instead of primary antibodies. In preadsorption tests, the diluted antibodies were mixed with their blocking peptides (synthetic human PACAP-38, 200 µg/mL) or PAC1-R blocking peptide (50 µg/mL, #BLP-VR003, Alomone labs) and were shaken overnight. Immunostaining was not present either in case of method control experiments or preadsorption tests.

Western blot analysis

The hemolymph (200 µL/animal; n = 5) was collected in a pre-cooled tube following the standard invasive protocol (Boisseaux et al. 2016; Hernadi et al. 2008) and stored on ice until use. The whole CNS (n = 10) and the heart (n = 5) were dissected from the animals, pooled, and extracted following a previously published protocol (Fodor et al. 2022; Hernadi et al. 2008). Briefly, the tissue samples were homogenized in 500 µL lysis buffer (1 M TRIS (pH = 6.8), 2% SDS, 10% glycerol, 1% mercaptoethanol, 10% protease inhibitory cocktail (#P2714-1BTL, Merck) with a Polytron PT 1200 homogenizer (Kinematica AG) and then further extracted with ultra-sonication. After centrifugation of the homogenates and the hemolymph (16,000×g for 5 min at 4 °C), the supernatants were placed in a new tube. 160 µL of the supernatants were diluted with 5x sample buffer (0.1 M TRIS (pH = 6.8), 0.5 M DTT, 2% SDS, 10% glycerol, 1 mM EDTA, 0.1% bromophenol blue) and denatured for 5 min at 95 °C.

For the WB analysis, following a previously published protocol (Fodor et al. 2022; Hernadi et al. 2008), 5 µL samples were run three times on a 10% SDS-PAGE, each with 3 µL protein marker. In the parallel SDS-PAGE, 20 µL samples were run three times for the MS analysis. The gel then was blotted for 3 h onto a nitrocellulose membrane (#GE10600002, Merck) following the standard semi-dry blotting protocol. The parallel gel was stained with Coomassie brilliant blue following the standard protocol. The membrane was incubated for 3 h with a blocking solution (10% non-fat dry milk powder and 6% BSA in Tris-buffered saline (TBS; 20 mM Tris, 150 mM NaCl, pH = 7.6), then washed thoroughly (3 × 5 min in TBS with 1% Tween-20 (TBST), then 3 × 5 min in TBS). The membrane was then incubated with the anti-PACAP-38 (#ab35342, Abcam) diluted 1:2000 in TBST-6% BSA overnight at 4 °C. After washing (3 × 5 min in TBS with 1% Tween-20, then 3 × 5 min in TBS), the membranes were incubated with a rabbit anti-sheep IgG secondary antibody conjugated with HRP (#ab97130, Abcam) diluted 1:5000 in TBS-Tween-20-BSA for 3 h at room temperature. After washing (3 × 5 min in TBST, then 3 × 5 min in TBS), the bands were visualized using enhanced chemiluminescence (Millipore). Synthetic PACAP-38 was used to the positive control experiments. In the negative control experiments, the samples were only probed with the secondary antibody and yielded no bands. In some cases, the anti-PACAP-38 antibody was preadsorbed with 10 µM of synthetic human PACAP-38 peptide overnight at 4 °C. This resulted in no bands (not shown).

Immunoprecipitation

The whole CNS (n = 15) was dissected from the animals, pooled, and homogenized in 1 mL lysis buffer (8 M urea, 20 mM TRIS-HCl pH = 8.0, 1 mM EDTA, 150 mM NaCl, 1% Triton X-100, 1 tablet of protease inhibitor cocktail (#11,836,170,001, Merck)) with a Dounce homogenizer and then further extracted with ultra-sonication. To remove urea and gain clean polypeptides and proteins, a chloroform-methanol total protein precipitation was performed. After centrifugation (16,000×g for 10 min at 4 °C), the supernatant was removed. The pellet was washed with methanol to remove the chloroform below the pellet. After centrifugation (16,000×g for 10 min at 4 °C), the pellet was resuspended in 1 mL buffer 1 (20 mM TRIS, 150 mM NaCl, 0.5% Triton X-100, pH = 7.5).

The magnetic immunoprecipitation was performed with a 12-Tube Magnet rack (#36,912, Qiagen). 200 µL suspensions of Protein A/G-conjugated magnetic beads (#88,802, Thermo Fisher Scientific) were placed in new tubes and the beads were fixed in the wall towards the magnet for 5 min. The containing buffer was removed and the fixed beads were washed with PBS three times. The anti-human PACAP-38 primary antibody diluted 1:500 in TBST-6% BSA (1 mL) was incubated with the magnetic bead – Protein A/G complexes for 1 h at room temperature with shaking. After the removal of the primary antibody solution, the whole resuspended sample (1 mL) was incubated with the magnetic bead – Protein A/G – primary antibody complexes for 2 h at room temperature with shaking. To check the reliability of the method, the procedure was also performed with 1 mL synthetic human PACAP-38 peptide solution (100 µg/mL) (positive control). The immunoprecipitated samples were washed one time with Buffer 1, two times with buffer 2 (20 mM TRIS, 150 mM NaCl, pH = 7.5), and two times with Buffer 3 (5 mM TRIS, pH = 7.5). Finally, the samples were eluted with 200 µL 1 M NH4OH and lyophilized.

HPLC-MS measurement

The lyophilized samples were re-suspended in a 6 M urea solution. Before the overnight enzymatic digestion with Trypsin/Lys-C Mix (#V5072, Promega, the USA), the samples were reduced and alkylated following the manufacturer’s technical bulletin. After the digestion, the peptides were cleaned on an Oasis HLB 96-well plate (HLB 30 mg, Waters, the USA), and the eluted samples were concentrated with an Eppendorf Concentrator plus system (Eppendorf, Hamburg, Germany). The concentrated peptides were resolved in 100 µL water containing 0.1% formic acid.

Nanoseparation-based proteomics analysis was performed with a Bruker Maxis 4G UHR-QTOF instrument (Bruker Daltonics, Bremen, Germany) coupled with Bruker EASY-nLC equipment. Three µL aliquots of the samples were injected and separated on a homemade C18 analytical column (3 μm, 75 μm x 150 mm) using a multistep gradient elution at a flow rate of 250 nL min− 1. We used the following eluent composition for the separation: A (aqueous formic acid solution: 0.1%) and B (acetonitrile/formic acid: v/v 99.9/0.1%). The scanning range was 300–2.200 m/z. The flow of nebulizer gas was 4 L min− 1 on 0.6 bar and the temperature was set at 200 °C. The capillary voltage was 4.5 kV, and the top 30 peptides were fragmented with a CID fragmentation cell.

For polypeptide/protein identification, the raw data were first processed with Data Analysis 4.4 software and then the fragment spectra were analyzed with the Mascot server V2.4.1 database search engine on the Swiss-Prot database or the given sequences we specified. Searching parameters were set to allow one missed cleavage site, accepting 50 ppm mass errors at the MS1 and 0.3 Da at the MS2 mode. For the polypeptide/protein identification, we used variable modification, including methionine oxidation and carbamidomethylation on cysteine as a fixed modification.

cAMP measurements in the heart

The heart (Supplementary Fig. 1) was dissected from individual snails (n = 10 animals/replicates) and pinned out on a Sylgard-coated dish containing Lymnaea physiological saline (50 mM NaCl, 2 mM KCl, 4 mM CaCl2, 4 mM MgCl2, 10 mM TRIS; pH = 7.5). The hearts were divided into control and synthetic PACAP-38-treated experimental groups (resulting in n = 5 animals/group/replicates). Based on a previously published protocol (Pirger et al. 2008, 2010b), specimens of the synthetic PACAP-38-treated group were first pre-incubated in Lymnaea physiological saline containing 0.5 mM 3-isobutyl-1-methyl-xanthine (#I5879, Merck) and 0.5 mM adenosine-triphosphate (#A9187, Merck) for 5 min, then incubated in Lymnaea Ringer containing 10 µM synthetic PACAP-38 for 10 min. The concentration of the PACAP-38 solution was chosen based on our preliminary results and our previous studies (Pirger et al. 2008, 2010a, b): the effects of vertebrate PACAP peptides have been investigated on snails at a concentration range of 10− 5 M − 10− 7 M with all concentrations were able to induce many physiological changes. In the present study, we chose the 10 µM (10− 5 M) concentration for a robust effect.

After the incubation, the hearts were pooled in both experimental groups, homogenized in cold acidic ethanol with a Polytron PT 1200 homogenizer (Kinematica AG), and centrifuged at 10,000×g for 10 min at 4 °C. The supernatants were collected, lyophilized (Vacuum Freeze Dryer, #BK-FD10S, Biobase), and dissolved in Tris–EDTA buffer. After centrifugation at 16,000×g for 10 min at 4 °C, the cAMP content was determined by the protein binding method using the Amersham cyclic AMP assay system (GE Healthcare UK Limited), according to the manufacturer instructions. Experiments were performed in 3 independent series.

Pharmacological experiments

For pharmacological experiments, we used the heart of L. stagnalis which is an excellent organ for studying chemical modulation (reviewed by Benjamin (Benjamin 2008). Its physiology is well-known: while heartbeat is generated by a muscle pacemaker located in the heart, its regulation is basically implemented by many neurotransmitters and neuropeptides which are released by several types of motoneurons into the heart (Benjamin and Kemenes 2013, 2020; Buckett et al. 1990a, b, c; Santama and Benjamin 2000; Willoughby et al. 1999a, b; Worster et al. 1998). All pharmacological experiments were carried out on isolated heart preparations using a previously developed refraction-based optical recording system (Pirger et al. 2023).

A three-way tap ensured rapid switching with minimum mixing of perfusates. The pressure heads containing Lymnaea physiological saline, 7 µM serotonin (5-HT) (#H9523-25MG, Merck), 2.5 nM acetylcholine (ACh) (#A6500-100G, Merck), and 10 µM synthetic PACAP-38 solutions were kept l-2 cm above the level of the heart. This was required for maintaining a steady beat. The test solutions were made just before the test. The concentration of 5-HT and ACh solutions was chosen based on the pharmacology of the heart of L. stagnalis described by previous studies (Benjamin 2008; Buckett et al. 1990a). Similarly to the cAMP measurement, the concentration of the PACAP-38 solution was chosen for a robust effect. Perfusions of test substance were of 1-min durations, Lymnaea physiological saline being perfused between each application until the heart had returned to a steady beat rate. The period of this normal saline perfusion varied according to the effects of the previous substance application but was never < 4 min.

Experiments were carried out to explore whether synthetic PACAP-38 had any direct or modulatory effect on heart physiology. To investigate the possible direct effect on the average amplitude of heart muscle contraction, 10 µM synthetic PACAP-38 solution was directly applied to the heart preparations (n = 16). To investigate the possible modulatory effects on the heart muscle responses (tonus, amplitude of contraction, relaxation), based on our previous experiments (Krajcs et al. 2015), the heart preparations (n = 6) were pre-treated with 10 µM synthetic PACAP-38 solution for 15 min before the direct application of 5-HT and ACh. The 7 µM 5-HT solution is known to increase both beat rate and amplitude, while 2.5 nM ACh solution is known to decrease both beat rate and amplitude (Buckett et al. 1990a).

Statistical analysis

Statistical analysis was carried out using OriginPro8 2018 software (OriginLab Corp., Northampton, Massachusetts, USA). The normality of the dataset was investigated using the Shapiro-Wilk test, homogeneity of variances between groups was investigated using the F-test. In the case of cAMP level measurements, differences between groups were analyzed by a two-sample t-test. In the case of average amplitude of contraction measurements, differences were analyzed by a paired t-test. In the case of 5-HT and ACh response experiments, comparisons between more than two independent groups were carried out using one-way ANOVA followed by multiple post hoc Tukey’s tests. Differences were considered statistically significant at least P < 0.05 (*,#). As a bias-reducing effort, the analysis of the data generated during the biochemical and pharmacological experiments was blinded.