Analysis of rhodopsin G protein-coupled receptor orthologs reveals semiochemical peptides for parasite (Schistosoma mansoni) and host (Biomphalaria glabrata) interplay

Schistosomiasis is a medically significant disease caused by helminth parasites of the genus Schistosoma. The schistosome life cycle requires chemically mediated interactions with an intermediate (aquatic snail) and definitive (human) host. Blocking parasite development within the snail stage requires improved understanding of the interactions between the snail host and the Schistosoma water-borne free-living form (miracidium). Innovations in snail genomics and aquatic chemical communication provide an ideal opportunity to explore snail-parasite coevolution at the molecular level. Rhodopsin G protein-coupled receptors (GPCRs) are of particular interest in studying how trematode parasites navigate towards their snail hosts. The potential role of GPCRs in parasites makes them candidate targets for new antihelminthics that disrupt the intermediate host life-cycle stages, thus preventing subsequent human infections. A genomic-bioinformatic approach was used to identify GPCR orthologs between the snail Biomphalaria glabrata and miracidia of its obligate parasite Schistosoma mansoni. We show that 8 S. mansoni rhodopsin GPCRs expressed within the miracidial stage share overall amino acid similarity with 8 different B. glabrata rhodopsin GPCRs, particularly within transmembrane domains, suggesting conserved structural features. These GPCRs include an orphan peptide receptor as well as several with strong sequence homologies with rhabdomeric opsin receptors, a serotonin receptor, a sulfakinin (SK) receptor, an allatostatin-A (buccalin) receptor and an FMRFamide receptor. Buccalin and FMRFa peptides were identified in water conditioned by B. glabrata, and we show synthetic buccalin and FMRFa can stimulate significant rates of change of direction and turn-back responses in S. mansoni miracidia. Ortholog GPCRs were identified in S. mansoni miracidia and B. glabrata. These GPCRs may detect similar ligands, including snail-derived odorants that could facilitate miracidial host finding. These results lay the foundation for future research elucidating the mechanisms by which GPCRs mediate host finding which can lead to the potential development of novel anti-schistosome interventions.


Scientific Reports
| (2022) 12:8243 | https://doi.org/10.1038/s41598-022-11996-x www.nature.com/scientificreports/ ment of Agriculture, Fisheries and Forestry (DAFF). Mice were euthanized with CO2 gas and their livers were perfused with chilled PBS. Eggs of S. mansoni were collected during perfusion of mice. Four infected mouse livers were sliced with scalpel blades and blended to a smooth consistency in 50 mL phosphate buffered saline (PBS). A two-day protocol was used to obtain relatively clean schistosome eggs and miracidia 34 . In brief, the mixture of eggs and mouse liver tissue were incubated with collagenase B, penicillin and streptomycin at 37 °C overnight, followed by fractionation using Percoll columns (8 ml Percoll + 32 ml of 0.25 M sucrose in 50 ml tubes). The egg pellets were washed using PBS twice on a second Percoll column (2.5 ml Percoll + 7.5 ml 0.25 M sucrose in a 15 ml tube). Purified eggs were transferred into a 200 ml hatching measuring cylinder wrapped completely in light-blocking black tape with the exclusion of the top 4 cm from the lip, thereby producing a light-gradient. The hatching cylinder was topped with pH neutral MilliQ water until ~ 1.5 cm above the tapecovered area and exposed to bright light at 27 °C. Eggs were incubated for 6 h post-hatch, and the top 10 ml of miracidium-containing water was collected for miracidia isolation. Hatched miracidia were isolated by centrifugation at 8,000 × g for 1 min at 4 °C, and washed twice with water. For RNA isolation, miracidia were collected at 6 h post-hatch and kept at -80℃. Total RNA was isolated from S. mansoni miracidia (6 h post-hatch in triplicate) using TRIzol reagent following the manufacturer's user guide (Invitrogen, USA), the RNA quantity and quality were assessed by UV spectrophotometry (NanoDrop ND-1000) and the RNA was sent to NovoGene (Hongkong) for next-generation Illumina 2500 platform RNA-seq. Raw sequence data was deposit into the GenBank NCBI under accession number SRR17224866. De novo transcriptome assembly of the S. mansoni miracidia raw sequence data was performed using Trinity, as previously described 35,36 and contigs were translated into protein sequences using Transdecoder 35 . Gene expression levels of the S. mansoni miracida were calculated by mapping raw sequence data against the S. mansoni reference genome derived from WormBase Parasite (https:// paras ite. wormb ase. org/ Schis tosoma_ manso ni_ prjea 36577/ Info/ Index/) using CLC Genomic Workbench with default parameters 37 .
Gene identification and functional annotation. The pipeline for identification of ortholog GPCRs shared between B. glabrata and S. mansoni is shown in Fig. 1 Putative GPCRs identified in S. mansoni miracidia were used to query (using tBLASTp) the B. glabrata GPCR database. Those sequences with E-values < 1.0E-20 were retrieved and screened for the presence of recurrent transmembrane motifs using TMHMM (http:// www. cbs. dtu. dk/ servi ces/ TMHMM/). Those containing 7 transmembrane (TM) domains were selected for further analysis. Multiple sequence alignments were created with Molecular Evolutionary Genetics Analysis (MEGA) software version 6.0 38 with the MUSCLE 39 algorithm. Phylogenetic trees were constructed using the neighbor-joining method with 1000 bootstrap replicates for node support. Gene ontology mapping and functional annotation were applied by using OmicsBox (BioBam) 40 . The final phylogenetic tree and heatmap were modified with iTOL v5 41 . Miracidia behaviour in response to test solutions. Test solutions. Synthetic FMRFa (FMRF-NH 2 ), buccalin (RLDRFGFAGQL-NH 2 ) and SK (NYGDYGIGGGRFGR) were provided by China Peptides (Shanghai, China) (purity ≥ 95%). Serotonin (5-hydroxytryptamine/5-HT) was provided by Sigma (Burlington, United www.nature.com/scientificreports/ States) (purity ≥ 98%). Stock solutions of FMRFa (100 μM), buccalin (100 μM), SK (100 μM) and 5-HT (5 nM) were prepared by dissolving in MilliQ water. MilliQ water was used as a negative control in bioassays.
Miracidia collection and assay. Isolation of S. mansoni miracidia at ~ 2 h post-hatch was performed as described in Wang et al. 21 . For each assay, 30 ± 5 actively swimming miracidia in pH neutral MilliQ water (~ 4 ml in total) were evenly distributed with a pipette to the central region of a Petri dish (100 mm × 15 mm) containing 4 ml of MilliQ water. The swimming area for the miracidia was covered to prevent light bias prior to analysis under the microscope. For assays, 2 µl of test solution were added to the central area of the Petri dish. Some diffusion of each molecule was expected over the 1 min test period. Assays were also tested at 10 × and 100 × serial dilutions.
To record miracidia movement before and after addition of the test solutions, an inverted compound microscope with videoing capacity (OLYMPUS CKX41), fitted with an OLMPUS DPI Digital Microscope Camera DP22 (2.8-megapixel image at a rate of 15 frames per second), was used. The real camera's field of view (FOV) was 2.500 × 1.875 mm. Miracidia movement was recorded for 1 min before and after the addition of each test solution and captured videos were processed using Tracker 4.87.
Analysis. Miracidia trajectories were tracked manually from entrance into the FOV to exit, or up to 1 min for those that remained within the FOV. Only miracidia that had been swimming for more than the length of the short edge (7.5 cm) of the FOV were included before and after solution addition (File S1); those that did not were considered statistically meaningless. The average time duration of miracidia staying within the FOV was considered as another key behavioural feature and was statistically compared. For those miracidia staying for more than 1 min after addition of solution, the time duration within that 1 min was used for comparison, and the mean acceleration value was calculated. Miracidium acceleration and velocity were calculated based on trajectories, with units converted to cm s -2 and cm/s. A paired two-tailed t-test was used to calculate P-values; in addition, wherever applicable, two-way ANOVA analysis 42 was performed to evaluate the significance of changes (acceleration, velocity and time) among the test solutions and the negative control.

Identification of candidate peptide ligands from B. glabrata-conditioned water extract.
To identify whether candidate peptide ligands were present in B. glabrata-conditioned water, two approaches were taken. First, mass spectrometry data derived from prior analysis of B. glabrata-conditioned water 43 was searched using target precursor proteins. Second, antibody-mediated dot blot analysis was performed using B. glabrata-conditioned water extracts. B. glabrata were washed with pH neutral MilliQ water and placed in beakers containing 20 ml water for 2 h at room temperature (RT). Snails were removed, conditioned water was collected from 20 snails (in different aquaria), and 20 ml methanol was added to each beaker and mixed thoroughly. The conditioned water was filtered through PVDF Millex-HV syringe filter units (0.45 µm) to remove particles and microbes. The filtrate was snap frozen and lyophilized. For negative controls, water not previously exposed to snails was similarly processed. When required for dot bot assay, samples were rehydrated with 200 μl MilliQ water and centrifuged at 12,000 rpm for 5 min. The supernatant was collected and diluted 1:1 in MilliQ water. Quantitation was performed using a NanoDrop 2000c UV-Vis spectrophotometer (Thermo Scientific, Waltham, U.S) at A280nm. B. glabrata extracts at 5 gµ/µ l, 500 ng/µ l and 50 ng/µ l were applied (2 μl) onto a nitrocellulose membrane (0.45 mµ , BioRad, Hercules, U.S) that had been pre-soaked in 1 × phosphate buffered saline (PBS) and air-dried at room temperature for 10 min. The membrane was incubated in blocking buffer (2% (v/v) skim milk in PBS) for 1 h, and primary antibody was added [1:500; rabbit anti-buccalin 44 , rabbit anti-FMRFa (Genscript, Piscataway, U.S), rat anti-5-HT (Sigma)] for 1 h at RT. Membranes were washed with PBS-Tween20 (0.1% (v/v)) and incubated for 1 h at RT with secondary antibody [1:5000; anti-rabbit Ig-IR 680 (LI-COR, Lincoln, U.S) or anti-rat Ig-IR 795 (LI-COR)]. Following washes (3x, 5 min) in PBST, blots were visualized using an Odyssey CLx, LI-COR machine.

Results
Identification of ortholog GPCRs shared between B. glabrata and S. mansoni miracidia. In total, 96 proteins with 7-TM domains were extracted from the S. mansoni transcriptome-derived protein models based on TMHMM prediction. Phobius prediction led to the identification of 139 proteins with 7-TM domains. Pfam profiling of both predictions led to the classification of 87 proteins (E-value < 0.0004) as rhodopsin-type receptors. BLASTp analysis of these GPCRs against all B. glabrata rhodopsin GPCRs showed significant matches (E-value < 1.0E−20) for 8 GPCRs (Table 1), with between 26 to 48% amino acid identity (File S2a). BLASTp searches using all S. mansoni and B. glabrata ortholog GPCRs against the NCBI non-redundant (nr) database returned best match hits for GPCRs, including 5-HT (serotonin), allatostatin-type A (AST-A), FMRFa, SK, orexin type 2, neuropeptide F, orphan peptide and opsin-like GPCRs (Table 1, Fig. 2A and File S2b). In B. glabrata, all GPCRs were expressed in the CNS, while the orphan rhodopsin GPCR had the widest tissue distribution (Fig. 2B). In S. mansoni miracidia, the rhodpopsin GPCR ortholog was found to show the highest average level of expression, compared to other receptors (Fig. 2B).  6,46 , so were tested for bioactivity on miracidia. The trajectories of miracidial movement before and after addition of the FMRFa and buccalin peptides (2 µl at 100 µM) are compared in Fig. 3A,B. Representative movies can be viewed in Movies S1 and S2. Before addition, miracidia generally swam across the FOV in a direct, linear path (Fig. 3A,B-Before). Following application, miracidia showed localized movement within the FOV, as well as increased circular swimming (Fig. 3A,B-After). Upon application of buccalin or FMRFa peptides, miracidia within the FOV swam for longer, except for buccalin at 10 µM (P-value = 0.2964) (Fig. 3C). The change in acceleration were significant after the addition of buccalin or FMRFa (100 µM and 10 µM) (Fig. 3D). Peptides stimulated a swimming pattern concentrated around the location of the peptides. In contrast, following application of 5-HT (5 nM; Movie S3), the time within the FOV was insignificant, yet the change of acceleration was significant (File S3). Peptides (100 µM) resulted in miracidia staying longer in the region where the peptide was added, but there was no significant change in average velocity, as indicated by two-way ANOVA analysis (File S4). There was also no significant change in the time within FOV, nor in average velocity, upon addition of the SK peptide at 3 different concentrations (100 µM, 10 µM and 1 µM; Movie S4). As negative controls, MilliQ water was tested on S. mansomi miracidia and no behavior changes was observed.

Detection of buccalin and FMRFa peptides in B. glabrata-conditioned water. The B. glabrata
buccalin precursor contains numerous buccalin-like peptides (Fig. 4A). The most conserved region in the buccalins is a C-terminal FXGGIG, which following post-translational processing, becomes an amidated peptide. Dot blots performed on conditioned-water extracts from B. glabrata showed the presence of a buccalin-like peptide at extract dilutions from 10 μg to 0.1 μg (Fig. 4B). The B. glabrata FMRFa precursor contains numerous FMRFa related peptides (FaRPs), including FMRFa, FLRFa and FIRFa (Fig. 4C). Analysis of proteomic mass spectrometry data derived from naïve B. glabrata snail-conditioned water 43 , identified 4 peptide matches for the FMRFa precursor, although not specifically within any FaRPs. Dot blots performed on conditioned-water extracts from B. glabrata showed the presence of peptide(s) with similarity to FaRPs at extract dilutions from 10 μg to 0.1 μg (Fig. 4D).

Discussion
S. mansoni miracidia respond to snail-derived biomolecules 43 , although the precise identity of the active biomolecule(s) has not been clearly defined. One study implicated "miracidia-attracting glycoproteins" present within the snail mucus 12 , while in silico analysis from B. glabrata snail conditioned-water proteins predicted interactions of uncharacterized S. mansoni proteins with B. glabrata proteins 43 . Peptides have also been implicated, whereby a snail-derived novel peptide (named P12) stimulated changes in the behaviour of the S. mansoni miracidia 47 .
In this study, to narrow down biomolecules potentially involved in the parasite and host interplay, we utilised gene resources from both the B. glabrata and S. mansoni to identify ortholog GPCRs that are likely used by each organism to detect similar ligands. We reported that 8 S. mansoni miracidia GPCRs share significant identity with 8 B. glabrata GPCRs, not only in GO mapping, but also within regions corresponding to putative TM domains. These include GPCRs with similarity to neuropeptide GPCRs that bind FMRFa, AST-A/buccalin and sulfakinin peptides. We propose that the miracidial ortholog GPCRs may be used for neural signaling, requiring a common ligand, and/or are used to detect semiochemical biomolecules present within the water. The latter expectation was validated by miracidial behaviour changes in the presence of snail FMRFamide and AST-A/buccalin peptides.  48 where they are involved in multiple functions such as inhibition of juvenile hormone biosynthesis and reduction of food intake 48 , AST-A-like neuropeptides have been identified in gastropods and bivalve molluscs, including Lottia gigantea, Theba pisana, Aplysia californica and Crassostrea gigas [49][50][51][52] . Buccalin, named following its first identification in the accessory radula closer muscle of A. californica 53 , has been implicated in various activities in molluscs such as the inhibition of muscle contraction, regulation of feeding and spawning [53][54][55] . Also in gastropods and bivalves, the AST-A/ buccalin receptor was identified through in silico analysis of publicly available genomic datasets including that of B. glabrata 56 . In our study, we identified an AST-A/buccalin receptor ortholog in S. mansoni, although there are no reports that S. mansoni has a buccalin-like peptide. In fact, a comprehensive neuropeptide investigation of 10 platyhelminth species showed that only the free-living turbellarian Macrostomum lignano has a buccalin-like peptide (npp-9 gene; GAYSGFLG) 57 . We identified a buccalin-like peptide in the B. glabrata conditioned water. Despite the presence of neuropeptides in mucus secretions having not been well investigated, we previously identified neuropeptides (including buccalin) within the salivary gland mucus of the land snail T. pisana 58,59 .
The FMRFa was first discovered in the hard clam Mercenaria mercenaria and it is thought to have a pleitropic role in molluscan physiology [60][61][62][63] . Extensive studies performed on the freshwater snail Helisoma showed that FMRFa and related peptides are densely concentrated not only in the nervous system but also within the salivary  66 . The S. mansoni genome contains a gene encoding a FLP precursor (npp-13 gene) that may be processed to release two RFamide peptides (HFMPQRFa and YTRFVPQRFa) 57 . A synthetic FLP (GNFFRFa) derived from non-schistosome platyhelminth precursors stimulates contraction of S. mansoni muscle fibres in vitro 67 . An FLP receptor has also been reported in the turbellarian flatworm Girardia tigrina based on sequence similarity and a receptor calcium mobilization assay 68 . The homolog receptor in S. mansoni miracidia was investigated in the current study due to its similarity with the B. glabrata FMRFa receptor. www.nature.com/scientificreports/ Our behaviour assays also indicated that snail-derived FMRFa can be detected by S. mansoni miracidia due to their staying significantly longer in FOV and the increased acceleration of miracidia, supported by the observed presence of FMRFa precursor peptides in B. glabrata conditioned water. However, as S. mansoni has the potential to generate endogenous FLPs, we cannot preclude the possibility that the applied FMRFa may stimulate endogenous effects, leading to the observed miracidial behaviour changes. The monoamine 5-HT plays a critical role in neural transmission and has been very well documented throughout eumetazoans, as has the conservation of 5-HT GPCRs. In adult S. mansoni, 5-HT stimulates motor activity 69 , while in the miracidia, an immunofluorescent approach localized 5-HT to within sensory nerves 70 . The 5-HT GPCR was identified within our interspecies GPCR ortholog analysis yet we found that 5-HT at 5 mM did not modify miracidial behaviour, while the significant change in acceleration could be attributed to its high concentration.
Sulfakinin is a sulfated neuropeptide best known for its function as a satiety (food intake) factor 71 . In silico data-mining showed that molluscan SK has the C-terminal RF(W)amide sequence common to insect sulfakinins, as well as the DY motif shared by both insect SKs and vertebrate cholecystokinin (CCK) 72 . Since vertebrate CCKs and insect SKs reveal similar biological function relating to digestive enzyme secretion, satiety and smooth muscle contraction 73 , it is possible that their molluscan counterparts have retained similar basic biological activities. In contrast, there is no obvious SK in S. mansoni, suggesting that the parasite may only recognize the B. glabrata SK, either as a secreted semiochemical, or once it penetrates the snail as a guidance peptide to navigate to the hepatopancreas where it proliferates 74 . Our behavior assays demonstrated that SK did not alter miracidial behavior (neither the velocity nor duration present under FOV were affected), and therefore it is more likely to act as an internal stimulus in S. mansoni.
FMRFa and buccalin peptides may contribute to a cocktail of biomolecules that could be used as an effective, species-specific attractant. Our serial dilution assays suggested sustained bioactivity for both buccalin and FMRFa peptides at a concentration of at least 1 µM. We also report 1 orphan peptide GPCR ortholog within B. glabrata and S. mansoni miracida, which is consistent with the possibility that uncharacterized species-specific peptides could help attract miracidia to the appropriate snail host due to its presence in many tissues of B. glabrata and its high expression level in S. mansoni miracidia.

Conclusions
To minimise transmission and reduce schistosomiasis prevalence, interference with the snail-miracidium interaction is a promising plan of biocontrol. We characterised ortholog GPCRs shared between B. glabrata and S. mansoni miracidia, important biomolecules commonly used for chemosensory communication. The B. glabrata buccalin and FMRFa GPCRs represented good targets for bioassay, the results from which indicated that buccalin and FMRFa stimulated miracidial behaviour changes, despite the fact that homologs of buccalin-like peptides are not present in S. mansoni. These GPCRs could present novel targets for the development of anti-helminthic compounds to be applied to lakes and specifically interfere with Schistosoma detection of snail host. For greater species-specificity, we suggest that deorphanizing the ortholog orphan peptide GPCR will be most advantageous. These findings further help our understanding of chemosensory interaction between parasites and their hosts, particularly within aquatic environments.