Bioinformatic analysis

Gene models coding for metabotropic 5-HT receptors were searched using the terms “5-hydroxytryptamine receptor”, “serotonin receptor”, “biogenic amine 5-HT receptor”, “G protein coupled 5 hydroxytryptamine receptor” in Echinococcus multilocularis [31], Echinococcus granulosus [32], Echinococcus canadensis [33] and Mesocestoides corti (http://parasite.wormbase.org/Mesocestoides_corti_prjeb510/Info/Index/, Helminth Genomes Consortium) databases. The retrieved gene models were also used as a query in the Blast tool from the NCBI for nucleotide (BLASTN 2.7.0+) and proteins (BLASTP 2.7.0+). The presence of conserved domains in the gene models found was determined using the Conserved Domain Database tool from the NCBI (https://www.ncbi.nlm.nih.gov/cdd/). The prediction of hypothetical open reading frames from nucleotide sequences was performed using the translate tool from the expasy proteomic tools (http://web.expasy.org/translate/). The localization and length of transmembrane domains was predicted using TMpred (http://embnet.vital-it.ch/software/TMPRED_form.html). The multiple sequence alignment was performed using the program MULTALIN [39] from PRABI (https://npsa-prabi.ibcp.fr/cgi-bin/npsa_automat.pl?page=/NPSAHLP/npsahlp_alignmultalin.html). The multiple sequence alignments were then visually inspected and manually edited when necessary. The molecular weight and isoelectric point of the predicted proteins was calculated with the program Compute pI/Mw tool (http://web.expasy.org/compute_pi/).

Cloned 5-HT_7Egran1, 5-HT_7Egran2 and 5-HT_7Mco1 and predicted cestode sequences gene models ECANG7_00799 (5-HT_1Egran1), ECANG7_02049 (5-HT_1Egran2) and MCOS_0000684301 (5-HT_7Mco2) were aligned with cloned serotonin receptor sequences from the following invertebrates; Dugesia japonica (5-HT_1Dj1 to 5-HT_1Dj3, 5-HT_4Dj1 to 5-HT_4Dj5, 5-HT_7Dj1 to 5-HT_7Dj7, PMCID: PMC4569474), Schistosoma mansoni (5-HT_7Sm, GenBank accession number KX150867), Caenorhabditis elegans (UniProt IDs G5EGH0 for 5-HT_1Ce, O17470 for 5-HT_2Ce and Q22895 for 5-HT_7Ce) and Drosophila melanogaster GenBank accession numbers CAA77570.1 (5-HT_1ADro), CAA77571.1 (5-HT_1BDro), CAA57429.1 (5-HT_2ADro), NP_001262373.1 (5-HT_2BDro) and NP_524599.1 (5-HT_7Dro). Alignment of amino acid sequences was performed with Clustal Omega v1.2 [40], gaps removed with Gap Strip/Squeeze v2.1.0 (75% gap tolerance) [41], and an unrooted maximum likelihood phylogenetic tree was generated with PhyML v3.1 (500 bootstrap replicates) [42].

For molecular modelling studies, proteins studied in this work were searched using BLAST [43] against UniProtKB/Swiss-Prot databases. Protein domains were screened against PFAM, and Prosite databases using PFAM_scan [44] or HMMscan 3.0. Protein structure models were obtained using PHYRE2 [45] and SWISS-MODEL [46–49]. PDB database was used for homology searching and 3SN6 and 4IAR with ergotamine ligand (http://www.rcsb.org/pdb) [50] was used for structural comparison analyses. For all the new sequences, the template used to model the transmembrane segment of the protein was the deposited structure 3SN6, and the internal segment (intracellular loop 3) modelled using Phyre using the template 2RH1. Ramachandran Plots were produced using Rampage (http://mordred.bioc.cam.ac.uk/~rapper/rampage.php). Relevant and conserved protein residues for the ligand binding were identified and its distances from the ligand were measured. Analysis of the 5-HT_7Egran1 modeled protein revealed that 91.7% of the residues are within the favoured regions, 4.3% in allowed regions and 4% in outlier regions of the Ramachandran plots. For the modeled 5-HT_7Egran2 protein, 88.6% of the residues are within the favoured regions, 8.3% in allowed regions and only 3,1% in outlier regions were as for 5-HT_7Mco1, 85.3% of the residues are within the favoured regions, 9.5% in allowed regions and 5,2% in outlier regions. All this data shows the good quality of the generated homology models (S1A, S1B and S1C Fig).

Chemicals

Compounds UCM120 and UCM2550 were synthesized as described previously [38, 51] and dissolved in DMSO before dilution to the needed concentration. Ergotamine and tryptamine were also dissolved in DMSO whereas Lysergic acid diethylamide (LSD), tyramine, octopamine, acetylcholine, histamine and dopamine were dissolved in distilled H₂O. 5-HT was dissolved in distilled H₂O at stock concentration of 5 mM. Stock solutions were either made up on the day of the experiment or taken from aliquots stored at -80°C for no longer than 1 week prior to use. After 0.22 μm filtration with Millex GV filter units (Millipore, Ireland), stock solutions were diluted to the corresponding final concentration (e.g. 0.1; 1; 10; 100; 500; 1000 and 2000 μM for 5-HT) in RPMI medium with high glucose (Gibco, USA). 5-HT (H9523), LSD (L7007), ergotamine (E1200000), tryptamine (193747), tyramine (T90344), octopamine (O0250), acetylcholine (A6625), histamine (H7125) and dopamine (H8502) were obtained from Sigma Chemical Company.

Parasite material

Echinococcus granulosus protoscoleces were obtained under sterile conditions by needle aspiration of hepatic hydatid cysts of porcine origin, provided by abattoirs from Buenos Aires and Santa Fe provinces, Argentina. The livers used for parasite extraction were from animals that were not specifically used for this study and all the material obtained was processed as part of the normal work of the abattoir. Samples from animals at the abattoir were collected under consent from local authorities. Protoscolex viability was assessed using the eosin exclusion test after three washes with PBS, with 50 μg/ml of gentamicin to remove cyst wall debris [52]. Only samples showing more than 95% viability were used. A fraction of the protoscoleces was used for probe imaging, other fraction for motility experiments and the remaining protoscoleces were used for species/genotype determination by sequencing a fragment of the mitochondrial cytochrome c oxidase subunit 1 (CO1), as previously described [53]. The resulting species and genotype of all protoscoleces used in this work were from Echinococcus canadensis G7. Three biological replicates were used with each replicate corresponding to protoscoleces obtained from a single cyst. Mesocestoides corti larvae (tetrathyridia) were maintained by alternate, serial passages in Wistar female rats and BALB/c female mice as previously described [54]. The tetrathyridia larvae used in drug testing were obtained after 3 months of intraperitoneal inoculation. Only tetrathyridia from up to the third serial passage in mice were used for the experiments. Three biological replicates were used with each replicate corresponding to tetrathyridia obtained from a single mouse host. Larvae freshly collected from mice were washed three times in PBS and used immediately for drug tests.

Wistar female rats and BALB/c female mice were housed at the animal facilities of Instituto de Investigaciones en Microbiología y Parasitología Médica (IMPaM), Facultad de Medicina, Universidad de Buenos Aires (UBA)-Consejo Nacional de Investigaciones Científicas y Tecnológicas (CONICET), Buenos Aires, Argentina, in a temperature-controlled light cycle room with food and water ad libitum.

Ethics statement

Experiments involving the use of experimental animals were carried out according to protocols approved by the Comité Institucional para el Cuidado y Uso de Animales de Laboratorio (CICUAL), Facultad de Medicina, Universidad de Buenos Aires, Argentina (protocol “in vivo passages of cestode parasites from Mesocestoides corti” number CD N° 1127/2015). Cyst puncture was performed following the approved protocol by the same institution (protocol “Hydatid cysts puncture from natural infections” number CD N° 3723/2014).

RNA extraction and cDNA synthesis

Total RNA from E. granulosus was extracted from protoscoleces that were crushed under liquid nitrogen and processed using Trizol reagent (Invitrogen). The RNA obtained was treated with RNase-Free DNase (Fermentas), ethanol precipitated and reverse transcribed using Superscript III reverse transcriptase (RT) (Invitrogen) and gene specific reverse primer complementary to 5-HT GPCR gene models (S1 Table). One cDNA for each selected gene model was synthesized. The same procedure was followed to obtain RNA from M. corti tetrathyridia and cDNA synthesis.

Amplification and cloning of cDNAs coding for serotoninergic GPCRs

The gene models ECANG7_06088 and ECANG7_06092 from E. granulosus and MCOS_0000684401 from M. corti, retrieved as explained in the section “Bioinformatic analysis” were used for primer design (S1 Table) and a PCR product for each serotoninergic GPCR was obtained. The cycling parameters for ECANG7_06088 were 95°C for 5´ (initial denaturalization), then 98°C for 20´´, 56°C for 30´´ and 72°C for 2´ (repeated 5 times), then 98°C for 20´´, 66°C for 30´´ and 72°C for 2´ (repeated 35 times) and finally, 72°C for 10´ (final extension). For ECANG7_06092 were 95°C for 5´ (initial denaturalization), then 98°C for 20´´, 55°C for 30´´ and 72°C for 2´ (repeated 5 times), then 98°C for 20´´, 64°C for 30´´ and 72°C for 2´ (repeated 35 times) and finally, 72°C for 10´ (final extension). Finally, for MCOS_0000684401 were 95°C for 5´ (initial denaturalization), then 98°C for 20´´, 55°C for 30´´ and 72°C for 2´ (repeated 5 times), then 98°C for 20´´, 72°C for 30´´ and 72°C for 2´ (repeated 35 times) and finally, 72°C for 10´ (final extension). PCR reactions were performed using the KAPA HIFI polymerase (Biosystems) using the cDNA previously obtained with each reverse primer as a template. Amplification products were visualized by agarose gel electrophoresis and Gel Red staining and the bands of interest were extracted from the gel using the QIAquick Gel Extraction Kit (Qiagen), used for a non-templated adenine adding or A-tailing procedure employing a nonproofreading DNA polymerase (Pegasus, Embiotec) and finally cloned into the TOPO TA Vector (Invitrogen). The recombinant plasmids were used for Escherichia coli (DH5α) transformation and the transformed bacteria were grown in LB with ampicillin and kanamycin.

The selected colonies were then used for plasmid purification using the GeneJet Plasmid miniprep kit (Fermentas) and sequencing using an Applied Biosystems Big Dye terminator kit (Applied Biosystems) on an ABI 377 automated DNA sequencer. The cloned cDNA products obtained with E. granulosus primers were designed from the ECANG7_06088 and ECANG7_06092 gene models and named 5-HT_7Egran1 and 5-HT_7Egran2 respectively (S1 Text) following the naming scheme proposed by Tierney [55]. The cDNA cloned from the gene model MCOS_0000684401 from M. corti was named 5-HT_7Mco1 (S1 Text).

Motility index measurement

For the motility index measurement, the method of Camicia et al. [16] was followed with minor modifications. In the case of E. granulosus, protoscoleces were mantained in RPMI medium with high glucose (Gibco, USA) with 50 μg/ml of gentamicin for no more than 48 hours in order to avoid the potential differentiation of the protoscolex towards premicrocyst [16]. The day before the experiment the protoscoleces were transferred to U-shape 96-well microplates (Greiner Bio-One, Germany) with approximately 125 parasites per 100 μl per well. This number of parasites per well was empirically determined as the best number of worms to perform the experiments. Movement was measured as described for C. elegans [56] using a worm tracker device (WMicrotracker Designplus SRL, Argentina) which determines the motility index by the light-microbeam (100 μm wide; k = 880 nm; intensity <1 mW) scattering produced by the movement of parasites. Quantification of the signal was performed by applying a mathematical algorithm previously described for sperm viability determination (Patent number: US4176953, year of submission: 1978, year of publication: 1979).

To study the effect of 5-HT on the motility of M. corti tetrathyridia, only one worm per well was used as this was the empirically determined optimal number. Addition of more tetrathyridia to each well plate did not result in a better signal. M. corti tetrathyridia were used fresh, immediately after the mice was opened. One tetrathyridium per well in 100 μl of RPMI with high glucose (Gibco) was pipetted to U-shape 96-well microplates and incubated overnight at 37°C in a 5% CO₂ atmosphere. The following day, basal activity recordings were performed for 2 h at 37°C in the worm tracker. After the initial 2 h, 25 μl of the same medium containing 5-HT at the indicated concentrations were added and the motility of parasites was recorded in the resultant 125 μl of medium for another 2 hours. Twenty-five microliters of fresh medium alone were added to the controls before registering motility in the resultant 125 μl. In order to avoid any potential difference between individuals not related to the effect of the 5-HT, the activity found in each well was divided by the activity found in the same well in the basal record. Each experiment was performed using 16 technical replicates per treatment condition. In order to count with biological replicates, four independent experiments, with tetrathyridia coming from different mice were performed.

Effect of the 5-HT GPCR agonist UCM2550 on cestode larval motility

We evaluated the effect of UCM2550, a 5-HT_1A agonist in mammals and the parent ligand of UCM120, [38, 57] on protoscolex movement. Protoscoleces were incubated in 96-well microplates as described in the section “Motility index measurement”, and basal activity was registered for 2 hours at 37°C. After this first acquisition they were subjected, in groups of 16 replicates, to the inidicated concentrations of the previously mentioned drug, and recorded for an additional 2 hours in the resultant 125 μl. Control wells were incubated in 125 μl of medium with or without 2% of DMSO (the vehicle used to dissolve the drugs) and no differences were observed between those groups. Three independent experiments, with protoscoleces from different hydatid cysts (biological replicates), were performed. The effect of the drug on protoscolex viability was tested for toxicity at the maximal concentrations used.

Measurement of motor activity by video imaging

In some experiments, alongside to the worm tracker motility analyses, the parasite motor activity was also measured with a digital video camera (Kodak easy share Z915 digital camera) coupled to an inverted microscope (Nikon, model TMS-F). Fresh parasites were processed in the same way as in previous sections in U-shape 96-well microplates, then incubated overnight at 37°C in a 5% CO₂ atmosphere and finally video recorded with a digital camera after drug treatment. Images were obtained for a period of 15 seconds and the data were analyzed with ImageJ software (version 1.48, NIH, USA). Protoscoleces showed complex motor movements characterized by sucker and rostellum movements and repeated body bends. To quantify this type of movements the method of Patocka et al. [18] was followed with some modifications. The videos were transformed to stacks of TIFF images by using the Filmora video editor (Wondershare Filmora, version 8.2.3, downloaded from https://filmora.wondershare.es/) and then imported into ImageJ. First, the stacks were first converted to 8 bit type images. Next, the background was subtracted and then converted to a binary image by threshold adjust. Then, the maximum intensity of Z projection (which shows all the elements in the movie) was calculated together with the minimum intensity projection (which shows only the constant elements in the movie). Then subtracting the minimal from maximal projections with the “image calculator” it was possible to remove everything constant in the movie, including immobile animals. Finally, the resultant pixels were quantified using the “Analyze particles” command and the integrated density in the summary result table showed the pixels that changed during the course of each movie. This procedure was repeated for each video and a graphic representation of pixel change was done for each treatment (S2 Fig). Eight technical replicates and three independent experiments, with protoscoleces coming from different hydatid cysts (biological replicates), were performed.

Whole mount probe imaging

Protoscoleces were fixed for 4 hours at room temperature (RT) in 4% (w/v) paraformaldehyde (PFA) in PBS (pH 7.4) and washed for 24 hours at RT in PBS containing 0.35% (v/v) triton X-100 (TX-100), 0.1% (w/v) sodium azide (NaN₃), 0.1% (w/v) BSA (Sigma), 6.25 x 10⁻³ (w/v) digitonin (Dig) and 0.5% DMSO (PBS/TX-100/NaN₃/BSA/Dig/DMSO). Approximately, 10 μl of pellet of protoscoleces were aliquoted in eppendorf tubes and incubated for 120 hours (five days) at 4° C in the presence of 100 μM of the probe UCM120 (which is a 5-HT_1A agonist in mammals labeled with a dansyl group as fluorescent tag) [38] in constant agitation in a final volume of 300 μl. Specimens were then washed three times with PBS and fixed for 3 hours at room temperature (RT) in 4% (w/v) paraformaldehyde (PFA) in PBS/TX-100/NaN₃/BSA/Dig/DMSO. The protoscoleces were washed again, mounted in glycerol/PBS (8:1 v/v) and then viewed under standard fluorescence microscope (Nikon Eclipse E600) or in a confocal scanning laser microscope (CSLM, Fluoview 1000 Olympus). The individual pictures obtained from the optical sections were used for the reconstruction of the whole specimen using the program ImageJ. In order to determine if the signal obtained was specific, several controls were run in parallel: protoscoleces without probe or protoscoleces in the presence of 1000 μM of serotonin as negative control. This experiment was performed several times with similar results.

Levels of GPCR expression

The transcriptional expression levels (in RPKM, or reads per kilobase per million reads) for each serotoninergic receptor in Echinococcus granulosus sensu stricto (G1 genotype) were from Zheng et al. ([32], Supplementary Table 28).

Cell culture and cAMP assays

HEK-293 cells (ATCC CRL-1573.3) were cultured in growth media [DMEM (Gibco), 10% heat inactivated fetal bovine serum (Gibco), penicillin (100units/mL), streptomycin (100μg/mL) and L-glutamine (290μg/mL)] and used for assays between passages 5 and 25. For cestode GPCR heterologous expression assays, cells were transfected (Lipofectamine 2000, Invitrogen) at 80% confluency approximately 16 hours after seeding within T-25 culture flasks with a 1:1 ratio of human codon optimized cestode GPCR cDNA (subcloned into a pcDNA3.1(-) mammalian expression vector) and cDNA encoding the pGloSensor 22-F plasmid (Promega). The following day, cells were trypsinized, centrifuged (300g/5min), resuspended in DMEM supplemented with 1% dialyzed FBS (Gibco) and plated in 96 well, solid white plates (Corning, cat # 3917). After overnight culture to allow adherence, media was exchanged for assay buffer (HBSS supplemented with 0.1% BSA, 20mM HEPES (pH 7.4), and GloSensor reagent (Promega). cAMP-luminescence assays were performed following addition of a phosphodiesterase inhibitor (3-isobutyl-1-methylxanthine, IBMX; 200μM) using a GloMax-Multi Detection System plate reader (Promega) as described previously [24].

Statistical analyses

Statistical analyses were carried out using the GraphPad Prism Software package, version 6 for Windows (GraphPad Software, CA, USA). To analyze the effects of serotonin in motility assays performed with the WMicrotracker device, the relative motility indices of all replicate wells for each concentration of the different independent experiments were made into one cohort for one-way ANOVA. To determine significant differences a Bonferroni and Dunnett post-tests were performed comparing all concentrations with the control. In the experiment evaluating the effects of UCM2550 on protoscolex motility, the mean motility index obtained with the WMicrotracker was compared before and after the addition of the drug with a Student’s t-test. For image based assays, pixel changes in control and treatments were calculated and analyzed with one way ANOVA and Dunnett post-test were performed comparing all the concentrations with the control. Differences from control with P < 0.05 were considered statistically significant inductions or inhibitions of motor activity.

Cloning, sequencing and bioinformatic analysis of serotoninergic GPCRs in two cestode species

Six gene models in total were retrieved by blast searches from which four gene models were from Echinococcus spp. databases and only two from the M. corti database. From all them, only ECANG7_06088, ECANG7_06092 (E. granulosus) and MCOS_0000684401 (M. corti), were used for primer design since the rest of the gene models were either incomplete (lacking one or more transmembrane segments), or missing important residues for function, or they could not be amplified in PCR reactions. By using sequence specific primers, three cDNAs were cloned by standard RT-PCR being two of the sequences obtained from E. granulosus (5-HT_7Egran1 and 5-HT_7Egran2) and one from M. corti (5-HT_7Mco1). The 5-HT_7Egran1 cDNA has 2100 bp and encodes a protein of 659 amino acids with a predicted molecular weight of 73.8 kDa and a pI close to 10 (pI = 9.83). The second cDNA cloned, 5-HT_7Egran2, has 1731 bp and encodes for a protein of 576 amino acids with a predicted molecular weight of 65 kDa and a pI close to 9 (pI = 9.12). Finally, the M. corti cDNA 5-HT_7Mco1, has 2091 bp and encodes for a protein of 696 amino acids with a predicted molecular weight of 78.6 kDa and a pI of 9.45. Domain searches showed that all the cloned sequences have best hits with the 5-HT₇ domain of serotonin receptors. S3 Fig shows a multiple sequence alignment between the Homo sapiens, C. elegans, S. mansoni, E. granulosus and M. corti sequences encoding for serotoninergic G-protein coupled receptors. The bioinformatic analysis of the cloned sequences shows several characteristics of the molecular signature of a GPCR. For example, all three sequences have seven conserved hydrophobic transmembrane segments consistent with GPCR architecture (S3 Fig). Topological modeling (TMpred) predicts an extracellular N-terminus and an intracellular C-terminus. Intracellular loop 3 and, for 5HT_7Egran1 and 5HT_7Mco1 the carboxy-terminal regions, were longer than the corresponding regions in the human counterpart (S3 Fig). The multiple sequence alignment also highlights conserved amino acid residues that are invariant within the rhodopsin family of GPCRs [58]. The first of them is the DRY motif present in the third transmembrane domain probably involved in the stabilization of the ground state [59, 60]. The second relevant motif found was the NPxxY motif located toward the cytoplasmic end of the seventh transmembrane domain and potentially implicated as crucial in the recruitment of β-arrestin, which is a mechanism of receptor inactivation [59–61]. The third is the PIF motif, which is a critical trigger motif for receptor activation [59, 60]. And the last one, in the transmembrane segment 6, there is a cluster of aromatic residues referred as “toggle switch” in which it was reported that the movement of some of these residues could be important for receptor activation [60]. Taking the residue position from the human 5HT_1B receptor as a reference sequence [62] and the Ballesteros and Weinstein nomenclature [63], the residues conserved in all the cestode sequences were: C122^3.25 which makes putative disulphide bond contact with the C199^El2 in the extracellular loop 2; D129^3.32, which forms a salt bridge with the positively charged amino group of ergotamine in the human receptor [62]; C133^3.36, which recognizes the amine portion of the ligand; T134^3.37, which forms hydrogen bond with the indole group; A213^5.46, which interacts with the indole ring in the ligand; W327^6.48, F330^6.51 and F331^6.52, which form the orthosteric binding pocket. The residue Y359^7.43 is hypothesized to form hydrogen bond with N6 of ergotamine in the human receptor. The residue conserved in all but not in 5-HT_7Egran2, is T216^5.43, which interacts with the indole ring in the ligand and is replaced by C in this position in the 5-HT_7Egran2 sequence. The conserved S215^5.42 of TM5 is replaced by alanine in all the cestode sequences and, interestingly, the same substitution was observed in the S7.1 receptor from the planaria D. japonica [15] and also in the Sm5HTR from S. mansoni [18]. TM5 residues (S4 Fig) are important players in ligand recognition [62, 64]: with the exception of the alanine at position 5.42, the other crucial residues (5.43 and 5.46) are conserved in mammalian serotonin receptors, particularly in subtypes 1 (5-HT1) and 7 (5-HT7). Besides ligand binding, other residues potentially involved in G protein contacts could be identified (S3 Fig). For example, the residue L316^6.37 in the cytoplasmic end of TM6 upon activation contacts a universally conserved leucine of the G protein in the β2 adrenergic receptor-Gs structure [65]. Moreover, in the V155 position of the human 5-HT1 receptor (F139 in the human beta 2 adrenergic receptor), located in the IL2 helix, it was reported a F (phenylalanine) or L (leucine) for Gs coupling which is coincident with the residues observed in serotonin receptors from cestodes at this position [66] (S3 Fig). The phylogenetic analysis (Fig 1) shows that all the cloned sequences grouped with the 5HT₇ subtype (S7-like clade) of invertebrate serotoninergic G-protein coupled receptors, which are Gs coupled [67].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Cladogram of cestode and invertebrate serotoninergic GPCRs.

The amino acid sequences of predicted cestode serotonin receptors were aligned with the repertoire of serotonin receptors cloned from various other flatworm and non-flatworm invertebrate model organisms; Dugesia japonica, 5-HT_Dj; Schistosoma mansoni, 5-HT_Sm; Caenorhabditis elegans, 5-HT_Ce; Drosophila melanogaster, 5-HT_Dm. The cestode GPCR sequences cloned and functionally expressed in this study (5-HT_7Egran1, 5-HT_7Egran2, 5-HT_7Mco1) cluster within a clade of 5-HT7 like receptors (green), as does an additional predicted M. corti sequence (5-HT_7Mco2). Other predicted cestode serotoninergic sequences (5-HT_1Egran1 and 5-HT_1Egran2) cluster within a clade of 5-HT1 like receptors (blue). See methods for complete list of UniProt and GenBank sequence identifiers used in this analysis. Analysis was bootstrapped with 500 replicates.

https://doi.org/10.1371/journal.pntd.0006267.g001

The transcriptomic data reveals a highly expressed receptor at the protoscolex stage

Transcriptomic data obtained from Zheng and coworkers [32], reveals that two of the receptors cloned here (5-HT_7Egran1 and 5-HT_7Egran2) are highly expressed in the activated protoscolex stage (S5 Fig) with the levels of expression of 5-HT_7Egran2 being particularly high. The levels of expression of both receptors in activated protoscolex stage are at least fourfold higher than those observed in other stages of the parasite. No expression of these two receptors was detected in the oncosphere.

Molecular modelling of 5-HT_7Egran1, 5-HT_7Egran2 and 5-HT_7Mco1

Molecular models of 5-HT_7Egran1, 5-HT_7Egran2 and 5-HT_7Mco1 were developed to explore potential structural similarities and differences between the parasitic and human receptors. Fig 2 shows the models of each cloned serotoninergic GPCR compared with the crystal structure of human 5-HT_1BR (4IAR) [62]. A striking structural similarity in the transmembrane domains can be observed (Fig 2, models A, C and E). According to the models obtained, four important interactions with ergotamine could occur at the transmembrane segments three (III) and five (V), (Fig 2, models B, D and F). The indicated residues comprise part of the orthosteric binding pocket of the cestode GPCRs.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Molecular modeling of serotoninergic receptors from cestodes.

The structural similarities between three cestode receptors and the published structure of the serotoninergic receptor from human (4IAR) are shown. A) Model of the serotonin receptor 5-HT_7Egran1 (Brown) superposed to the published structure of the human serotonin receptor 5-HT1B (blue). B) Close view of the putative ergotamine interaction with residues of the transmembrane domains III and V from 5-HT_7Egran1 (lateral chains of the amino acids involved are indicated in red), molecular distances between atoms are indicated. C) Model of 5-HT_7Egran2 (violet) superposed to 5-HT1B (blue). D) Close view of the ergotamine interaction with some residues of the transmembrane domains III and V from 5-HT7_Egran2. E) Model of 5-HT_7Mco1 (Brown) superposed to 5-HT1B (blue). F) Close view of the ergotamine interaction with some residues of the transmembrane domains III and V from 5-HT_7Mco1. In all the representations, the molecule of ergotamine was marked in green and the residues in transmembrane domains potentially involved in ergotamine interaction were marked in red.

https://doi.org/10.1371/journal.pntd.0006267.g002

Serotonin increases the motility of Mesocestoides corti tetrathyridium

Previously, we have shown that 5-HT induces E. granulosus protoscolex motility [16]. In spite of some evidence about 5-HT action on M. corti motility [3], the kinetics of this important neurotransmitter in motility over longer periods of time (>30 minutes) was unknown. For this reason, we analyzed further the effect of 5-HT on the larvae of M. corti. The addition of serotonin to tetrathyridia resulted in the stimulation of the motility above basal levels (S6 Fig, S1 Video). The channel activity traces show a stimulation in frequency of movement at high doses of 5-HT compared to the basal condition (S6A Fig). The dose-response curve showed a significant rise of motility starting from 100 μM of 5-HT (*p< 0.05) until the highest concentration tested of 2000 μM over a period of two hours (****p< 0.001; S6B Fig). Dose-response curves were also collected over different intervals of time (30, 60, 90 and 120 minutes). These graphs showed that the shape of the curve changes with time (S6C Fig). At the highest concentrations tested (500 to 2000 μM), the relative motility tends to drop with time after the first 30 minutes of incubation, losing their significance respect to control, probably reflecting an inactivation process. At the lowest dose (100 μM) the motility tends to rise during the first 90 minutes of incubation and then also it tends to fall after 120 minutes of incubation (S6C Fig). Similar concentrations of 5-HT (1000 μM; ***p< 0.001) are necessary to see significant changes in motility over a period of 30 minutes by video imaging (S6D Fig).

Functional expression and pharmacological profiling show activation by 5-HT with distinct pharmacological properties

Based on the bioinformatic prediction that these sequences encode serotoninergic GPCRs, responses to 5-HT were compared between cells expressing individual cestode GPCRs and untransfected cells. For each of the clones, addition of 5-HT in transfected cells evoked a rapid and dose dependent stimulation of cAMP accumulation (Fig 3A, 3B & 3C). One notable feature of two of the GPCRs– 5-HT_7Egran2 and 5-HT_7Mco1 –was resolution of a background rate of cAMP accumulation immediately on IBMX addition prior to addition of 5-HT (the period between the arrows in Fig 3B & 3C). This was not observed with the 5-HT_7Egran1 clone (Fig 3A). Because of this basal coupling, the fold-change in luminescence signal on addition of 5-HT was lower with– 5-HT_7Egran2 (~3.3-fold) and 5-HT_7Mco1 (~2.5-fold) than observed in cells expressing 5-HT_7Egran1 (~20-fold). Full dose response relationships were performed for each GPCR, and compared with endogenous responses to the same concentrations of 5-HT in untransfected cells (Fig 3D, 3E & 3F). The EC₅₀ for 5-HT evoked cAMP generation was 171±57nM in cells expressing 5-HT_7Egran1. For the two GPCRS exhibiting basal coupling, the EC₅₀ for 5-HT evoked cAMP production was in the picomolar range measured as 543±107pM for 5-HT_7Egran2 and 255±63pM for 5-HT_7Mco1. Responses to 5-HT were considerably smaller in untransfected cells (Fig 3D, 3E & 3F). To investigate the specificity of these GPCRs for 5-HT, responses to higher concentrations (10μM) of other neurotransmitters (tryptamine, tyramine, octopamine, histamine, dopamine and acetylcholine) were examined (Fig 4). In all cases, 5-HT elicited maximal cAMP accumulation. At higher doses (10 μM), tryptamine also proved an activator of 5-HT_7Egran2 (EC₅₀ ~0.8μM) and 5-HT_7Mco1 (EC₅₀ ~1μM) but not 5-HT_7Egran1 (Fig 4). The first 2 receptors were at least 1000-fold less sensitive to tryptamine than to 5-HT. Responses to other ligands were considerably smaller, likely representing responses from endogenous GPCRs in HEK-293 cells (histamine, dopamine) or comparable with levels of luminescence seen with vehicle control. From this dataset we conclude that each of the cestode sequences encodes a bona fide serotoninergic GPCR.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Heterologous expression of GPCRs reveals 5-HT evoked cAMP accumulation.

Time resolved measurements of cAMP accumulation in cells expressing individual cestode serotoninergic GPCRs (A) 5-HT_7Egran1, (B) 5-HT_7Egran2 and (C) 5-HT_7Mco1 before and after addition of IBMX (1^st arrow, 200μM) and different doses of 5-HT (2^nd arrow, doses indicated in legend). (D, E & F) dose response relationships to 5-HT measuring peak amplitude of 5-HT evoked luminescence change in cells expressing the indicated GPCR (solid circles), or untransfected HEK-293 cells (open circles).

https://doi.org/10.1371/journal.pntd.0006267.g003

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Profiling neurotransmitter specificity against cestode serotoninergic GPCRs.

Responses (raw luminescence units, RLU) in cells transfected with (A) 5-HT_7Egran1, (B) 5-HT_7Egran2 and (C) 5-HT_7Mco1 to indicated neurotransmitters at the indicated time points in the arrow (10μM), with the exception of 5-HT (700nM in (A), 3nM in (B) and 2nM in (C)).

https://doi.org/10.1371/journal.pntd.0006267.g004

Our modelling studies supported an interaction of the ergot alkaloid ergotamine with the binding pocket of the cestode GPCRs (Fig 2B, 2D & 2F). Therefore we assayed responsiveness of each GPCR to ergotamine, as well as the synthetic hallucinogen lysergic acid diethylamide (LSD) which has proved a useful probe ligand for studying 5-HT receptor properties [59]. Ergotamine activated each of the cestode 5HT₇Rs but with variable efficacy (Fig 5A, 5B & 5C), behaving as a low efficacy partial agonist at 5-HT_7Egran1 and a higher efficacy partial agonist at 5-HT_7Egran2 and 5-HT_7Mco1. Each GPCR responded to ergotamine over a similar concentration range (EC₅₀s of 290±49nM, 234±37nM and 354±62nM for 5-HT_7Egran1, 5-HT_7Egran2 and 5-HT_7Mco1 respectively, Fig 5A, 5B & 5C). Responses to LSD were then examined. Two of the receptors (5-HT_7Egran1 and 5-HT_7Mco1) were potently activated by addition of LSD but no response was observed in cells expressing 5-HT_7Egran2 when probed in ‘agonist’ mode (Fig 5D, 5E & 5F). Rather, LSD attenuated the basal coupling of 5-HT_7Egran2 and blocked cAMP accumulation in response to a subsequent addition of 5-HT (‘antagonist mode’), demonstrating LSD acted as an antagonist of 5-HT_7Egran2. Full dose response relationships to LSD were performed (Fig 5G, 5H & 5I) which demonstrated EC₅₀s for activation of 35±6nM (5-HT_7Egran1) and 4.4±1.0nM (5-HT_7Mco1), and an IC₅₀ for inhibition of 5-HT responses of 2.9±0.2nM (5-HT_7Egran2). Collectively, these functional expression data demonstrate that while each of these GPCRs is activated by 5-HT, they exhibit distinct pharmacological properties (sensitivity to 5-HT, differential responsiveness to tryptamine, ergotamine and LSD).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Ergot alkaloid activity at cestode 5-HT GPCRs.

(A-C) Dose response relationships describing sensitivity and magnitude of peak cAMP accumulation versus maximal 5-HT response to various doses of ergotamine at (A) 5-HT_7Egran1, (B) 5-HT_7Egran2 and (C) 5-HT_7Mco1 (D-F). Real time kinetic profiling of cAMP accumulation in response to an addition of either LSD and IBMX (10μM and 200μM, respectively, closed circles) or IBMX alone (200μM, open circles) at first arrow and then subsequent addition of 5-HT (5 μM, second arrow) for (D) 5-HT_7Egran1 or 5-HT (1 μM, second arrow) for (E) 5-HT_7Egran2 and (F) 5-HT_7Mco1. Dose response relationships for activation or blockade of 5-HT evoked cAMP accumulation by LSD at (G) 5-HT_7Egran1, (H) 5-HT_7Egran2 and (I) 5-HT_7Mco1.

https://doi.org/10.1371/journal.pntd.0006267.g005

Serotoninergic GPCR expression is localized in particular cellular territories of E. granulosus protoscoleces

In order to assess the localization of serotoninergic G-protein coupled receptors in larval stages of cestodes, we used a fluorescent probe with high affinity for mammalian serotoninergic GPCRs. Since the first attempts with M. corti tetrathyridia were unsuccessful and owing to the very limited availability of the probe, we concentrated our efforts in protoscoleces of E. granulosus, taking into account the higher size of M. corti tetratyridia compared to E. granulosus protoscoleces. The fluorescent probe—UCM120, a high affinity agonist of the human 5-HT_1A receptor (K_i = 2 nM, Fig 6A) [38]—displayed a distinctive staining pattern in protoscoleces of E. granulosus. A discontinuous and punctiform pattern of staining could be observed particularly concentrated in suckers but some of this pattern could also be observed in the body (Fig 6B and 6C). Several controls were performed in order to determine the specificity of pattern observed. When protoscoleces where incubated in the presence of the probe and an excess of serotonin (1 mM) was added, the staining pattern previously mentioned disappeared (Fig 6D). The hooks remained strongly stained as seen in controls (Fig 6D), suggesting this represented non-specific staining or autofluorescence. When protoscoleces where incubated in the presence of the non fluorescent analog UCM2550, motility was inhibited (Fig 6E, S2 Video). The dose-response curve obtained with the WMicrotracker device showed a significant fall of motility (relative to basal) starting from 100 μM (*p< 0.05) until the highest concentration tested of 1000 μM during a period of two hours (*p< 0.05; Fig 6E). However, using a video based method, lower concentrations produced significant inhibition of motility, starting from 0.1 μM (**** P ≤ 0.0001; Fig 6F).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. Protoscoleces of Echinococcus granulosus stained with the fluorescent probe UCM120 under the confocal scanning laser microscope and effect of the parent compound UCM2550 on the motility.

- Protoscoleces were fixed, after washing they were incubated in the presence of the probe (100 μM) during several days, washed, fixed, washed again, mounted, and then observed by confocal microscopy. (A) Structure of the fluorescent probe UCM120 with the structure of the parent compound UCM2550 shown in black and the dansyl group shown in light brown. Images of two protoscoleces obtained from superficial (B) or from an entire stack of laser scanning (C). To assess specificity, protoscoleces were labeled under the same conditions with the probe (100 μM) in the presence of an excess (1000 μM) of 5-HT and the image projection was obtained from an entire stack of pictures (D). Relative motility observed in the presence of increasing concentrations of the parent compound UCM2550 in the WMicrotracker device measured after two hours of incubation (E); Image based motility quantification measured as a pixel change (F). Further details are provided in the Methods section. Asterisks indicate treatments found to be significantly different from the controls (*P ≤ 0.05) with t-test (E) or ANOVA and Dunnet post comparison tests (****P ≤ 0.0001, F). R: rostellum; S: suckers and Bo: body. The scale bar represents 50 μm in (A), (B) and (C) and 100 μm in (D). Arrowheads indicate localization of the probe.

https://doi.org/10.1371/journal.pntd.0006267.g006