The FMRF-NH2 gated sodium channel of Biomphalaria glabrata: Localization and expression following infection by Schistosoma mansoni

The neglected tropical disease schistosomiasis impacts over 700 million people globally. Schistosoma mansoni, the trematode parasite that causes the most common type of schistosomiasis, requires planorbid pond snails of the genus Biomphalaria to support its larval development and transformation to the cercarial form that can infect humans. A greater understanding of neural signaling systems that are specific to the Biomphalaria intermediate host could lead to novel strategies for parasite or snail control. This study examined a Biomphalaria glabrata neural channel that is gated by the neuropeptide FMRF-NH2. The Biomphalaria glabrata FMRF-NH2 gated sodium channel (Bgl-FaNaC) amino acid sequence was highly conserved with FaNaCs found in related gastropods, especially the planorbid Planorbella trivolvis (91% sequence identity). In common with the P. trivolvis FaNaC, the B. glabrata channel exhibited a low affinity (EC50: 3 x 10−4 M) and high specificity for the FMRF-NH2 agonist. Its expression in the central nervous system, detected with immunohistochemistry and in situ hybridization, was widespread, with the protein localized mainly to neuronal fibers and the mRNA confined to cell bodies. Colocalization of the Bgl-FaNaC message with its FMRF-NH2 agonist precursor occurred in some neurons associated with male mating behavior. At the mRNA level, Bgl-FaNaC expression was decreased at 20 and 35 days post infection (dpi) by S. mansoni. Increased expression of the transcript encoding the FMRF-NH2 agonist at 35 dpi was proposed to reflect a compensatory response to decreased receptor levels. Altered FMRF-NH2 signaling could be vital for parasite proliferation in its intermediate host and may therefore present innovative opportunities for snail control.


Introduction
Pond snails of the genus Biomphalaria (Mollusca: Gastropoda: Heterobranchia; Planorbidae) serve as intermediate hosts for Schistosoma mansoni, the causative agent for the most widespread form of intestinal schistosomiasis [1,2]. Within their snail hosts, larval trematodes multiply and transform into the cercarial form that can infect humans [3,4]. Strategies for elimination of schistosomiasis include improved sanitation, large-scale preventive chemotherapy, and snail control [5,6,7].
Neuropeptide signaling systems are promising molecular targets for pesticide and parasiticide drug development [8,9,10]. In contrast to the classical neurotransmitter systems that are presently common targets for pest control, some neuropeptides and their receptors are limited to specific invertebrate clades, reducing concerns of widespread toxicity [10,11,12]. The FMRF-NH 2 family of neuropeptides holds potential for drug development due to its pervasive role in the behavior and neuromuscular physiology of major arthropod and helminth parasites and pests [13,14,15]. FMRF-NH 2 was initially purified from a bivalve mollusc (sunray venus clam, Macrocallista nimbosa; [16]) and has been intensively studied in several gastropod species [17,18,19]. To date, however, the potential utility of this peptide signaling system for snail control interventions has not been explored.
Ionotropic DEG/ENaC channels are expressed in numerous cell types and tissues and are gated by varied stimuli, including mechanical forces and protons [29,30,31]. In the nervous system, members of this channel family participate in a range of functions, including mechanotransduction, nociception, and synaptic plasticity [32,33,34]. Involvement of the FMRF-NH 2 activated sodium channel (FaNaC) in the neural circuits that regulate gastropod physiology and behavior has not been established.
Recently, we examined the precursor organization of the B. glabrata FMRF-NH 2 -related neuropeptides and localized their expression in the CNS [35]. As our understanding of this neuropeptide system will be broadened by defining its complementary receptors, the present study characterized the B. glabrata FMRF-NH 2 -activated sodium channel, localized its expression in the CNS, and explored whether its expression may be modified following exposure to Schistosoma mansoni.

Ethics statement
All protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Puerto Rico Medical Sciences Campus (Snails: Protocol #3220110; Xenopus: Protocol #9470110). All animal care and experimental procedures followed guidelines and regulations specified in the Guide for the Care and Use of Laboratory Animals (National Research Council) [36].

Specimens
Histological protocols were performed on B. glabrata snails bred in the animal facility at the Institute of Neurobiology, University of Puerto Rico Medical Sciences Campus. Snails were maintained in aquaria at room temperature under a 12:12 light-dark cycle and fed lettuce ad libitum.
B. glabrata were exposed to S. mansoni miracidia at the Biomedical Research Institute (BRI, Rockville MD). Schistosome eggs were obtained from mouse livers and hatched using BRI protocols [37]. Snails were incubated with miracidia (target: 5 per snail) for two hours. Tissues from infected specimens were dissected and collected at the BRI at 20 days (prepatent) and 35 days (shedding) post-infection (dpi). Shedding was verified by exposure of snails to direct light.

Electrophysiology
The cDNA encoding the Bgl-FaNaC sequence was codon-optimized for Xenopus laevis and synthesized by Integrated DNA Technologies (IDT, Coralville IA). The Flag-tag protein affinity sequence (DYKDDDDK; Sigma-Aldrich) was added to the N-terminal to enable confirmation of expression. The cDNA was subcloned into the Xenopus expression vector pGEM HE as described previously [38]. The Bgl-FaNaC full-length RNA was transcribed in vitro, capped and polyadenylated using the T7 mScript Standard mRNA Production System (CellScript, Madison WI).
Oocytes were obtained from dissected ovaries of adult Xenopus laevis specimens (Xenopus Express, Brooksville FL). They were dispersed with type II collagenase and manually defolliculated. RNA injections were performed in pre-selected oocytes from stages V and VI. Oocytes were injected with 38.6 nL of Bgl-FaNaC encoding RNA at a concentration of 20 ng/uL using a Nanoliter 2000 microinjector (World Precision Instruments, Sarasota FL).
Protein expression was monitored daily for six consecutive days. Eight oocytes were homogenized (83 mM NaCl, 1 mM MgCl 2 , 10 mM HEPES, 5 mM EDTA, pH 7.9) using a glass micro-homogenizer. Homogenates were centrifuged at 14,000 G (4˚C) and supernatants were stored at -80˚C until needed. Protein lysate quantification was performed with the Precision Red Advance Protein Assay (Cytoskeleton Inc., Denver CO) following the manufacturer's instructions. An equal amount (5 μL) of protein lysate was loaded on a 12% polyacrylamide gel. Following electrophoresis and transfer to a PVDF membrane, immunodetection was performed with an anti-Flag primary antibody diluted 1:200 in blocking solution (4˚C, overnight rocking). Membranes were washed in TBS-T (3 x 10 min) and incubated in goat-anti rabbit peroxidase conjugated second anti body (1:4000; 1 h). Membranes were then washed with Tris-buffered saline, 0.01% Tween 20 (TBS-T; 3 x 10 min) and processed for chemiluminescent signal detection using the Super Signal West-Femto Maximum Sensitivity Substrate kit (Thermo Fisher Scientific, Waltham MA) following the manufacturer's instructions. Maximum expression levels were observed 6 days following injection.
Injected oocytes were placed at the bottom of a plastic recording chamber (3 ml volume) lined with a nylon mesh and continuously perfused with ND96 (96 mM NaCl, 2 mM KCl, 1.8 mM CaCl 2 , 1 mM MgCl 2 , 5 mM HEPES, pH 7.6; see detailed methods in [38]). An OC725B oocyte clamp (Warner Instruments LLC., Hamden CT) was used to clamp the membrane potential at -60 mV with independent microelectrodes for recording (0.1 M potassium chloride) and passing current (3 M potassium chloride). Data were acquired on a Digidata 1200 interface with Clampex (V.6) and analyzed with AxoScope 10.7 (Axon Instruments, Union City CA). All experiments were carried out at room temperature (20-25˚C).

Antibody validation
Affinity purified rabbit polyclonal antibodies were generated against the amino terminus intracellular domain of the B. glabrata FMRF-NH 2 -gated sodium channel (KYTSPDAKPSMSTS-C; residues 2-15, Figs 1A and 2) by GL Biochem Ltd., Shanghai, China (ELISA titer > 1:128,000). Solid phase specificity was confirmed with dot blots of serial antigen dilutions (2 μL) applied to nitrocellulose membrane (Bio-Rad 0.45 μm; Fig 1B). Membranes were allowed to air dry (1 h) and then incubated with blocking buffer (1 h, room temperature, shaking). Membranes were incubated overnight with the anti-FaNaC primary antibody diluted to 5 μg/ml in blocking solution (4˚C, shaking). For preabsorption controls, the antibody was pre-incubated with the peptide antigen (10 −4 M) overnight prior to application to the membrane. The membranes were then washed three times for 15 minutes and incubated with goat anti-rabbit IgG (H+L) second antibody conjugated to HRP (0.25 μg/ml in blocking solution, 1 h room temperature). They were then washed in TBS-T (4 x 15 min) and transferred to SuperSignal West Femto Maximum Sensitivity Substrate (Thermo Fisher Scientific, Waltham MA). The enzyme-substrate reaction was allowed to proceed for five minutes before visualization.
Preabsorption experiments on fixed Biomphalaria nervous tissue also verified the specificity of antigen detection. A 1:200 dilution of the Bgl-FaNaC antibody produced strong immunofluorescence ( Fig 1C) that was eliminated when the antibody was preabsorbed with the antigen prior to tissue incubation (5 x 10 −4 M, overnight; Fig 1D). Together, the preabsorption experiments supported the sensitivity and specificity of the antibody used in this study. They also provided guidance for primary and secondary antibody dilutions to use for antigen detection. Signals were eliminated when primary antibody incubation was omitted from the protocol (not shown).

Wholemount immunohistochemistry
Standard wholemount immunohistochemical protocols were followed [39,40]. Tissues were dissected in normal saline (51.3 mM NaCl, 1.7 mM KCl, 1.5 mM MgCl 2 , 4.1 mM CaCl 2 , 5 mM HEPES, pH 7.8.), and pinned in a Petri dish lined with Sylgard (World Precision Instruments, Cat. No. SYLG184). Ganglia were incubated in protease (Type XIV, Sigma #P5147; 0.5% in conformation are labeled (see [31]). The antibody used in this study was generated against a 14-residue domain near the amino terminus of the channel (asterisk). The selected sequence was based upon an analysis of antigenicity performed by GL Biochem (Shanghai, China). b: Dot blot controls demonstrate specificity of the Bgl-FaNaC antibody. Upper row: serial dilutions of a 2 μM antigen solution were blotted (2 μl) and probed with a 1:200 dilution of the antibody used in this study. Lower row: Preabsorption of the antibody with the antigen peptide (1 x 10 −4 M, overnight) eliminated recognition of the blotted peptide. c: Wholemount immunolabeling of the ventral surface of the visceral and left parietal ganglia. Intense FaNaC labeling (green) was present in fiber systems coursing through the ganglia. d: FaNaC labeling was eliminated following antibody preabsorption with the antigen. e: The signal produced with the digoxygenin in situ hybridization protocol was confined to the cell bodies of neurons in the visceral and left parietal normal saline; 7-10 min), washed thoroughly with normal saline, and fixed in 4% paraformaldehyde (1 h, room temperature).
Fixed tissues were washed 5 x 20 min in PTA (0.1 M phosphate buffer containing 2% Triton X-100 and 0.1% sodium azide) at room temperature. Samples were pre-incubated with normal goat serum (NGS; 0.8%, 3-12 h, room temperature) and then transferred to the primary antibody (5 μg/ml in PTA, 3-5 days). Samples were washed (5 x 20 min in PTA) and incubated in second antibodies conjugated to a fluorescent marker (Alexa 488 goat anti-rabbit IgG (H+L) conjugate; Molecular Probes, Eugene OR) at dilutions ranging from 1:500 to 1:1,000. No differences were evident across the ranges of NGS blocking times, antibody incubation durations, or second antibody concentrations. Results were assessed on a Nikon Eclipse epi-fluorescent microscope prior to confocal imaging on a Nikon A1R Confocal Laser Microscope using the NIS Elements AR software package. Image processing and analysis were performed with Fiji (NIH, GitHub open source) and figures were prepared with Microsoft PowerPoint (v.16.69.1).
Hybridization Chain Reaction (HCR) fluorescence in situ hybridization (FISH). HCR RNA-FISH methods were adapted from the Molecular Instruments, Inc. (Los Angeles CA) protocols website (https://www.molecularinstruments.com/). The B. glabrata CNS was dissected in normal saline and pinned on Sylgard-lined plates. Tissues were exposed to protease (0.5%, Type XIV, Sigma) diluted in normal saline (7-10 min) and fixed in 4% paraformaldehyde overnight. The CNS was washed 5 times for 15 minutes each with PTwA (0.1 M phosphate buffer containing 2% Tween 20 and 0.1% sodium azide). Samples were pre-hybridized in hybridization buffer (Molecular Instruments, Inc.) for 30 minutes, and then hybridized ganglia. f: Detection of FaNaC mRNA using the Hybridization Chain Reaction technique produced defined labeling (cyan) in the cell bodies of neurons in the visceral and left parietal ganglia. All calibration bars = 50 μm.
https://doi.org/10.1371/journal.pntd.0011249.g001  [42]), and Aplysia kurodai (GenBank ID: AB206707.1 [43]). Amino acid numbering corresponds to Bgl-FaNaC. Sequence of the peptide used to generate the antibody used in this study is shaded light green. Orange shading highlights conserved cysteine residues that are thought to contribute to disulfide bridges in the finger region of the channel. Light blue shading denotes the two highly conserved transmembrane domains, TM1 and TM2. The less conserved region that is proposed to account for peptide recognition is shaded light red [44] [45]. Within this domain, four specific residues that influence the concentration-response relationship are outlined (see [46]). The alignment was generated with the T-Coffee web server [47]. https://doi.org/10.1371/journal.pntd.0011249.g002

PLOS NEGLECTED TROPICAL DISEASES
overnight at 37˚C with a probe set generated for the Bgl-FaNaC transcript (FaNaC/ LOT PRI987). Multiplexed detection of transcripts was achieved with a cocktail of probe sets that also included the Bgl-FaRP1 precursor (LOT PRI087; 361 nucleotides, coding sequence, excluding the sequence shared with Bgl-FaRP2; see [35]) and the Bgl-FaRP2 precursor (LOT PRI298; 487 nucleotides, coding sequence, excluding the sequence shared with Bgl-FaRP1). Probes were diluted in hybridization buffer at a final concentration of 4 nmol/μL each. Samples were washed 4 x 15 minutes with wash buffer (Molecular Instruments, Inc.) and incubated at room temperature with amplification buffer (Molecular Instruments, Inc.) for 30 minutes. Aliquots of the hair pin amplifiers (h1 and h2; 5 μL each, from 100 μM stock) were heated (95˚C for 90 s) and then placed in a dark box for 30 minutes at room temperature. Following cooling, 5 μL of each hair pin was added to 250 μL of amplification solution. Tissues were transferred to the amplification solution and incubated overnight at room temperature in a dark box. The following day, samples were washed 5 times (10 minutes each) with 5x SSCT (sodium chloride-sodium citrate buffer, 0.1% Tween) at room temperature. Image acquisition, image analysis, and figure preparation were performed as described above (Wholemount immunohistochemisty).

Quantification of expression
As the fluorescence mRNA detection provided superior clarity and definition (Fig 1E and 1F), all expression measurements were obtained from samples using the HCR method. Images from control and infected samples were obtained using identical settings on the NIS Elements data acquisition program. Labeled neurons on the dorsal and ventral surfaces of each ganglion were counted using the Fiji image processing package (ImageJ.org) by an experimenter blinded to the treatment. The mean gray value cut-off for positive expression was set at 15. Neurons with a diameter less than 10 μm were excluded from the analysis. Overall mean gray values were obtained from a region of interest (ROI) demarcated with the Fiji "free hand selection" tool. For analysis of Bgl-FaRP1 and Bgl-FaRP2, specific clusters (B group, F group and E group; see [35]) were selected as the ROI for gray value measurements. As Bgl-FaNaC expression was more dispersed, the perimeter of each ganglion was traced to define the ROI.

Statistical analysis
Data are presented as mean ± standard error of the mean (SEM). Individual measurements are plotted for each condition in the bar graphs. Statistical significance was determined with the Brown-Forsythe and Welch one-way analysis of variance (ANOVA) and Dunnett's multiple comparison post hoc test. Tests were performed and graphs were generated with GraphPad Prism version 9.2.0.

The B. glabrata FaNaC structure and function
A transcriptome generated from twelve pooled B. glabrata central nervous systems [35] yielded a transcript encoding the B. glabrata FMRFamide-activated amiloride-sensitive sodium channel-like protein previously derived from genomic sequence [28]. This 4444 nucleotide transcript (GenBank Accession number OP066530) included a 1395 nucleotide 5' untranslated sequence, an open reading frame (ORF) encoding a 621 amino acid protein termed Bgl-FaNaC, and a 1186 nucleotide 3' untranslated sequence. The Bgl-FaNaC amino acid sequence was identical to that deduced from the genome (GenBank Accession Number XP_013063507; [28]).

Bgl-FaNaC localization
A polyclonal rabbit antibody was generated against residues 2-15 of the B. glabrata FaNaC (Figs 1A and 2). Immunohistochemical processing of wholemount central nervous systems labeled a widespread network, with cell bodies located in all ganglia, and abundant fiber systems within the peripheral nerves (Figs 5 and 6). Dense fiber tracts also coursed through the central neuropil of the ganglia, often obscuring underlying cell bodies (Figs 1C and 5).
The two methods used to detect Bgl-FaNaC mRNA in wholemount nervous systems produced comparable results (see Materials and Methods). Chromogenic detection of cRNA probes ( Fig 1C) and the fluorescent Hybridization Chain Reaction (HCR; Fig 1D) methods both produced labeling that was confined to cell somata, facilitating visualization of channel expressing cells (Fig 1E and 1F). Bgl-FaNaC expression assessed with the HCR method was widespread throughout the B. glabrata CNS (Fig 7, cyan). All eleven ganglia contained neurons expressing Bgl-FaNaC. Large intensely labeled cells were dispersed among the ganglia (Fig 8). A giant neuron was located in the inferior and lateral border of each pedal ganglion (Fig 8A). Each pleural ganglion contained one giant Bgl-FaNaC neuron (25-40 μm) near the ventromedial inferior border (Fig 8B and 8C). A single giant neuron in the ventromedial region of the right parietal ganglion was also labeled with HCR in situ hybridization for the Bgl-FaNaC transcript (Fig 8D).
The multiplexing capacity of the HCR system enabled comparison between Bgl-FaNaC localization and expression of the FaRP precursors. In agreement with previous immunohistochemical observations [35], strong expression of the tetrapeptide Bgl-FaRP1 precursor occurred in a single pair of neurons in the buccal ganglia (Figs 7 and 9A). The buccal ganglia were devoid of neurons expressing Bgl-FaRP2 (Figs 7 and 9B). While strong labeling of the Bgl-FaNaC receptor was also observed in a single pair of buccal neurons (Fig 9C), multiplexed hybridization showed that the receptor expression did not colocalize with its peptide agonist (Fig 9D).
Colocalization of the Bgl-FaRP1 and Bgl-FaNaC messages was observed in the left ventral lobe (VL) of the cerebral ganglion, a lateralized CNS region involved in penile control (Fig 10; [ 35,52,53]. While the majority of VL neurons that expressed Bgl-FaRP1 did not label for Bgl-FaNaC, colocalization did occur in a cell cluster in the anterolateral region of the lobe (Fig  10D-10F). Colocalization of receptor and agonist expression was also observed in two larger

PLOS NEGLECTED TROPICAL DISEASES
left cerebral ganglion cells that were not within the VL (Fig 10A-10C and 10G-10I). These observations suggest that the Bgl-FaNaC could play a presynaptic or autoreceptor role in the circuits that control male reproductive behavior ( [53], see Discussion).
Localization of Bgl-FaRP1 and Bgl-FaRP2 expression in the visceral and left parietal ganglia agreed with immunohistochemical findings obtained with precursor specific antibodies [35]. Abundant Bgl-FaRP1 expression was observed in the anterolateral E group (Egp) of cells in the visceral ganglion and the B group (Bgp) of the left parietal ganglion (Figs 7 and 11A). The heptapeptide Bgl-FaRP2 precursor mRNA was expressed in the posterolateral F group (Fgp) of the visceral ganglion and in a posteromedial cluster in the left parietal ganglion (Figs 7 and 11B). Bgl-FaNaC was expressed in neurons spanning the region between the Egp and the Fgp on the ventral surface of the visceral ganglion (Fig 11C). Although a few of the Bgl-FaNaC cells overlapped with the E and F groups of the visceral ganglion, no co-expression of the peptide precursors and the receptor was detected (Fig 11D).

FaRP precursor and Bgl-FaNaC expression following infection
Due to the multiplexing capability and high resolution attained with the HCR protocol, this method was utilized for experiments testing potential effects of S. mansoni infection on peptide and receptor expression in the left parietal and visceral ganglia (Table 1 and Figs 12-15). Nervous systems were dissected from snails that were not exposed to miracidia, from size-matched specimens at 20 days post infection (dpi), and from 'shedding' snails at 35 dpi. Shedding was verified by stimulating release of cercariae upon exposure to light. Expression was quantified by counting the number of cells with HCR signals above background intensity levels and by measuring the average intensity of labeled cells (Table 1). Summary data showed that the number of cells expressing the Bgl-FaRP2 heptapeptide precursor was unchanged at the time points examined (Table 1 and   the Bgl-FaRP2 HCR signals were also unchanged ( Table 1 and  Unlike the heptapeptide precursor, expression of the Bgl-FaRP1 tetrapeptide precursor was affected by infection by S. mansoni ( Table 1). The B group of the left parietal ganglion and the E group of the visceral ganglion were analyzed to determine whether these effects occurred generally or in specific cell groups. No changes in the number of cells expressing Bgl-FaRP1 or their mean intensity were detected in the left parietal Bgp (Table 1). However, the number of visceral ganglion Egp cells expressing the Bgl-FaRP1 tetrapeptide precursor increased at 35 dpi ( Fig 13A-13C Fig 13E). Together, these observations indicate that expression of the Bgl-FaRP1 tetrapeptide (FMRF-NH 2 ) precursor is increased in specific cell clusters relatively late in the infection chronology. This increase appears to occur primarily in Egp cells with low Bgl-FaRP1 expression levels prior to infection.
As the Bgl-FaNaC was not expressed in discrete clusters, changes were assessed on cell numbers and gray values obtained from the entire visceral and left parietal ganglia surfaces (Table 1 and Figs 14 and 15). No significant changes were observed in the number of Bgl-FaNaC expressing cells in either ganglion (Table 1). However, overall gray values were decreased on the ventral surface of the visceral ganglion (control: 37.50 ± 5.17; 20 dpi:  Fig 14A-14D). Bgl-FaNaC mRNA labeling was also decreased on the dorsal surface of the visceral ganglion of infected snails (control: 24.29 ± 3.39; 20 dpi: 19.15 ± 2.11; shedding: 11.21 ± 2.13; ANOVA: Bgl-FaNaC expression was also decreased in the left parietal ganglion of infected snails (  Fig 15E-15K). The decrease in Bgl-FaNaC expression on the ventral surface of the left parietal ganglion reached significant levels at 20 dpi (Fig 15H).

Properties of the Bgl-FaNaC
The FaNaC receptors that have been studied to date exhibit at least a 35-fold range of efficacy, with EC 50 values for FMRF-NH 2 varying from 2 x 10 −6 M in Cornu aspersum [25] to 7 x 10 −5

PLOS NEGLECTED TROPICAL DISEASES
M in Planorbella trivolvis [41]. The EC 50 of 3.3 x 10 −4 M observed for the Biomphalaria FaNaC in the present investigation may be interpreted in the context of structure-activity studies on the Cornu and Planorbella receptors. Chimeras constructed from these receptors indicated that the peptide recognition site is located in the extracellular region following TM1 (shaded red in Fig 2; [44,45]. Within this region, site-directed mutagenesis was used to substitute Planorbella amino acids for the Cornu aspersum residues at positions Y131Q, N134T, or I160F (enclosed by rectangles in Fig 2; [46]). Each substitution produced a FMRF-NH 2 EC 50 value that was significantly higher than the native C. aspersum receptor [46]. Notably, these three 'low affinity' residues are conserved between the planorbids Planorbella and Biomphalaria. A

PLOS NEGLECTED TROPICAL DISEASES
fourth substitution that also significantly reduced the affinity of FMRF-NH 2 , D154K (dashed rectangle , Fig 2), is not shared between Planorbella and Biomphalaria and may contribute to the apparent 4-fold difference in their efficacy.
Divergent peptide recognition sequences could also account for species differences in agonist specificity. FLRF-NH 2 , which is present in two copies on the B. glabrata FMRF-NH 2 precursor [35], activates the C. aspersum FaNaC with an EC 50 of 11 μM [25]. In contrast, responses of the Planorbella FaNaC to FLRF-NH 2 were negligible [41], in agreement with our observations on Biomphalaria. The high level of sequence conservation between the planorbids Planorbella and Biomphalaria (>90%) may therefore confer an extraordinary degree of

PLOS NEGLECTED TROPICAL DISEASES
agonist specificity in addition to the reduced efficacy of FMRF-NH 2 . The sensitivity of the Bgl-FaNaC to amiloride was not tested. Interestingly, while amiloride blocks the FaNaC of Cornu aspersa [25], it potentiates responses to FMRF-NH 2 in the Planorbella FaNaC [41]. Clearly, even within the panpulmonates, the FaNaCs exhibit striking divergent properties.

Localization of the BgFaNaC
Our findings that Bgl-FaNaC mRNA was confined to cell bodies and its abundant expression in the subesophageal (visceral and parietal) ganglia were consistent with observations in Planorbella trivolvis and Helix aspersa [51]. The predominant localization of the Bgl-FaNaC protein to neuronal processes was also in agreement with immunohistochemical observations in Planorbella trivolvis. When FMRF-NH 2 was applied to isolated giant dopaminergic neurons (GDN; corresponding to LPeD1 of Biomphalaria; Fig 6A and 6B) large inward currents were produced with focal application near the axon hillock, leading to the suggestion that newly synthesized membrane channels were inserted at a high density prior to translocation to distal sites [51].
It is well established that the FMRF-NH 2 related peptides participate in multiple neural circuits in gastropods, including the control of feeding motor programs [54,55,56], male mating behavior [57,58,59], and cardiorespiratory regulation [60,61]. Bgl-FaNaC involvement in each of these circuits was supported by its expression patterns in the buccal (Fig 9), cerebral (Fig 10), and visceral (Fig 11) ganglia, respectively. In the feeding and cardiovascular networks, Bgl-FaNaC was expressed in neurons that were in close proximity to cells that express the Bgl-FaRP1 precursor.
Co-expression of the Bgl-FaNaC and the message for the FMRF-NH 2 tetrapeptide precursor was rare, but instances were detected in regions of the cerebral ganglion that control male mating behavior (Fig 10). Such co-expression of an agonist and its receptor could enable neurons to form autapses (see [62]). Excitatory autaptic signaling has been shown to produce long-lasting after-discharges in gastropod feeding and reproductive systems [63,64,65]. Autapses are proposed to Table 1. Expression of FaRP precursors and Bgl-FaNaC in the left parietal and visceral ganglia of B. glabrata. Data were obtained from ganglia dissected 20 and 35 days post infection (dpi) with Schistosoma mansoni. Ganglia were processed with multiplexed HCR in situ hybridization (Fig 11D). provide a mechanism whereby a brief stimulus can produce a prolonged response required to drive a motor circuit. In gastropods, male copulation consists of a stereotyped sequence of actions, including preputium eversion, probing, penis eversion, and intromission [53]. Each action lasts for several minutes, probably persisting after termination of its initiating stimulus. Interestingly, application of FMRF-NH 2 to the water surrounding Biomphalaria caused preputium eversion, a behavior that lasts several minutes in the copulatory sequence [66]. The larger cerebral neurons in which Bgl-FaNaC and Bgl-FaRP1 are co-expressed (Fig 10G-10I) could provide opportunities to examine the involvement of FaNaC receptors in autapses.

Response to infection
Due to the pleiotropic functions of the FaRPs in gastropods, this neuropeptide signaling system is considered a potential target for schistosome larvae [67,68]. The increase in Bgl-FaRP1 expression observed here (Fig 13 and  pulmonate snail Lymnaea stagnalis and the avian schistosome Trichobilharzia ocellata, significant increases in FMRF-NH 2 gene expression were measured across the post-infection chronology [69]. The early onset of this increase (>300% at five hours) was suggested to reflect a direct effect of parasitism on the host brain. A lower increase at later time points (<100% at 6 and 8 weeks post-infection) was proposed to contribute to the schistosome survival strategy during the shedding stage of infection [69], when host energy resources are redirected toward the large numbers of cercariae inhabiting the snail (see [70,71]). Elevated levels of FMRF-NH 2 were also detected with liquid chromatography tandem mass spectrometry (LC-MS/MS) in B. glabrata nervous systems at 12 days post-infection with S. mansoni [72]. Of 39 CNS peptides that exhibited >1.5-fold changes, FMRF-NH 2 was one of only 6 that was increased. It was proposed that the increased expression of FMRF-NH 2 could contribute to enhanced metabolic activity during the pre-patent phase of infection [72]. Our observations suggest one potential source of elevated levels of FMRF-NH 2 in infected snails (Fig 15l). Increased precursor expression was limited to Bgl-FaRP1 (Fig 13 and Table 1), the tetrapeptide precursor that encodes FMRF-NH 2 , the sole Bgl-FaNaC agonist [35]. No changes in expression were observed for the heptapeptide precursor Bgl-FaRP2 (Fig 12 and  Table 1). Moreover, the increased expression was limited to a subset of Bgl-FaRP1expressing neurons in the visceral ganglion and was primarily observed late in the infection chronology. In contrast, down-regulation of the Bgl-FaNaC receptor appeared to commence earlier and occurred throughout the visceral and left parietal ganglia.
We propose that increased Bgl-FaRP1 expression could reflect a compensatory mechanism that occurs in response to decreased receptor expression (Fig 15l). Such homeostatic increases in neuropeptide expression would only occur in neurons that are presynaptic to neurons that express Bgl-FaNaC. Diverse mechanisms, including altered neurotransmitter release, are known to contribute to maintenance of signals following perturbation of synapses (see [73,74]). In the case of rapid signaling by FMRF-NH2 via the Bgl-FaNaC, such compensatory mechanisms could include increased precursor gene expression in response to decreased availability of postsynaptic receptors (Fig 15l). Future studies should explore the role of Bgl-FaNaC in synaptic signaling and examine whether such signaling is maintained despite reduced receptor expression levels following infection. The participation of this signaling pathway in multiple vital physiological and behavioral circuits, coupled with its extraordinary agonist specificity and apparent limitation to heterobranch taxa, could lead to novel strategies for control of snail pests.   Table 1. e-g: Decreased Bgl-FaNaC hybridization on the ventral surface of the left parietal ganglion in infected specimens. Calibration bar = 30 μm, applies to e-g. h: Group data show a significant reduction in the mean grey value on the ventral surface of the left parietal ganglion at 20 dpi and at 35 dpi. Dunnett's post hoc test: *, p < 0.05. **, p < 0.01. See values in Table 1. i-k: Large ventral left parietal ganglion neuron (panels e-g) shown at higher magnification. Calibration bar = 10 μm, applies to ik. l: Proposed responses of Bgl-FaNaC synapses to S. mansoni infection. Increased expression of the FMRF-NH 2 peptide agonist (red circles) is hypothesized to reflect a compensatory response to decreased Bgl-FaNaC expression (green receptors). Such opposing actions could exert a stabilizing influence on FMRF-NH 2 ligand-gated signaling.