Pharmacological characterization of novel heteromeric GluCl subtypes from Caenorhabditis elegans and parasitic nematodes.

BACKGROUND AND
PURPOSE: Macrocyclic lactones are the most widely used broad-spectrum anthelmintic drugs for the treatment of parasitic nematodes affecting both human and animal health. Macrocyclic lactones are agonists of the nematode glutamate-gated chloride channels (GluCls). However, for many important nematode species, the GluCls subunit composition and pharmacological properties remain largely unknown. To gain new insights into GluCl diversity and mode of action of macrocyclic lactones, we identified and pharmacologically characterized receptors made of highly conserved GluCl subunits from the model nematode Caenorhabditis elegans, the human filarial nematode Brugia malayi and the horse parasite Parascaris univalens. EXPERIMENTAL APPROACH: AVR-14, GLC-2, GLC3 and GLC-4 are the most conserved GluCl subunits throughout the Nematoda phylum. For each nematode species, we investigated the ability of these subunits to form either homomeric or heteromeric GluCls when expressed in Xenopus laevis oocytes and carried out detailed pharmacological characterization of the functional channels. KEY
RESULTS: A total of 14 GluCls were functionally reconstituted, and heteromers formation was inferred from pharmacological criteria. The GLC-2 subunit plays a pivotal role in the composition of heteromeric GluCls in nematodes. We also found a novel GluCl subtype, combining GLC-2/GLC-3 subunits, for which a high concentration of the anthelmintics ivermectin and moxidectin reversibly potentiates glutamate-induced response. CONCLUSION AND IMPLICATIONS: This study brings new insights into the diversity of GluCl subtypes in nematodes and promotes novel drug targets for the development of the next generation of anthelmintic compounds.

glutamate‐gated chloride channel

What is already known

Macrocyclic lactones are the most widely used anthelmintics for the control of parasitic nematodes.

Macrocyclic lactones act as agonists of nematode pentameric glutamate‐sensitive chloride channels (GluCls).

What does this study adds

New heteromeric GluCls from parasites as molecular targets for structurally different macrocyclic lactones.

Characterization of a novel nematode GluCl subtype on which macrocyclic lactones potentiate glutamate‐induced responses.

What is the clinical significance

Identification of targets of macrocyclic lactones in two of the most important parasitic nematode groups.

New targets for rational drug screening for the identification of next generation anthelmintic compounds.

INTRODUCTION

The phylum Nematoda is divided into five major clades (I to V) that include free living and parasitic species affecting human, animal and plant health (Blaxter & Koutsovoulos, 2015; Blaxter et al., 1998). Among these parasitic nematodes, Filarioidea and Ascaridoidea belonging to clade III are considered as those most affecting both humans and animals (Pullan et al., 2014). In this study, we focus our research on two parasitic nematode species representative of human filarids and animal ascarids: Brugia malayi, a human lymphatic filarid which is the causative agent of chronic elephantiasis in south and south‐east Asia (Global programme to eliminate lymphatic filariasis: progress report, 2016, 2017), and Parascaris sp., which is responsible for equine ascaridiosis (Reinemeyer & Nielsen, 2009; Salle et al., 2020).

Without effective vaccines or alternative strategies (Matthews et al., 2016), the use of anthelmintic treatments remains the standard control strategy for parasitic nematodes. Among the available anthelmintic drugs, the broad‐spectrum macrocyclic lactones (MLs) are highly effective and are widely used in human and veterinary medicine (Laing et al., 2017). The macrocyclic lactones include avermectins (ivermectin, doramectin, eprinomectin, abamectin, selamectin and emamectin and milbemycins (moxidectin, milbemycin‐oxime and nemadectin) that are potent against both endoparasites and ectoparasites (Campbell, 2012). The morbidity and socio‐economic impact of human lymphatic filariasis motivated control programmes led by the World Health Organization (WHO) with ivermectin as a spearhead of eradication operations (Ramaiah & Ottesen, 2014). For the control of Parascaris sp. infestations, currently, three drug classes have marketing approval including benzimidazoles, pyrantel and the macrocyclic lactones (ivermectin and moxidectin), with the last being the most widely used family. Unfortunately, because of the intensive use of macrocyclic lactones, treatment failures and resistant parasites have been reported worldwide (Kaplan & Vidyashankar, 2012; Peregrine et al., 2014). In this context, a better understanding of the mode of action of the macrocyclic lactones is essential for the development of novel treatment strategies.

In the free‐living model nematode Caenorhabditis elegans, macrocyclic lactones act as allosteric modulators of glutamate‐gated chloride channels (GluCls) (Cully et al., 1994). Exposure to the macrocyclic lactones hyperpolarizes the cell membrane and inhibits neurotransmission (Cook et al., 2006; Dent et al., 1997; Holden‐Dye & Walker, 2006), leading to a stationary paralysis of the worms (Dent et al., 2000; Hernando & Bouzat, 2014). GluCls are made of five subunits combining together to form either homomeric or heteromeric receptors (Degani‐Katzav et al., 2016; Hibbs & Gouaux, 2011). In order to investigate the subunit composition and the pharmacological properties of recombinant nematode GluCls, the Xenopus laevis oocyte has proved to be an efficient heterologous expression system (Bianchi & Driscoll, 2006).

In C. elegans, six GluCl genes were identified and named avr‐14 (Dent et al., 2000; McCavera et al., 2009), avr‐15 (Dent et al., 1997; Vassilatis et al., 1997), glc‐1 (Cully et al., 1994), glc‐2 (Cully et al., 1994; Vassilatis et al., 1997), glc‐3 (Horoszok et al., 2001) and glc‐4 (Cully et al., 1996). With the exception of GLC‐4, all the subunits are able to form functional homomeric receptors when expressed in X. laevis oocytes. All functional homomeric receptors are ivermectin‐sensitive, with the exception of Cel‐GLC‐2 (Cully et al., 1994). Interestingly, it has been reported that the C. elegans GLC‐2 subunit co‐assembles with Cel‐GLC‐1 (Cully et al., 1994) or Cel‐AVR‐15 (Vassilatis et al., 1997) to form two distinct ivermectin‐sensitive heteromeric GluCl subtypes with different pharmacological properties.

In contrast to C. elegans, few functional GluCls have been characterized so far in parasitic nematodes. The AVR‐14B subunit was reported to form functional homomeric GluCls in Haemonchus contortus (Atif et al., 2017; Cheeseman et al., 2001; McCavera et al., 2009), Cooperia oncophora (Njue et al., 2004) and Dirofilaria immitis (Yates & Wolstenholme, 2004). Interestingly, in C. oncophora and H. contortus, GLC‐2 also combined with AVR‐14B to form a heteromeric GluCl subtype sensitive to ivermectin (Atif et al., 2019; Njue et al., 2004). As GluCl investigations have mainly focused on clade V nematodes, GluCl diversity and the mode of action of macrocyclic lactones remain poorly understood in human and animal parasitic nematodes from the clade III.

In the present study, we reasoned that GluCls composed of the highly conserved AVR‐14, GLC‐2, GLC‐3 and GLC‐4 subunits could be involved in the broad‐spectrum activity of macrocyclic lactones on nematodes. Using Xenopus oocytes, we reconstituted a panel of functional homomeric and heteromeric GluCls from C. elegans, B. malayi and Parascaris univalens. We characterized a novel heteromeric GLC‐2/GLC‐3 GluCl subtype not previously described. Finally, our results provide new insights on the pharmacology of nematode GluCl subtypes as well as the mode of action of a wide range of macrocyclic lactones, thus highlighting the heteromeric GluCls as potential contributors to sensitivity to macrocyclic lactones in nematodes.

METHODS

Nematodes

C. elegans used in this study are Bristol N2 wild‐type strain worms supplied by the Caenorhabditis Genetics Center (CGC), St. Paul, MN, USA, which is funded by the NIH Office of Research Infrastructure Programs (P40 OD010440). B. malayi L4 larvae were supplied by NIH/NIAID Filariasis Research Reagent Resource Center, University of Georgia, Athens, GA, USA (www.filariasiscenter.org). Adult P. univalens were collected in faeces of naturally infested foals from UEPAO (Experimental Unit of Orfrasière Animal Physiology, INRAE Centre Val de Loire, Nouzilly 37380, France). Adult worms were collected, 50 h after a treatment with ivermectin. All samples were stored at −80°C in RNAlater solution (Ambion, Austin, TX, USA) before use. Note that because Parascaris equorum and P. univalens are morphologically identical, we confirmed the species status by karyotyping Parascaris eggs before collecting the adult worms. Briefly, parascaris eggs were extracted from faecal samples from four foals from the UEPAO. Faeces were mixed with tap water and deposited on two sieves stacked in order of size (125 μm on the top and 63 μm on the bottom). Eggs were collected and washed with tap water on the 60‐μm sieve. Karyotyping was performed as described previously (Martin et al., 2018). Whereas P. univalens has one pair of chromosome, P. equorum has two (Nielsen et al., 2014). P. univalens was identified as the single Parascaris species present in the naturally infected foals from UEPAO, because all eggs had a single pair of chromosomes (Figure S1).

Molecular biology

Total RNA was extracted from a pool of adults for C. elegans and from a pool of L4 larvae for B. malayi. For P. univalens, total RNA was extracted from the head of a single worm, including pharynx. Total RNA was isolated with Trizol reagent (Invitrogen, Carlsbad, CA, USA) as previously described (Charvet et al., 2018). cDNA synthesis was performed with 0.5–5 μg of total RNA using the Maxima H minus Reverse Transcriptase kit (Thermo Scientific, Waltham, MA, USA) according to the manufacturer's recommendations.

PCR amplifications were performed according to the manufacturer's recommendations with the Phusion High Fidelity Polymerase (New England BioLabs, Ipswich, MA, USA) using cDNA as template. Full‐length coding sequences were cloned into the transcription vector pTB‐207, and RACE‐PCR product were cloned into pGEM‐T (Promega, Madison, WI, USA). Eurofins Genomics (Luxembourg) sequenced all constructs. Sequences of Cel‐AVR‐14B (CAA04170), Cel‐GLC‐2 (AAA50786), Cel‐GLC‐3 (CAB51708) and Cel‐GLC‐4 (NP_495489.2) from C. elegans were available on GenBank as well as Pun‐AVR‐14B (ABK20343) subunit coding sequence from P. univalens. Using the GluCl sequences of C. elegans and Pun‐AVR‐14B as queries, tBLASTn searches in WormbaseParasite databank (https://parasite.wormbase.org/Tools/Blast?db=core) allowed the identification of full‐length coding sequence of GluCls from B. malayi (Bma‐GLC‐2: XM_001893073.1; Bma‐GLC‐4: XM_001900205.1; Bma‐AVR‐14B: supercontig:Bmal‐4.0:Bm_v4_Chr3_scaffold_001:1374074:1388079:1; Bma‐GLC‐3: Bmal‐4.0:Bm_v4_Chr4_scaffold_001:1513094:1542759:‐1) and partial sequences of GluCls from P. univalens (Pun‐GLC‐2: NODE_2545302; Pun‐GLC‐3: NODE_2129897 and NODE_2250308; Pun‐GLC‐4: NODE_1817943, NODE_2402242 and NODE_2418647). Primer sequences designed for RACE PCR and amplification of the full‐length coding sequences of each subunit are provided in Table S1. For GluCl subunits of P. univalens, the corresponding 5′ and 3′ cDNA ends were obtained by nested RACE‐PCR experiments as previously described (Courtot et al., 2015). The full‐length coding sequence of all GluCl subunits was obtained by nested PCR using the proofreading Phusion High‐Fidelity DNA Polymerase (Thermo Scientific, Waltham, MA, USA). Subsequently, PCR products were cloned into the transcription vector pTB‐207 as previously described (Charvet et al., 2018). Recombinant constructs were sequence checked and linearized with MlsI, PaeI or PacI restriction enzymes (Thermo Scientific, Waltham, MA, USA) depending on the construct. Linearized plasmids were used as DNA templates for cRNA synthesis using the mMessage mMachine T7 transcription kit (Ambion, Austin, TX, USA). cRNAs were precipitated with lithium chloride and were resuspended in a suitable volume of RNAse‐free water and stored at −20°C before use.

Sequence analysis and phylogeny

Deduced amino‐acid sequences of GluCl subunits from B. malayi, C. elegans, H. contortus and P. univalens were aligned using the MUSCLE algorithm included in the Seaview 5.0.4 package (Gouy et al., 2010). Percentage of identity between deduced amino acids of mature protein without peptide signal was obtained using the EMBOSS Needle software (Madeira et al., 2019). Signal peptide predictions were performed using the SignalP 3.0 server (Bendtsen et al., 2004), and membrane‐spanning regions were predicted using the SMART server (Schultz et al., 1998). Phylogenetic analyses were performed on coding sequences predicted from genomic data available in databases or cloned cDNA sequences when available. Maximal likelihood phylogeny reconstruction was performed using PhyML V3.1 included in the Seaview 5.0.4 package. Significance of the internal tree branches was estimated using bootstrap with 100 replicates. The tree was edited using the FigTree software (http://tree.bio.ed.ac.uk/software/figtree/). The sequences used in this study are available on GenBank under the following accession numbers: B. malayi (Bma): AVR‐14B (MW196269), GLC‐2 (MW196266), GLC‐3 (MW196267), GLC‐4 (MW196268); C. elegans (Cel): AVR‐14B (MW196270), AVR‐15A (CAA04171), AVR‐15B (CAA04170), GLC‐1 (AAA50785), GLC‐2 (AAA50786), GLC‐3 (CAB51708), GLC‐4 (NP_495489.2), UNC‐49B (AAD42383); H. contortus (Hco): AVR‐14B (CAA74623), GLC‐2 (CAA70929); P. univalens (Pun): AVR‐14B (MW187941), GLC‐2 (MW187938), GLC‐3 (MW187939), GLC‐4 (MW187940).

Electrophysiological recording in oocytes

Distinct batches of defolliculated X. laevis oocytes were purchased from Ecocyte Bioscience (Dortmund, Germany) and micro‐injected using the Drummond nanoject II microinjector with 72 ng of cRNA when subunits were expressed singly or with 50 ng for each subunit when expressed in combination (1:1 ratio). Two‐electrode voltage‐clamp recordings were performed at a holding potential of −80 mV to assess the expression of the GluCl channels as previously described (Lamassiaude et al., 2021). Currents were recorded and analysed using the pCLAMP 10.4 package (Molecular Devices, San Jose, CA, USA). All recordings were performed at room temperature (20 ± 1°C).

Dose–responses relationships for glutamate were carried out by challenging oocytes with 5–10 s of application of increasing concentration of glutamate (from 1 μM up to 30 mM, depending on the receptor) with 2‐min washing steps with Ringer solution between each application. The peak current values were normalized to the maximum response obtained with a saturating concentration of glutamate. Comparison of the effects of the macrocyclic lactones on receptors was performed as described previously (Njue et al., 2004; Yates & Wolstenholme, 2004). Briefly, 1‐mM glutamate was first perfused on the oocytes as a reference peak current (maximum response) before application of a macrocyclic lactone at 1 μM for 5 s. Current responses were normalized to the maximum current amplitude obtained with 1‐mM glutamate. For Pun‐GLC‐2/GLC‐3, the potentiating effect of macrocyclic lactones was evaluated by a first application of each macrocyclic lactone alone for 5 s, followed by the co‐application with 100‐μM glutamate for 5 s. Then, in order to monitor a potential reversible effect, the oocytes were washed for 2 min with Ringer solution before a subsequent application of 100‐μM glutamate. The observed responses were normalized to the response induced by 100‐μM glutamate (corresponding approximately to the glutamate EC50 for Pun‐GLC‐2/GLC‐3) alone, performed prior to challenging with the macrocyclic lactone.

Data and statistical analysis

The data and statistical analysis comply with the recommendations of the British Journal of Pharmacology on experimental design and analysis in pharmacology (Curtis et al., 2018). Each experiment corresponds to current recordings made on at least five different oocytes (n) from at least five distinct defolliculated egg batches. Only one drug was tested on each oocyte. In order to minimize variation, current amplitudes were normalized to the maximal effect obtained with a saturating glutamate concentration. None of the experiments were blinded or randomized as pharmacological characterization of the GluCls were performed with a set of selected drugs. The group sizes varied with the type of experiment performed. In order to investigate putative functional recombinant homomeric or heteromeric GluCls, responsive or not to 1‐mM glutamate, at least eight recordings were performed on independent oocytes for each putative receptor. For the heteromeric GluCls selected for further pharmacological characterizations (i.e., AVR‐14B/GLC‐2 and GLC‐2/GLC‐3), expression robustness (variability between oocytes) was evaluated with at least 17 recordings performed on independent oocytes. For each GluCl subtype, glutamate dose–response and agonist/antagonist experiments were performed on at least five independent oocytes. Statistical analysis was undertaken only for studies where each group size was at least n = 5. All statistical analyses were performed using the software R (v3.6.1). All data were tested for normality with a Shapiro–Wilk test, and outliers were included in data analysis and presentation. When data followed a normal distribution, statistical significance was calculated using one‐way ANOVA with post hoc Tukey's HSD test using the Bonferroni correction for comparisons among more than two groups (Figure 4). Post hoc tests were run only if F achieved P < 0.05 and there was no significant variance in homogeneity using Levene's test. P < 0.05 was considered statistically significant. The Mann–Whitney test was performed to compare the effect of a single drug on similar receptors from distinct nematode species (Figure 6). For paired data (Figure 7), a non‐parametric Friedman test followed by a post hoc Conover multiple comparison test using the Bonferroni correction was used. Dose–response data were modelled as a log‐logistic regression with two parameters, that is, the curve slope and the associated EC50 concentration, using the drc package v3.0‐1.

Materials

Glutamate, piperazine and the macrocyclic lactones (selamectin, ivermectin, doramectin, emamectin, eprinomectin, abamectin and moxidectin) were purchased from Sigma‐Aldrich (Saint‐Quentin Fallavier, France). These macrocyclic lactones were first dissolved in DMSO at 10 mM and then diluted in Ringer solution to the required concentration with a final concentration of DMSO which did not exceed 1%. Glutamate was directly prepared in recording solution.

Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in the IUPHAR/BPS Guide to PHARMACOLOGY http://www.guidetopharmacology.org and are permanently archived in the Concise Guide to PHARMACOLOGY 2021/22 (Alexander, Mathie et al., 2021).

RESULTS

Four distinct GluCl subunits are conserved between , B. malayi and P. univalens

Searches for homologues of GluCl subunit encoding genes from C. elegans (i.e., glc‐1, glc‐2, glc‐3, glc‐4, avr‐14 and avr‐15) (Cully et al., 1994; Dent et al., 1997, 2000; Horoszok et al., 2001; Vassilatis et al., 1997) in B. malayi and P. univalens genomic/transcriptomic databases allowed the identification of four independent sequences corresponding to putative homologues of avr‐14, glc‐2, glc‐3 and glc‐4 from both species. In contrast, homologues could not be found in B. malayi and P. univalens for glc‐1 or avr‐15. The full‐length cDNAs of avr‐14, glc‐2, glc‐3 and glc‐4 from the three nematode species were cloned, sequenced and submitted to GenBank under the accession numbers provided in Section 2. All GluCl subunits identified including AVR‐14B, GLC‐2, GLC‐3 and GLC‐4 from C. elegans, B. malayi and P. univalens present the typical characteristics of a Cys‐loop receptor subunit. This includes a predicted signal peptide in the extracellular N‐terminal part (with the exception of Bma‐GLC‐2), the first Cys‐loop domain specific for ligand‐gated ion channels (LGIC) composed of two cysteines that are 13 amino‐acid residues apart, the second Cys‐loop domain found in GluCls, and four transmembrane domains (TM1–4) (Figures S2 and S3). In comparison with C. elegans, the GluCl‐deduced amino‐acid sequences of B. malayi and P. univalens, respectively, show an identity of 78.4% and 80.3% for AVR‐14B, 64.1% and 67.2% for GLC‐2, 66.5% and 67.8% for GLC‐3 and 64.8% and 66.6% for GLC‐4 (Table S2). A phylogenetic analysis including GluCl sequences from C. elegans (Cel), B. malayi (Bma), H. contortus (Hco) and P. univalens (Pun) confirmed the orthologous relationship of the parasitic subunit sequences with their respective counterparts in C. elegans (Figure 1). The identified sequences were named according to the nomenclature proposed by Beech et al. (2010). The full‐length cDNA sequences of all the subunits (including Bma‐GLC‐2 lacking a predicted signal peptide) were cloned into a transcription vector for subsequent analysis.

FIGURE 1

Maximal likelihood tree generated with GluCl subunit deduced amino‐acid sequences from Brugia malayi (Bma), (Cel) and Parascaris univalens (Pun). The bootstrap values (100 replicates) are indicated at each node. Scale bar represents the number of substitutions per site. Accession numbers for the sequences used in this analysis are provided in Section 2. The B. malayi, and P. univalens GluCl subunit sequences investigated in the present study are highlighted in green, red and blue, respectively. The GABA receptor subunit UNC‐49B from was used as an outgroup

subunits are more prone to reconstitute robustly expressed homomeric glutamate‐sensitive receptors than their parasite counterparts

In order to test the ability of C. elegans, B. malayi and P. univalens GluCl subunits to form functional homomeric receptors, their respective cRNAs (avr‐14b, glc‐2, glc‐3 and glc‐4) were injected singly into X. laevis oocytes. Four days after injection, currents elicited by 1‐mM glutamate application were recorded using the two‐electrode voltage‐clamp technique (Figure 2a).

FIGURE 2

Functional expression of GluCl subunits from , Brugia malayi and Parascaris univalens in oocytes. (a,b) Box plots depict response elicited by 1‐mM glutamate application on Xenopus oocytes expressing homomeric (a) or heteromeric (b) receptors for , B. malayi and P. univalens. HR (high response) corresponds to subunit combinations that led to robust expression of receptors responding to 1‐mM glutamate with peak current in the μA range. LR (low response) corresponds to subunit combinations that respond to 1‐mM glutamate with moderate currents in the nA range. NR (no response) corresponds to combinations that did not respond to 1‐mM glutamate application

Here, the C. elegans AVR‐14B, GLC‐2 and GLC‐3 GluCl subunits were used as positive controls as their ability to form functional glutamate‐sensitive homomeric channels has been previously reported (Cully et al., 1994; Dent et al., 2000; Horoszok et al., 2001; McCavera et al., 2009). In our hands, Cel‐AVR‐14B or Cel‐GLC‐2 alone formed functional homomeric receptors with robust glutamate‐elicited currents in the μA range: (as medians) 1.6 μA (n = 16) and 9.1 μA (n = 32) for Cel‐AVR‐14B and Cel‐GLC‐2, respectively. Although Cel‐GLC‐3 was also able to form functional receptors, the currents elicited by 1‐mM glutamate were in the nA range (median, 176 nA, n = 50). Finally, no 1‐mM glutamate‐induced current was observed on oocytes injected with Cel‐GLC‐4 cRNA (n = 18). These results are in agreement with previous studies (Cully et al., 1994, 1996; Dent et al., 2000; Horoszok et al., 2001).

Unlike C. elegans, where three out of four GluCl subunits formed functional homomeric receptors, for B. malayi, only the expression of AVR‐14B in oocytes allowed the recording of reliable currents (median, 134 nA, n = 23). For P. univalens, AVR‐14B and GLC‐3 reconstituted functional homomeric receptors, with 1‐mM glutamate‐elicited median currents of 103.6 nA (n = 9) and 70 nA (n = 11), respectively. Finally, in oocytes injected with Bma‐glc‐2 (n = 11), Pun‐glc‐2 (n = 11), Bma‐glc‐3 (n = 14), Bma‐glc‐4 (n = 13) or Pun‐glc‐4 (n = 12), 1‐mM glutamate application did not produce detectable responses.

GLC‐2 plays a pivotal role in heteromeric GluCl composition

Because previous studies highlighted the involvement of GLC‐2 in heteromeric GluCls (Atif et al., 2019; Cully et al., 1994; Vassilatis et al., 1997), oocytes expressing a combination of GLC‐2 with AVR‐14B, GLC‐3 or GLC‐4 were challenged with 1‐mM glutamate (Figure 2b).

Firstly, we investigated the AVR‐14B/GLC‐2 combination for the three different nematode species. As expected, the co‐expression of both subunit cRNAs from C. elegans led to the robust expression of GluCls with a median current amplitude of 5.2 μA (n = 21). Surprisingly, the B. malayi AVR‐14B/GLC‐2 combination led to the robust expression of receptors responsive to 1‐mM glutamate with a median current amplitude of 5.1 μA (n = 31), corresponding to a 38‐fold increase, in comparison with the homomeric receptor made of Bma‐AVR‐14B alone. Note that such a result was unexpected as Bma‐GLC‐2 lacks a predicted signal peptide. The P. univalens AVR‐14B/GLC‐2 combination also led to the robust expression of functional receptors with 1‐mM glutamate‐elicited currents of 2.0 μA (n = 17), corresponding to a 19‐fold increase in comparison with Pun‐AVR‐14B alone. These first results indicated that AVR‐14B and GLC‐2 could assemble into functional heteromeric receptors potentially distinguishable from homomeric receptors.

Secondly, we tested the ability of the GLC‐2 subunit to assemble with GLC‐3. For the C. elegans GLC‐2/GLC‐3 combination, we recorded strong currents elicited by 1‐mM glutamate (median 3.8 μA, n = 59). Strikingly, for P. univalens, the combination of GLC‐2/GLC‐3 also led to the robust expression of glutamate‐sensitive receptors with a median current amplitude of 3.8 μA (n = 45), corresponding to a 54‐fold increase in comparison with the currents recorded on the Pun‐GLC‐3 homomeric receptor. These results suggested that Pun‐GLC‐2 and Pun‐GLC‐3 could assemble into heteromeric receptors. In contrast, for B. malayi, the GLC‐2/GLC‐3 combination failed to give rise to a functional receptor responsive to 1‐mM glutamate (n = 8). Because Bma‐GLC‐2 has proved to be functional in combination with AVR‐14B, we addressed the functionality of Bma‐GLC‐3 by co‐injecting Bma‐glc‐3 together with Pun‐glc‐2 in Xenopus oocytes. The recordings of glutamate‐elicited currents in the μA range resulting from a functional chimeric receptor made of Pun‐GLC‐2 and Bma‐GLC‐3 confirmed that Bma‐GLC‐3 is a functional subunit (Figure S4). Therefore, the Bma‐GLC‐2 and Bma‐GLC‐3 subunits do not assemble to form a functional heteromeric channel responsive to 1‐mM glutamate although both are functional subunits.

Thirdly, we tested the combination of GLC‐2 with GLC‐4 for all species. For the C. elegans GLC‐2/GLC‐4 combination, median currents of 592 nA (n = 14) were recorded, corresponding to a 15‐fold decrease in comparison with Cel‐GLC‐2 alone. Note that such a reduction of glutamate‐induced response has been previously reported, when Cel‐GLC‐2 was co‐expressed with Cel‐GLC‐1 (Cully et al., 1994). In contrast with C. elegans, the B. malayi and the P. univalens GLC‐2/GLC‐4 combination failed to produce detectable glutamate responses (n = 8 and n = 11, respectively).

Finally, we tested the combination of GLC‐3 with the GLC‐4 subunit for all species. No currents could be recorded with 1‐mM glutamate application on oocytes injected with Bma‐GLC‐3/GLC‐4 (n = 13), although nA range currents were recorded for Cel‐GLC‐3/GLC‐4 and for Pun‐GLC‐3/GLC‐4. Taken together, these results highlight the role of GLC‐2 from the three different species in the formation of glutamate‐sensitive heteromeric receptors, including either AVR‐14B or GLC‐3.

The AVR‐14B/GLC‐2 GluCl subtype is targeted by a wide range of MLs

In order to distinguish the putative Cel‐AVR‐14B/GLC‐2 heteromeric receptor from the Cel‐AVR‐14B and Cel‐GLC‐2 homomeric channels, we determined their respective glutamate concentration–response curves. Glutamate EC50 values (shown as geometric means with 95% confidence intervals) of 112.2 (88.3 ‐ 142.8) μM (n = 5), 214.7 (189 ‐ 243.9) μM (n = 5) and 19.4 (17.2 ‐ 22)   μM (n = 8) were obtained for the Cel‐AVR‐14B, Cel‐GLC‐2 and Cel‐AVR‐14B/GLC‐2 receptors, respectively (Figure 3a). Cel‐AVR‐14B/GLC‐2 channels exhibited a significantly higher sensitivity for glutamate in comparison with the homomeric receptors formed by AVR‐14B or GLC‐2, with 5‐ and 10‐fold lower EC50, respectively. These results strongly support the view that Cel‐AVR‐14B and GLC‐2 can combine to form a functional heteromeric GluCl receptor distinguishable from the respective homomers.

FIGURE 3

Glutamate concentration–response relationships for the AVR‐14B/GLC‐2 receptors from , Brugia malayi and Parascaris univalens. Glutamate concentration–response curves for AVR‐14B/GLC‐2 (red), AVR‐14B (blue) and GLC‐2 (green) from C. elegans (a), B. malayi (b) and P. univalens (c). Current amplitudes were normalized to the maximal effect obtained with a saturating glutamate concentration. Solid line stands for the fitted log‐logistic regression curve, and shaded area indicates 95% confidence interval. For both and B. malayi, the glutamate EC50s were significantly different (P<0.05) between the heteromeric AVR‐14B/GLC‐2 receptors and the respective homomeric AVR‐14 and GLC‐2 channels

For B. malayi, we determined the glutamate EC50 values for Bma‐AVR‐14B and Bma‐AVR‐14B/GLC‐2 but not for GLC‐2 alone as it did not form a functional homomeric receptor responsive to 1‐mM glutamate. For the AVR‐14B homomeric channel, the glutamate EC50 value was 761.4 (618.2 ‐ 937.7) μM (n = 5), whereas the EC50 of the Bma‐AVR‐14B/GLC‐2 combination was 32.5 (24 ‐ 44.1) μM (n = 7) (Figure 3b). This drastic shift of glutamate sensitivity (23‐fold) provides strong evidence that AVR‐14B and GLC‐2 subunits from B. malayi associate to form a heteromeric GluCl subtype, distinct from the homomeric receptor composed of AVR‐14B.

For the P. univalens homomeric AVR‐14B receptor, the glutamate EC50 value could not be determined. Indeed, the application of high concentrations of glutamate only induced small currents that did not reached a plateau value. Note that even with 100‐mM glutamate applications (n = 7), the receptor was not saturated. In sharp contrast, the co‐injection of avr‐14B and glc‐2 cRNAs led to the robust expression of a functional receptor with a glutamate EC50 value of 10.9 (8.8 – 13.6) μM (n = 5) (Figure 3c). Taken together, these results confirmed that Pun‐AVR‐14B/GLC‐2 form a heteromeric GluCl.

Subsequently, in order to get new insights into their respective pharmacological properties, the heteromeric AVR‐14B/GLC‐2 GluCls from C. elegans, B. malayi and P. univalens were challenged with a wide range of marketed macrocyclic lactone compounds used as anthelmintic, insecticide or acaricide (abamectin, doramectin, emamectin, eprinomectin, ivermectin, moxidectin and selamectin (Figure 4 and Table 1). Recordings revealed that all the macrocyclic lactones tested in the present study acted as potent agonists on these receptors, inducing their permanent activation. Large currents were recorded in response to 1 μM for all the macrocyclic lactones tested. Representative currents induced by the application of macrocyclic lactones on oocytes expressing the AVR‐14B/GLC‐2 receptor from each species are presented in Figure 4a,c,e. As previously described, receptor activation by the macrocyclic lactones was not reversible and slower in comparison with glutamate. Interestingly, the insecticide emamectin was found to be more potent than the standard drug ivermectin on both Cel‐AVR‐14B/GLC‐2 and Bma‐AVR‐14B/GLC‐2. In contrast, no significant difference of sensitivity between ivermectin and the other macrocyclic lactones was observed on Pun‐AVR‐14B/GLC‐2 (Figure 4b,d,f).

FIGURE 4

Effects of macrocyclic lactones on AVR‐14B/GLC‐2 receptors from , Brugia malayi and Parascaris univalens. (a,c,e) Representative recording traces from a single oocyte injected with AVR‐14B and GLC‐2 of (a), B. malayi (b) and P. univalens (c) induced by 1‐μM ivermectin and 1‐μM emamectin after a first application of 1‐mM glutamate. Application times are indicated by the black bars. (b,d,f) Boxplots show the effects of the macrocyclic lactones (1 μM) after 5‐s application on Cel‐AVR‐14B/GLC‐2 (c), Bma‐AVR‐14B/GLC‐2 (d) and Pun‐AVR‐14B/GLC‐2 (e). All responses were normalized to the maximum responses obtained with glutamate at 1 mM. *P < 0.05, significantly different from effects of ivermectin

TABLE 1

Effect of 1‐μM application of different macrocyclic lactones on , Brugia malayi and Parascaris univalens AVR‐14B/GLC‐2 receptors

	IVM	MOX	ABA	DOR	EMA	EPR	SEL
C. elegans AVR‐14B/GLC‐2	17.6 ± 2.7 (n = 6)	27.3 ± 4.7 (n = 7)	11.9 ± 2.0 (n = 9)	33.4 ± 6.9 (n = 8)	77.7 ± 7.3 (n = 7)	35.9 ± 2.2 (n = 8)	32.0 ± 4.4 (n = 8)
B. malayi AVR‐14B/GLC‐2	27.0 ± 7.1 (n = 7)	63.1 ± 6.7 (n = 9)	34.1 ± 9.7 (n = 7)	41.8 ± 4.8 (n = 7)	80.5 ± 6.0 (n = 7)	30.8 ± 5.4 (n = 7)	47.9 ± 7.1 (n = 11)
P. univalens AVR‐14B/GLC‐2	19.0 ± 3.8 (n = 7)	9.3 ± 1.1 (n = 5)	9.4 ± 2.3 (n = 6)	5.4 ± 1.8 (n = 5)	24.9 ± 4.4 (n = 5)	15.1 ± 4.4 (n = 9)	2.5 ± 0.6 (n = 5)

Note: Values are means ± SEM, representing percentage of 1‐mM glutamate‐induced response.

Abbreviations: ABA, abamectin; DOR, doramectin; EMA, emamectin; EPR, eprinomectin; IVM, ivermectin; MOX, moxidectin; SEL, selamectin.

GLC‐2/GLC‐3: A new functional heteromeric GluCl subtype with original pharmacological properties

In order to distinguish the putative Cel‐GLC‐2/GLC‐3 heteromeric receptor from the Cel‐GLC‐2 and Cel‐GLC‐3 homomeric receptors, we compared their respective glutamate EC50 values (Figure 5a).

FIGURE 5

Glutamate concentration–response relationships for GLC‐2/GLC‐3 receptors from and Parascaris univalens. (a,b) Glutamate concentration–response curves for GLC‐2 and the GLC‐2/GLC‐3 combination (a); P. univalens GLC‐3 and the GLC‐2/GLC‐3 combination (b). Current amplitudes were normalized to the maximal effect obtained with glutamate. Solid line stands for the fitted log‐logistic regression curve, and shaded area indicates 95% confidence interval. For both nematode species, the glutamate EC50 values were significantly different between the heteromeric and homomeric receptors

For Cel‐GLC‐2/GLC‐3, the glutamate EC50 value was 46.8 (40.3 – 54.2) μM (n = 5), corresponding to a 4‐fold reduction in comparison with the glutamate EC50 value previously determined for Cel‐GLC‐2 alone (214.7 (189 – 243.9) μM, n = 5) and a 40‐fold reduction in comparison with the glutamate EC50 value of Cel‐GLC‐3 (1.9 ± 0.03 mm) previously reported by Horoszok et al. (2001). The shift in these EC50 values indicates that C. elegans GLC‐2 and GLC‐3 can combine to form a new GluCl subtype with a higher affinity for glutamate, than that found with the homomeric receptors.

For P. univalens GLC‐2/GLC‐3, the glutamate concentration–response curve revealed an EC50 of 120.2 (109.2 – 132.3) μM (n = 21) whereas the EC50 for Pun‐GLC‐3 was 1482.2 (1275.6 – 1723.6) μM (n = 11) (Figure 5b). This 12‐fold shift of EC50 values also confirmed that Pun‐GLC‐2/GLC‐3 forms a novel subtype of heteromeric GluCl distinct from the Pun‐GLC‐3 channel.

In order to evaluate the involvement of GLC‐2/GLC‐3 as putative molecular targets for macrocyclic lactones, we focused our attention on ivermectin, moxidectin and eprinomectin representing the most commonly used anthelmintic macrocyclic lactones in equids and ruminants (i.e., ivermectin and moxidectin used for Parascaris sp. treatment; eprinomectin used for clade V parasites on lactating animals). Strikingly, whereas the Cel‐GLC‐2/GLC‐3 receptor was activated by 1‐μM ivermectin, moxidectin or eprinomectin, surprisingly, for P. univalens, none of the tested macrocyclic lactones induced a current (Figure 6). This last result was further confirmed using a wider panel of macrocyclic lactones (Figure S5a). In order to investigate if the drug application time could potentially alter the effects of the macrocyclic lactone, 1‐μM ivermectin was perfused for 90 s on Pun‐GLC‐2/GLC‐3. Even with this long‐lasting application, ivermectin remained poorly effective as an agonist with a median current amplitude corresponding to 0.37% (n = 7) of the maximum response obtained with glutamate (Figure S5b).

FIGURE 6

Effects of ivermectin, moxidectin and eprinomectin on GLC‐2/GLC‐3 receptors from and Parascaris univalens. Boxplots depict 1‐μM ivermectin, moxidectin or eprinomectin effects after 5‐s application on GLC‐2/GLC‐3 from and P. univalens. All currents were normalized to 1‐mM glutamate‐elicited currents. Ivermectin, moxidectin or eprinomectin applications induced higher currents in GLC‐2/GLC‐3 receptors from *P < 0.05, significantly different from

Because of this unexpected lack of activity as agonists, we hypothesized that the macrocyclic lactones could act both as antagonists or potentiators of the glutamate effect on Pun‐GLC‐2/GLC‐3. In order to test this hypothesis, 100‐μM glutamate (corresponding approximately to the receptor glutamate EC50) was applied before, during and after the addition of 1‐μM ivermectin or moxidectin on the receptor (Figure 7a,b). Strikingly, both drugs potentiated the effects of glutamate on Pun‐GLC‐2/GLC‐3. Median current amplitudes in response to 100‐μM glutamate were increased by 50% (n = 5) and 75% (n = 8), when co‐applied with ivermectin or moxidectin, respectively (Figure 7c,d). This effect was reversible for both ivermectin and moxidectin. Whereas it has been previously reported that concentrations of ivermectin that were too low to directly activate the channels would nonetheless potentiate the effects of glutamate (Cully et al., 1994), this is, to our knowledge, the first evidence of a nematode GluCl response potentiated by macrocyclic lactones at high concentrations, thus representing a novel receptor subtype with unique pharmacological properties.

FIGURE 7

Modulation of glutamate effect by ivermectin and moxidectin on GLC‐2/GLC‐3 receptor from Parascaris univalens. (a,b) Representative current traces induced by glutamate 100 μM followed by a co‐application of ivermectin 1 μM (a) or moxidectin 1 μM (b) with glutamate 100 μM and a final glutamate 100 μM application to monitor a potential reversible effect. Application times are indicated by the black bars. (c,d) Boxplots of summary data showing potentiating effect of ivermectin (c) and moxidectin (d) on Pun‐GLC‐2/GLC‐3 normalized and compared with the response to 100‐μM glutamate. Note that the effects of the macrocyclic lactones were reversible on washout. *P < 0.05, significantly different as indicated

DISCUSSION

Diversity of GluCls in distantly related nematodes

Among the six distinct genes encoding GluCl subunits in the free‐living model nematode C. elegans, avr‐14, glc‐2, glc‐3 and glc‐4 are highly conserved in distantly related nematode species from different phylogenetic clades (Williamson et al., 2007). Here, we identified GluCls composed of the GLC‐2, GLC‐3 and AVR‐14B subunits from C. elegans, B. malayi and P. univalens, two parasitic nematodes presenting a major impact for human and equine health, respectively. In C. elegans, we confirmed that homomeric receptors composed of AVR‐14B or GLC‐2 are responsive to glutamate in the μM range with elicited currents at the μA range, whereas, in contrast, the GLC‐3 homomeric receptor only responds to mM glutamate application resulting in small currents at the nA range. Strikingly, for the latter, we showed that addition to GLC‐2 led to the formation of a novel heteromeric GLC‐2/GLC‐3 receptor, responsive to more physiologically relevant glutamate concentrations. Even though it remains speculative to consider that such differences could reflect the existence of such a subunit combination in vivo, it clearly highlights the need to further investigate heteromeric GluCls in nematodes.

In clade V nematode species such as H. contortus (Atif et al., 2017, 2019) and C. oncophora (Njue et al., 2004), homomeric glutamate‐sensitive channels made of AVR‐14B or GLC‐2 have also been described, suggesting that such recombinant homomeric GluCl channels could also be obtained from other parasitic nematode species. However, in the present study, we showed that none of the GluCl subunits from the parasites B. malayi or from P. univalens (i.e., AVR‐14B, GLC‐2, GLC‐3 and GLC‐4) gave rise to robust functional channels when expressed in the Xenopus oocytes (i.e., no glutamate elicited current, or small current at the nA range elicited by mM range glutamate). These results further supported the need to explore heteromeric GluCls in target parasitic species.

The GLC‐2 subunit is a key player for heteromeric GluCls

Previous studies reported that GLC‐2 associates with GLC‐1 or AVR‐15 in C. elegans and with AVR‐14B in H. contortus and C. oncophora (Atif et al., 2019; Cully et al., 1994; Njue et al., 2004; Vassilatis et al., 1997). In addition, the C. elegans AVR‐14, GLC‐2, GLC‐3 and GLC‐4 subunits have been shown to be expressed in pharyngeal neurons, suggesting potential interactions between these subunits in the worm (Cao et al., 2017; Hutter & Suh, 2016). It is noteworthy that the down‐regulation of glc‐2 in C. elegans is embryonically lethal (Sonnichsen et al., 2005) and that the avr‐14, avr‐15, glc‐1 triple mutant displays only subtle behavioural changes and is resistant to 4 μg·ml−1 ivermectin (Dent et al., 2000). Interestingly, the avr‐14, avr‐15, glc‐1, glc‐3 quadruple mutant strain is also viable, fertile with very subtle phenotype but is able to survive up to 50 μg·ml−1 ivermectin exposure (Wever et al., 2015). Taken together, these data clearly support a major role for the GLC‐2 subunit in C. elegans but also highlight GLC‐3 as an important contributor to ivermectin susceptibility.

For C. elegans, B. malayi and P. univalens, the combination of AVR‐14B/GLC‐2 led to the robust expression of glutamate‐sensitive receptors. The drastic reduction of the glutamate EC50 of AVR‐14B/GLC‐2 in comparison with their respective homomeric receptors strongly support the association of the two distinct subunits into functional heteromeric receptors. Subsequently, we describe for the first time that GLC‐2 and GLC‐3 can associate to form a novel glutamate‐sensitive channel subtype in C. elegans and P. univalens, but surprisingly not in B. malayi. Although the Bma‐GLC‐2 and Bma‐GLC‐3 subunits were shown to be functional with Bma‐AVR‐14B and Pun‐GLC‐2, respectively, they did not give rise to a heteromeric GluCl. Therefore, we can only speculate that additional subunits combining to GLC‐3 alone or GLC‐2/GLC‐3 of B. malayi are required to form functional receptors. In addition, we cannot exclude that ancillary proteins might be required for the functional expression of GluCl receptor in Xenopus oocytes as reported for some ACh‐receptor subtypes (Boulin et al., 2008, 2011; Courtot et al., 2015). Interestingly, as in C. elegans, the glc‐3 gene is present in H. contortus and C. oncophora, but the heteromeric combination with GLC‐2 has never been tried so far. Therefore, the present study opens the way for the characterization of the new GLC‐2/GLC‐3 receptor subtype in other nematode species. In summary, our results highlight that subunit combination is critical for clade III parasites to form functional GluCls system but also contribute to the diversity of GluCl subtypes in clade V nematodes, including the model species C. elegans.

New insights into the mode of action of macrocyclic lactones in and parasitic nematodes

In accordance with previous studies performed on AVR‐14B/GLC‐2 of H. contortus (Atif et al., 2019) and C. oncophora (Njue et al., 2004), we showed that AVR‐14B/GLC‐2 from C. elegans, B. malayi and P. univalens were sensitive to ivermectin and moxidectin, corresponding to the two macrocyclic lactones with marketing authorization in human health (de Moraes & Geary, 2020) and the most widely used in equine health with abamectin and doramectin (Gokbulut & McKellar, 2018).

In addition, we showed that this GluCl subtype is also sensitive to a wide range of macrocyclic lactones including abamectin, doramectin, emamectin, eprinomectin and selamectin. Presently, only ivermectin has marketing authorization for a wide variety of hosts and parasites. Other macrocyclic lactones have more specific authorization, such as emamectin, which is mostly used as an insecticide in veterinarian aquaculture as well as in terrestrial agriculture (Lees et al., 2008). Interestingly, the present work highlights emamectin as a more potent agonist than ivermectin on the AVR‐14B/GLC‐2 receptors from C. elegans and B. malayi. However, whether the high potency of emamectin on AVR‐14B/GLC‐2 could be correlated or not with an efficacy of this drug on the worms remains to be established. Interestingly, this also raises the question of the relative contribution of the different nematode GluCl subtypes in the sensitivity to macrocyclic lactones. Here, we report a new subtype of functional GluCls made of GLC‐2/GLC‐3 in C. elegans and P. univalens. However, depending on the nematode species, they presented very contrasting pharmacological properties. For C. elegans, ivermectin, moxidectin and eprinomectin acted as agonists on GLC‐2/GLC‐3, thus underpinning a potential role for this heteromeric channel in its sensitivity to MLs. In that respect, our results provide an explanation for the drastic increase of ivermectin resistance in the avr‐14, avr‐15, glc‐1, glc‐3 quadruple mutant in comparison with the avr‐14, avr‐15, glc‐1 triple mutant (Dent et al., 2000; Wever et al., 2015). In sharp contrast, for P. univalens, ivermectin, moxidectin and eprinomectin potentiated glutamate effect on its GLC‐2/GLC‐3 receptor. Such differences in pharmacological properties between nematode species highlight the need for future investigations of GLC‐2/GLC‐3 in other parasitic species as it could provide important information about amino acids involved in the receptor binding sites.

Interestingly, in some case of ivermectin resistance, moxidectin remains effective to treat lambs infected with H. contortus (Williamson et al., 2011), suggesting that each molecule could preferentially activate distinct pharmacological targets in the worms. Because stable transformation remains an elusive goal for numerous parasitic nematode species (Boyle & Yoshino, 2003; Selkirk et al., 2012), RNAi experiments could represent an attractive alternative to investigate the respective role of the distinct GluCls subtype in determining the susceptibility to macrocyclic lactones (Blanchard et al., 2018). RNAi has been successfully developed in B. malayi (Song et al., 2010) and in the pig nematode parasite Ascaris suum that is closely related to P. univalens (McCoy et al., 2015). Undoubtedly, such an approach combined with phenotypic assays (Storey et al., 2014) or the recently developed in vivo imaging system (IVIS) optimized to study B. malayi on jird (Liu et al., 2020) would represent a major opportunity to investigate in more detail the putative contribution of AVR‐14B/GLC‐2 and GLC‐2/GLC‐3 GluCl subtypes in the sensitivity to macrocyclic lactones.

In conclusion, our study provides new insight about the GluCl diversity and highlights the importance of GLC‐2 as a core subunit in heteromeric GluCls from C. elegans and the clade III nematodes B. malayi and P. univalens. This work opens the way for the systematic investigation of heteromeric GluCl subtypes in target parasitic species in order to lay a strong basis for the rational use of macrocyclic lactones and the discovery of novel drug targets for the development of next generation anthelmintics.

AUTHOR CONTRIBUTIONS

N.L., E.C. and C.L.C. performed the experiments and analysed the two‐electrode voltage‐clamp data; N.L. and A.C. performed molecular biology; N.L., C.L.C. and C.N. designed the study; and N.L., E.C., C.L.C. and C.N. interpreted data and drafted the manuscript.

DECLARATION OF TRANSPARENCY AND SCIENTIFIC RIGOUR

This Declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research as stated in the BJP guidelines for Design & Analysis, and as recommended by funding agencies, publishers and other organizations engaged with supporting research.

Table S1. List of primers used for PCR and cloning experiments of GluCl subunits from (Cel), Brugia malayi (Bma) and Parascaris univalens (Pun).

Click here for additional data file.

Table S2. Percentage of amino acid sequence identities between GluCl subunits from , B. malayi and P. univalens.

Click here for additional data file.

Figure S1. A single pair of chromosomes is representative of P. univalens.

Karyotype was performed on DAPI‐stained Parascaris eggs during the first mitotic division. (x400).

Click here for additional data file.

Figure S2. Amino‐acid alignments of AVR‐14B (A), GLC‐2 (B) and GLC‐3 (C) subunit sequences from the four nematode species Brugia malayi (Bma), (Cel), Haemonchus contortus (Hco) and Parascaris univalens (Pun).

Predicted signal peptides in the N‐terminal region are highlighted in grey. Amino acids share between the four species are highlight in blue. The four transmembrane segments (TM 1–4) and cys‐loops are indicated by the black bars.

Click here for additional data file.

Figure S3. Amino‐acid alignment of GLC‐4 from the four nematode species Brugia malayi (Bma), (Cel), Haemonchus contortus (Hco) and Parascaris univalens (Pun).

Predicted signal peptides in the N‐terminal region are highlight in grey. Amino acids share between the four species are highlight in blue. The four transmembrane segments (TM 1–4) and cys‐loops are indicated by the black bars.

Click here for additional data file.

Figure S4. Representative current traces after application of glutamate 1 mM on Xenopus oocytes expressing chimeric receptors made of the Pun‐GLC‐2 and Bma‐GLC‐3 subunits (n = 12).

Application times are indicated by the black bars.

Click here for additional data file.

Figure S5. Pharmacological characterization of GLC‐2/GLC‐3 from P. univalens.

a. Comparison of the macrocyclic lactones effects at 1 μM after 5 s application on Pun‐GLC‐2/GLC‐3. All responses are normalized to the maximum responses obtained with glutamate at 1 mM.

b. Representative current traces induced by ivermectin at 1 μM during 90 s on Pun‐GLC‐2/GLC‐3.

Application times are indicated by the black and purple bars.

Click here for additional data file.