Inhibition of cholinergic pathways in Drosophila melanogaster by α-conotoxins

Abstract

Nicotinic acetylcholine receptors (nAChRs) play a pivotal role in synaptic transmission of neuronal signaling pathways and are fundamentally involved in neuronal disorders, including Alzheimer's disease, Parkinson's disease, and schizophrenia. In vertebrates, cholinergic pathways can be selectively inhibited by α-conotoxins; we show that in the model organism Drosophila, the cholinergic component of the giant fiber system is inhibited by α-conotoxins MII, AuIB, BuIA, EI, PeIA, and ImI. The injection of 45 pmol/fly of each toxin dramatically decreases the response of the giant fiber to dorsal longitudinal muscle (GF-DLM) connection to 20 ± 13.9% for MII; 26 ± 13.7% for AuIB, 12 ± 9.9% for BuIA, 30 ± 11.3% for EI, 1 ± 1% for PeIA, and 34 ± 15.4% for ImI. Through bioassay-guided fractionation of the venom of Conus brunneus, we found BruIB, an α-conotoxin that inhibits Drosophila nicotinic receptors but not its vertebrate counterparts. GF-DLM responses decreased to 43.7 ± 8.02% on injection of 45 pmol/fly of BruIB. We manipulated the Dα7 nAChR to mimic the selectivity of its vertebrate counterpart by placing structurally guided point mutations in the conotoxin-binding site. This manipulation rendered vertebrate-like behavior in the Drosophila system, enhancing the suitability of Drosophila as an in vivo tool to carry out studies related to human neuronal diseases.—Heghinian, M. D., Mejia, M., Adams, D. J., Godenschwege, T. A., Marí, F. Inhibition of cholinergic pathways in Drosophila melanogaster by α-conotoxins.

Nicotinic acetylcholine receptors (nAChRs) are a family of neurotransmitter-gated ion channels that belong to the Cys-loop receptor superfamily (1) and are essential for synaptic transmission in the central and peripheral nervous systems. They are implicated in disorders such as Alzheimer’s disease, schizophrenia, hyperactivity, and depression disorders and are central to nicotine addiction pathways. nAChRs are pentameric, with 5 homologous subunits comprised of a ligand-binding domain (LBD), a transmembrane domain, and an intracellular region. In vertebrates, there are 12 identified neuronal nAChR subunits, α2–10 and β2–4, which can combine in different arrangements and stoichiometry depending on their functional role (1). Regardless of their composition and biologic role, all these pentameric ligand-gated channels are activated by the neurotransmitter, acetylcholine (ACh). The homomeric α7 nAChR is one of the most prevalent nAChR subtypes expressed in the nervous system and is a common target of nAChR probes (2, 3).

nAChRs are expressed in the CNS of the Drosophila melanogaster fly, a model organism widely used for genetic and neurophysiological studies, with ACh as the primary excitatory neurotransmitter (4). D. melanogaster has 10 identified nAChR subunits: Dα1–7 and Dβ1–3 (5). Dα7 subunits have the highest sequence homology to the vertebrate α7 nAChR, with 81% sequence homology in the LBD (6). Dα7 subunits form homopentamers that are essential for the function of the giant fiber system (GFS), which mediates the fly’s escape response (7,8). The GFS has 2 neuronal circuit pathways: the giant fiber (GF) to dorsal-longitudinal muscle (DLM) pathway and the GF to tergo-trochanteral muscle (TTM) pathway. Both circuits contain multiple synaptic connections with different properties, which are dependent on mixed chemical and electrical synapses (cholinergic and gap junction), ending with a glutamatergic neuromuscular junction. Only the GF-DLM pathway contains a synaptic connection that is solely dependent on the Dα7 nAChR (Figs. 1A and 2C) (9). We have previously shown with various cholinergic antagonists that the GF-DLM pathway can be specifically inhibited, whereas the GF-TTM pathway remains unaffected (10). These multifarious properties of the GFS make it a suitable in vivo model to differentiate specific and nonspecific effects of compounds on α7 nAChRs function, such as its involvement in synaptic plasticity, sensory processes, and behaviors associated with addiction (3, 11).

Figure 1.

Activity of α-conotoxin panel on TTM and DLM pathways in WT flies. A) Schematic placement of the GFS within the anatomy of D. melanogaster. Injection of α-conotoxins takes place through the ocellus while simultaneously stimulating the GF circuit (10). B) Effect of α-conotoxins MII, PeIA, AuIB, BuIA, EI, and ImI on the FF of the GF-TTM pathway of the GFS (FF measured at 50 Hz, n = 10 for each conotoxin, *P < 0.05). No significant change is detected in FF for any conotoxin tested compared with the control injection of 0.7% saline. C) Effect of MII, PeIA, AuIB, BuIA, EI, and ImI on the FF of the GF-DLM pathway of the GFS (FF measured at 50 Hz, n = 10 for each conotoxin, *P < 0.05) compared with the control injection of 0.7% saline.

Figure 2.

Biochemical and functional characterization of α-conotoxin BruIB. A) Isolation of BruIB by reversed-phase HPLC of a low-molecular-weight fraction obtained by size exclusion chromatography of C. brunneus venom. The circled peak yielded BruIB. B) Sequence of BruIB showing disulfide connectivity. O, 4-hydroxyproline. *N-terminal amidation. C) Schematic of a GFS in which BruIB was screened for activity. BruIB reduced following frequency of the GF-DLM pathway (boxed in gray). D) Responses of the GF-DLM pathway to 10 GF stimulations given at 50 Hz before and after injection with BruIB. Stars indicate no response after stimulation. E) Effect of BruIB on the FF of the GF-TTM pathway of the GFS (FF measured at 50 Hz, n = 10, *P < 0.05) where no significant change is detected in FF compared with the control injection of 0.7% saline. F) Effect of BruIB on the FF of the GF-DLM pathway of the GFS. Decrease of 100% to 43.7 ± 8.02% at 20 min after injection (FF measured at 50 Hz, n = 14, *P < 0.05) compared with the control injection of 0.7% saline.

Inhibition of cholinergic synapses can be accomplished by ligands, ranging from small molecules to larger peptidic toxins. Among such ligands are the α-conotoxins, a class of small cystine constrained peptides (12–20 residues) found in the venom of cone snails, a genus that comprises >750 species of predatory marine gastropods. α-Conotoxins are potent and selective nAChR inhibitors widely used as tools to probe nAChR structure and function (12). α-Conotoxins have a distinctive CC-C-C disulfide framework, where the number and attributes of amino acids between the second and third cysteine (loop 1) and third and fourth cysteine (loop 2) are variable and determine their nAChR subtype selectivity (12). Different Conus species produce unique α-conotoxins that can vary considerably in selectivity; the discovery of α-conotoxins with unique functional properties would advance our understanding of the roles of nAChR subtypes in synaptic transmission. Functionally, α-conotoxins bind to the interface of 2 subunits of a given nAChR subtype and hold it in a resting or closed position, thus preventing the necessary conformational changes that results in channel opening on ACh binding (13).

Exploring directly relevant cholinergic pathways in vivo is challenging because of the scope and complexity of model vertebrate organisms, along with difficulties in accessing their CNS. To overcome this problem, we decided to use the advantages of the tiny fruit fly to develop an efficacious and facile assay to test in vivo picomolar quantities of a panel of α-conotoxins that have varying selectivity toward vertebrate nAChRs. We also used this assay as a tool for bioassay-guided fractionation to isolate novel α-conotoxins. We proceeded to genetically manipulate Dα7 nAChRs by humanizing the conotoxin-binding site to mimic the functional profile of vertebrate nicotinic receptors (14). The ability to readily carry out these genetic transformations adduces transgenic fruit flies as surrogate models for defining structure-function relationships in cholinergic synaptic transmission relevant to vertebrate systems.

MATERIALS AND METHODS

Functional characterization of α-conotoxins in the GFS

α-Conotoxins were tested using the D. melanogaster GFS with the paired electrophysiology/nanoinjection bioassay as described previously (10, 15). P[Gaw- B]OK307 (Stock #6488; Bloomington Stock Center, Bloomington, IN, USA; referred to as A307 henceforth) fly stocks were kept at either 22°C or 25°C in vials containing standard media. α-Conotoxins MII, AuIB, BuIA, EI, and PeIA were synthesized as described previously (16–21). All conotoxins were resuspended in 0.7% saline for injection and initially tested at 60 pmol/fly for activity. If any changes in the responses were detected, the conotoxins were further tested at 45 and 15 pmol/fly (n = 10 for each concentration), respectively. Control flies (n = 10) were injected with a 0.7% saline solution. The GF-TTM and GF-DLM pathways were evaluated for changes in the following frequency (FF), which is the total number of responses recorded for each pathway when stimulated with 10 trains of 10 stimuli given at 50 Hz with a 1 s interval between the trains. This evaluation was performed before and 1, 5, 10, 15, and 20 min after. Statistical analysis was performed with SigmaPlot software (Systat Software, San Jose, CA, USA). All groups underwent a nonparametric Kruskal-Wallis 1-way ANOVA followed by a Tukey test.

Venom extraction, fractionation, and peptide characterization

Conus brunneus specimens (35–70 mm) were collected off the Pacific Coast of Costa Rica at depths ranging from 0 to 5 m. Venom extraction and purification were performed as described previously (22, 23). Reduction and alkylation of cysteine residues and peptide sequencing were performed as described previously (23). Positive ion matrix-assisted laser desorption ionization-time of flight mass spectrometry was carried out on an AB Voyager-DE STR spectrometer (Applied Biosystems, Foster City, CA, USA). Peptide samples were dissolved in 0.1% trifluoro acetic acid and 60% acetonitrile and applied on a cyano-4-hydroxycinnamic acid matrix. C-terminal amidation was determined by the difference between the calculated and experimental molecular weight and then confirmed by nano-NMR spectroscopy (23). Peptide concentrations were determined spectrophotometrically using a NanoDrop 2000 (Thermo Fisher Scientific, Wilmington, DE, USA).

Testing of BruIB in Xenopus laevis ooctyes

Oocyte preparation and nAChR subunit expression in X. laevis oocytes were performed as described previously (22, 24). In brief, plasmids with cDNA encoding the rat α3, α4, α10, β2, β4, α1, β1, γ, and δ nAChRs and human α7 and α9 subunits subcloned into the oocyte expression vector pNKS2 were used to prepare mRNA using the mMESSAGE mMACHINE Kit (Ambion, Carlsbad, CA, USA). All oocytes were injected with 5 ng cRNA 2–5 d before recording and kept at 18°C in ND96 buffer (96 mM NaCl, 2 mM KCl, 1 mM CaCl₂, 1 mM MgCl₂, and 5 mM HEPES, at pH 7.4) supplemented with 50 mg/L gentamycin and 5 mM pyruvic acid. Two-electrode voltage-clamp recordings from oocytes were conducted at room temperature using a GeneClamp 500B amplifier (Molecular Devices, Sunnyvale, CA, USA) at a holding potential of −80 mV. Oocytes were continuously perfused in a recording chamber with a volume of 10 μl, with ND96 solution at 1 ml/min, applied by a pump perfusion system. nAChR-mediated currents were evoked by pipetting 10 μl ACh into the bath when perfusion was temporarily paused. ACh concentration was 50 μM unless specified otherwise. Oocytes were preincubated with the peptide for 3 min, and then ACh and the peptide were coapplied. Peak ACh-evoked current amplitude was recorded before and after peptide incubation using the pClamp 9 software (Molecular Devices). The effects of BruIB on ACh-evoked currents were defined as differences in peak current amplitudes relative to the average peak current amplitude of 3–5 control ACh applications recorded before application of the peptide.

Testing of α-conotoxins in flies expressing wild-type Dα7 and D77T/L117Q/I196P Dα7

The generation of upstream activation sequence (UAS)-Dα7, UAS-Dα7-D77T/L117Q/I196P, and gfA^PΔEY6 has been described previously (8, 14). UAS constructs were expressed in gfA^PΔEY6 using the A307 Gal4 driver. Female gfA^PΔEY6_;A307 flies were crossed to UAS-Dα7, UAS-Dα7-D77T/L117Q/I196P and reared at 25°C. Male offspring was tested using the paired electrophysiology/nanoinjection bioassay described previously (10, 15). BruIB, PeIA, BuIA, and ImI were injected at 45 pmol/fly (n = 10 for each peptide), and flies were evaluated for changes in the FF of the GF-DLM and GF-TTM pathway at 50 Hz in the UAS-Dα7-D77T/L117Q/I196P line. As a positive control, UAS-Dα7 wild-type (WT) rescue flies were tested with each peptide and 0.7% saline solution was used as a negative control (n = 10) for both the WT rescue and triple mutant flies.

Homology modeling of BruIB bound to the Dα7 receptor

Homology models of BruIB complexed with native Dα7 nAChR and the D77T/L117Q/ I196P Dα7 triple mutant were generated using Modeler v9.11 (25) based on the X-ray crystal structure of ImI complexed with the Aplysia californica acetylcholine binding protein (AChBP; Protein Data Bank ID code 2C9T) (26). Verify3D (27) was used to analyze the models generated and Chimera v1.6.1 (28) was used for the molecular graphics analysis and to generate Ramachandran plots. Sequence alignments were carried out using Clustal Omega v1.2.0 (29).

RESULTS

Neuromodulation of α-conotoxins in the GFS of WT D. melanogaster

α-Conotoxins MII (20), AuIB (24), BuIA (30), EI (21), PeIA (19, 31), and ImI (16, 32, 33), which have different selectivities toward vertebrate nAChR subtypes (Table 1), were tested in the GFS of WT flies (n = 10/conotoxin). All conotoxins showed inhibition of responses during neuronal stimulation of the GF-DLM pathway. The responses were measured by testing the FF of the GFS. The FF is the number of recorded responses when the GFS is stimulated 10 times with 10 stimuli given at 50 Hz, and the average of responses in percent was determined before and after conotoxin injection. A significant inhibition (P < 0.05) of FF responses was observed between 5 and 20 min following injection compared with injection of a control 0.7% saline solution (Fig. 1C). For WT flies, the FF baseline (before injection or 0 min) response was 99.6 ± 0.2%. Twenty minutes after injection, the percent response of the GF-DLM decreased dramatically (MII, 20 ± 13.9%; AuIB, 26 ± 13.7%; BuIA, 12 ± 9.9%; EI, 30 ± 11.3%; PeIA, 1 ± 1%; ImI, 34 ± 15.4%). None of the conotoxins in the panel showed statistically significant effects on the GF-TTM pathway (Fig. 1B).

TABLE 1.

Panel of α-conotoxins tested in the GFS

α-Conotoxin	Species	Sequence	nAChR selectivity	Reference
BruIB	C. brunneus	DYCCRROTCIPIC*	Dα7	This work
ImI	C. imperialis	GCCSDPRCAWRC*	α3β2 (40.8 nM), α7 (595 nM), α3β4 (3.4 μM)	(16, 32, 33)
BuIA	C. bullatus	GCCSTPPCAVLYC*	α6/α3β2 (0.26 nM), α7 (272 nM), α3β2 (5.72 nM), α3β4 (27.7 nM), α6/α3β4 (1.54 nM), α6/α3β2β3 (0.46 nM)	(30)
AuIB	C. aulicus	GCCSYPPCFATNPDC*	α3β4 (2.5 μM)	(24)
PeIA	C. pergrandis	GCCSHPACSVNHPELC*	α9α10 (54.9 nM), α3β2 (97.5 nM), α3β4 (480 nM)	(19, 31)
MII	C. magus	GCCSNPVCHLEHSNLC*	α3β2 (3.0 nM), α6β2* (4.4 nM)	(20)
EI	C. ermineus	RDOCCYHPTCNMSNPQIC*	α1β1γδ (0.37 nM)	(21)

Includes the cone snail species venom source, the amino acid sequences, and their corresponding selectivity toward nAChRs with their corresponding IC₅₀ values in parentheses. O, 4-hydroxyproline; *, N-terminal amidation.

Discovery and functional characterization of α-conotoxin BruIB

We used the Drosophila GF assay (10, 15) to screen venom fractions from C. brunneus, an Eastern Pacific worm-hunting cone snail species. The HPLC fraction eluting at ∼40 min (Fig. 2A) showed inhibition of the GF-DLM pathway (Fig. 2D) but had no effect on the GF-TTM pathway in WT flies (Fig. 2C). After further purification, a 45 pmol/fly injection of this fraction decreased the FF of the GF-DLM pathway when stimulated at 50 Hz from 100% to 43.7% (±8.02%, P < 0.05, n = 14; Fig. 2F). This decrease in the reliability of the GF-DLM response, while maintaining the GF-TTM response (Fig. 2E), was reminiscent of the activity profile of the panel of α-conotoxins shown in Table 1 and Fig. 1. This fraction does not affect the function of gap junctions or glutamatergic synapses of the GF circuit, suggesting that it contains an α-conotoxin that targets Drosophila α7 nAChRs.

Biochemical characterization of the active venom fraction yielded a 13-residue α4/3 conotoxin, BruIB, with posttranslational modifications of hydroxylation of Pro at position 7 and N-terminal amidation (Fig. 2B). BruIB was found to be unstructured by NMR experiments (Supplemental Fig. 1). BruIB was tested at vertebrate neuronal nAChRs (α₃β₂, α₃β₄, α₄β₂, α₄β₄, α₇, α₉α₁₀, αβδε) by examining its effect on ACh-evoked currents mediated by nAChR subunit combinations expressed in X. laevis oocytes. Surprisingly, 10 μM BruIB had no effect on ACh-evoked currents for any of the vertebrate nAChR subtypes, indicating that BruIB selectively inhibits Dα7 nAChRs. BruIB also did not exhibit significant binding to Lymnaea stagnalis or A. californica AChBPs at a 10 μM concentration.

In vivo effects of BruIB on flies expressing WT Dα7 and D77T/L117Q/I196P Dα7 nAChRs

To assess the specificity of BruIB, we transgenically expressed WT Dα7 and (D77T/L117Q/I196P) Dα7 triple mutant subunits in a Dα7-null mutant (gfA^PΔEY6) background as described previously (14). These 3 amino acids were markedly different in the sequence of the conotoxin binding site of Dα7 compared with human α7 (Fig. 3); we replaced them with their human counterparts in a humanized Dα7 triple mutant. We tested the activity of BruIB, ImI, PeIA, and BuIA against FF at 50 Hz of the GF-DLM pathway in both genotypes. As expected, BruIB (45 pmol/fly) still inhibited the function of the GF-DLM pathway in null mutants (gfA^PΔEY6/>,A307/UAS-Dα7) rescued with transgenic expression of WT Dα7 (Fig. 4A). However, BruIB had no effect in null mutants expressing the Dα7 (D77T/L117Q/I196P) triple mutation (gfA^PΔEY6/>,A307/UAS-Dα7-D77T/L117Q/I196P; Fig. 4B). In comparison, ImI, PeIA, and BuIA at 45 pmol/fly were still able to disrupt the FF at 50 Hz of the GF-DLM pathway gradually over time in both genotypes (Fig. 4B). This confirmed that the Drosophila α7 nAChR is the molecular target of BruIB and that the 3 amino acids mutated (D77T/L117Q/I196P) play a critical role in BruIB binding to the Dα7 nAChR. Additionally, this demonstrated that the 3 mutated amino acids are not critical for the binding of ImI, PeIA, and BuIA, suggesting differences in the interactions of BruIB with respect to other α-conotoxins tested when binding α7 nAChRs.

Figure 3.

Sequence alignment of the Dα7 nAChR and human α7 nAChR ligand-binding regions. The upper part of the sequence depicts the 1–108 segment, while the lower part of the sequence is its continuation (108–213). Amino acids highlighted in yellow are key amino acids for conotoxin binding to the primary side of the LBD. Sequence numbering at the top corresponds to the human α7 nAChR. Residues in blue are key for conotoxin binding to complementary side of the LBD (26). Amino acids with x underneath highlighted in red are sites of mutations for mutant Dα7 studies. *Amino acid identity. ^:Amino acid homology. ^•Similar amino acids.

Figure 4.

Effect of α-conotoxins on flies expressing WT Dα7 and Dα7-D77T/L117Q/I196P in a Dα7-null mutant background. A) Activity of α-conotoxins BruIB (red), PeIA, BuIA, and ImI on the GF-DLM pathway in gfA^PΔEY6/>,A307/UAS-Dα7 flies (FF measured at 50 Hz, n = 10 for each conotoxin, *P < 0.05). B) Activity of BruIB (red), PeIA, BuIA, and ImI on the GF-DLM pathway in gfA^PΔEY6/>,A307/UAS-Dα7-D77T/L117Q/I196P flies (FF measured at 50 Hz, n = 10 for each conotoxin, *P < 0.05).

Interaction of BruIB with the WT Dα7 and triple mutant Dα7 receptor

Homology models of the complex between BruIB and Dα7 nAChR LBDs (WT and triple mutant) were generated using the crystal structure of the Ac-AChBP bound to α-conotoxin ImI as a template (Protein Data Bank ID code 2C9T) (26). These models were used to assess structural interactions between BruIB and Dα7 and BruIB and the triple mutant. BruIB–Dα7 and BruIB–triple mutant complexes (Fig. 5) showed differential ligand-receptor interactions through hydrogen bonds (purple), salt bridges (black dotted), and van der Waals interactions (orange). There are more ligand-receptor interactions in the BruIB–Dα7 complex (Fig. 5A) than in the BruIB–triple mutant complex (Fig. 5B). A single salt bridge between R5 of BruIB and D167 (complementary binding site) of the triple mutant replaces a set of stacked salt bridges found between the same amino acids of BruIB and the Dα7. Additionally, salt bridges between R6 of BruIB and D197 (principal binding site) of the C-loop are present in the BruIB–Dα7 complex but not in the BruIB–triple mutant complex. ImI inhibits WT and mutant Dα7 receptors, and the interactions seen in the crystal structure of ImI bound to Aplysia AChBP (13, 26, 34) correspond with the predicted interactions in models of ImI–Dα7 and ImI–triple mutant Dα7 (Supplemental Fig. 2). The experimental results for the inhibition of Dα7 and the triple mutant by α-conotoxins correlate excellently with the predicted interactions by homology models. This correlation validates the structural framework that we used for the Dα7 LBD.

Figure 5.

Homology models of BruIB (yellow) in complex with the (A) WT Dα7 nAChR and the (B) humanized triple mutant D77T/L117Q/I196P Dα7 nAChR. The figures depict the conotoxin binding at interface between 2 subunits. The subunits that contain the primary binding sites are shown in cornflower blue; the subunits that contain the complementary binding sites are shown in light blue; negatively charged residues are in red; positively charged residues are in dark blue; mutated residues are in green; hydrogen bonds are in purple; salt bridges are black dashed lines; and van der Waals attractions are orange lines.

DISCUSSION

Cholinergic pathways in model organisms can be effectively explored with inhibitory probes. We chose the D. melanogaster GFS to screen for neuronal probes, because its small size (1.1 mg/fly) requires only minimal quantities of material for testing, which allows us to directly test in vivo native cone snail venom fractions. Central to our assay is the Dα7 nAChR within the GFS. Our results show that α-conotoxins, which are quintessential vertebrate nAChR inhibitors, effectively target Dα7 nAChRs. This demonstrates that there is a functional overlap between the conotoxin binding site of vertebrate receptors and Dα7.

Dα7 was inhibited by all of the α-conotoxins we tested. However, these α-conotoxins exhibit differential selective inhibition profiles for vertebrate nAChR subtypes (Table 1). Surprisingly, Dα7 was inhibited by EI, an α-conotoxin selective for the αβδε nAChR at the vertebrate neuromuscular junction. The inhibition of Dα7 by these conotoxins can be rationalized using homology modeling, where in all cases strong interactions between the conotoxins with the C-loop in the principal binding were predicted (i.e., binding of ImI with Dα7; Supplemental Fig. 2). Dα7’s nonselective nature toward α-conotoxins makes it a desirable target for initially screening and identification of nAChR modulators from venom fractions or other natural extracts. Furthermore, because this receptor can be genetically manipulated, it is possible to ascertain structure-activity relationships using mutated Dα7 in transgenic flies, allowing fine-tuning of cholinergic events in neuronal transmission.

In our Drosophila-based screenings, we identified α-conotoxin BruIB, the first conotoxin to selectively target Dα7 and not affect vertebrate nAChRs. Dα7 and vertebrate α7 differ at 3 nonconserved amino acids at the conotoxin-binding site (Fig. 3). To test whether interactions with these 3 residues are responsible for Dα7–BruIB binding, we designed a triple mutant of the Dα7 receptor (D77T/L117Q/I196P) (14). This triple mutant, which resembles the human α7 receptor at the conotoxin-binding site, was not inhibited by BruIB. This confirmed that these 3 amino acids are critical for α-conotoxin binding to vertebrate nAChRs within the Drosophila model. Our findings were further supported using homology models of the complex between Dα7 and BruIB, which showed multiple interactions at the principal binding site and the complementary binding site that resulted in locking the C-loop in a resting position, thereby hindering ACh binding and hampering the conformational changes needed for the channel to open (13). In contrast, the model of the Dα7 triple mutant and BruIB shows diminished interactions between BruIB and the C-loop. Analogously, BruIB cannot interact effectively with the C-loop in vertebrate nAChRs to inhibit their function.

BruIB’s specificity for Dα7 nAChRs can be rationalized as an evolutionary adaption at the molecular level, in which inhibition depends on whether or not specific conotoxin residues interact with specific sites in the receptor. BruIB is an α4/3 conotoxin with unique traits, such as extended N terminus (DY) and an –RROT– amino acid string in loop 1. These traits contribute to BruIB binding to Dα7, because both are critical for conotoxin selectivity (12, 22). The more common –SDPR– sequence in loop 1 is critical for binding of α-conotoxins, such as ImI, to vertebrate neuronal nAChRs. Other loop 1 sequences such as –SYPP– and –SYPP–, found in BuIA and AuIB, respectively, are also selective for vertebrate neuronal nAChRs. The markedly different RROT and RROO loop 1 sequences are only found in the venom of C. brunneus and C. regius (35), respectively, which are related species belonging to subgenus Stephanoconus that inhabit the Americas and preferentially prey on polychaete worms (36). It appears that these species can express α-conotoxins that selectively target invertebrate nAChRs but not their vertebrate counterparts.

We showed that the Dα7-dependent pathway of the Drosophila escape response circuit can be inhibited by α-conotoxins, and we can functionally correlate these findings with vertebrate cholinergic pathways (14). Dα7 and its D77T/L117Q/I196P triple mutant are effective tools for in vivo screening of α-conotoxins, with the advantage of using picomolar quantities of the native natural product inhibitors. Genetic manipulation of Dα7 aimed to mimic other human nAChR subtypes yielded transgenic flies that reproduce cholinergic synapses found in higher organisms providing an efficient and low-cost in vivo model for testing novel inhibitors of specific nAChR subtypes and enhancing the suitability of Drosophila as an in vivo tool for drug discovery and the study of human neuronal diseases.

Glossary

ACh

acetylcholine

AChBP

acetylcholine binding protein

DLM

dorsal-longitudinal muscle

FF

following frequency

GF

giant fiber

GFS

giant fiber system

LBD

ligand-binding domain

nAChR

nicotinic acetylcholine receptor

TTM

tergo-trochanteral muscle

UAS

upstream activation sequence

WT

wild-type