A Change in the Ion Selectivity of Ligand-Gated Ion Channels Provides a Mechanism to Switch Behavior

Behavioral output of neural networks depends on a delicate balance between excitatory and inhibitory synaptic connections. However, it is not known whether network formation and stability is constrained by the sign of synaptic connections between neurons within the network. Here we show that switching the sign of a synapse within a neural circuit can reverse the behavioral output. The inhibitory tyramine-gated chloride channel, LGC-55, induces head relaxation and inhibits forward locomotion during the Caenorhabditis elegans escape response. We switched the ion selectivity of an inhibitory LGC-55 anion channel to an excitatory LGC-55 cation channel. The engineered cation channel is properly trafficked in the native neural circuit and results in behavioral responses that are opposite to those produced by activation of the LGC-55 anion channel. Our findings indicate that switches in ion selectivity of ligand-gated ion channels (LGICs) do not affect network connectivity or stability and may provide an evolutionary and a synthetic mechanism to change behavior.


Introduction
Mapping the neural connections of nervous systems is often considered to be a fundamental step in understanding behavior [1,2]. However, a neural connectivity map carries no information about the activity of neurons and the nature of the connections that each neuron makes. Neurons are embedded in neural networks, which require a delicate balance between excitation and inhibition to maintain network stability [3,4]. Homeostatic processes, conserved from invertebrates to humans, can adjust synaptic and neuronal excitability to keep neural circuits functioning within their stable dynamic range [5][6][7][8]. In these circuits, ligand-gated ion channels (LGICs) are the principal signaling components that mediate fast inhibitory and excitatory neurotransmission. The Cys-loop LGIC receptors, which include the cation-selective nicotinic acetylcholine receptors (nAChRs), serotonin type 3 receptors (5HT 3 Rs), and anion-selective GABA A and glycine receptors, form pentameric complexes in the plasma membrane [9,10]. Each individual subunit contains an extracellular N-terminal domain that harbors the ligand binding domain and four transmembrane spanning domains (M1-M4) [11]. The charge selectivity of both anion and cation-selective channels is determined by residues in the M2 domain (Fig 1A and 1B). In vitro studies have shown that LGIC channels can be switched from excitatory cation-selective to inhibitory anion-selective and vice versa through substitutions in the intracellular loop between M1 and M2 [12][13][14][15]. However, it is not known whether these channels with switched ion selectivity are functional in vivo. By switching the sign of a synapse, can the behavioral output of a neural circuit be reversed, or will a switch in the sign of a synapse cause defects in network development and stability? Neuronal specification, receptor clustering, homeostatic processes, and behavioral feedback mechanisms may preclude such manipulations.
The nematode Caenorhabditis elegans, the only animal with a completely defined neural wiring diagram [16,17], is particularly suited to addressing these questions. The neural circuit that mediates the C. elegans escape response has been well characterized. The biogenic amine tyramine coordinates backward locomotion and the suppression of head movements during the C. elegans escape response elicited by touch to the anterior half of the body [18]. C. elegans has a single pair of tyraminergic neurons, the RIMs, which activate the homomeric tyraminegated chloride channel, LGC-55 [19,20]. LGC-55 belongs to the Cys-loop LGIC family of receptors and is the only ionotropic tyramine receptor expressed in neurons and muscles that are directly postsynaptic to the tyraminergic neurons. Activation of LGC-55 induces the suppression of head movements and backward locomotion through the hyperpolarizaton of the neck muscles and premotor interneurons that drive forward locomotion. Here we changed the ion selectivity of LGC-55, from an inhibitory tyramine-gated anion channel to an excitatory tyramine-gated cation channel, and reintroduced the excitatory channel in the native circuit. We show that switching the sign of the synapse within the escape circuit does not affect circuit development or stability and results in opposite behavioral responses.
which includes an additional substitution at the 20' position of the M2 segment (R to D) ( Fig  1B). The 20' position of the M2 segment has been reported to increase the cation conductance [21,22]. To determine the ion selectivity of the engineered LGC-55 receptor, we recorded tyramine-elicited whole-cell currents in cultured muscle cells obtained from C. elegans strains that ectopically expressed either the wild type or engineered LGC-55 channel in body wall muscles. We analyzed current-voltage (I-V) relationships in varying ionic conditions: standard solution (ES1), low Na + (ES2), and low Cl -(ES3). The reversal potential (E rev ) of the wild-type LGC-55 anion channel in ES1 was -26.8 ± 3.1 mV (n = 4) near the predicted E rev for a C. elegans anion-selective channel under our conditions. A reduction of extracellular chloride concentration lead to a rightward shift of the reversal potential (E rev in ES3 = -1.9 ± 2.3 mV, n = 4), while no significant differences in the E rev values for the LGC-55 anion receptor were observed when we reduced the Na + concentration (ES2), consistent with our previous findings ( Fig 1C) [19]. The reversal potential of the engineered LGC-55 cation-II channel in standard solution was 2.4 ± 1.2 mV (n = 5), near the Goldman-Hodgkin-Katz (GHK)-predicted value for a cation-selective channel in our conditions. Reduction of the extracellular Clconcentrations did not lead to significant changes in this value (E rev in ES3 = 1.7 ± 0.9 mV, n = 4), whereas a shift to more negative potentials is observed when we decreased the extracellular Na + concentration (E rev in ES2 = -21.9 ± 2.6 mV, n = 5).
To analyze the relative Cland Na + permeabilities (P Cl /P Na ), we performed recordings using extracellular buffers containing different NaCl dilutions (1, 0.5, and 0.25 relative to the intracellular solution NaCl concentration; see Materials and Methods), and determined reversal potentials from current-voltage curves (S1 Fig). The E rev values obtained for LGC-55 anion and engineered LGC-55 cation-II receptors were plotted against extracellular Clactivity (S1C Fig), and P Cl /P Na values were obtained (see Materials and Methods [14,23]). Wild-type LGC-55 exhibited a P Cl /P Na of 18.8, further confirming that these receptors are anion selective (S1 Fig). In contrast, the P Cl /P Na value of the engineered LGC-55 cation-II channel was 0.19 (S1 Fig), indicating that the current passing through the chimeric LGC-55 cation-II channel is mainly carried by Na + and that the Cldependent component is negligible (Fig 1C). To further characterize the permeability properties of the LGC-55 cation-II channel, we analyzed E rev shifts after altering extracellular K + and Ca 2+ concentrations (S2 Fig). An increase in the external K + concentration significantly shifted the E rev towards more positive membrane potentials, whereas changes in the external Ca 2+ had no significant effects on the E rev value (S2 Fig), indicating that the engineered LGC-55 is mainly permeant to monovalent cations. Our observations are consistent with previous reports showing that similar mutations in the M1-M2 linker of GlyR dramatically increase the permeability to monovalent cations but not to calcium [14].

A Switch in Ion Selectivity Does Not Affect Synapse Formation
Can the LGC-55 cation channel assemble into a functional synapse? The tyraminergic RIM neuron make synaptic outputs onto the neck muscles and several head neurons that express LGC-55. To visualize tyraminergic synapses, we expressed the synaptic vesicle marker, mCherry::RAB-3 in the RIM neurons. Expression of mCherry::RAB-3 in the RIM neurons localized to axonal puncta along the ventral process and in the nerve ring, consistent with presynaptic specializations with the AVB premotor interneurons, the neuromuscular junction (NMJ) and head motor neurons, respectively ( Fig 3A) [16]. To examine the localization of the tyramine-gated chloride channel, we expressed a rescuing LGC-55 anion::GFP (GFP, green fluorescent protein) translational fusion under control of the lgc-55 promoter. LGC-55 anion::GFP receptors formed high-density clusters opposite presynaptic tyramine release sites in the nerve ring and the ventral process of the AVB premotor interneurons (Fig 3B). In transgenic animals that expressed LGC-55 cation-II::GFP, we observed clustering to synaptic specializations opposite tyramine release sites in the nerve ring and along the ventral process similar to animals expressing the LGC-55 anion channel ( Fig 3B). To quantify the localization of the receptor to the post-synapse, we analyzed the preand post-synaptic densities of synapses from the RIM onto the AVB (S3A Fig). Both LGC-55 anion::GFP and LGC-55 cation-II::GFP cluster in discrete regions of the ventral process of the AVB ventral process opposite the tyramine release sites (S3B- S3D Fig). However, the RIM-AVB synaptic markers were slightly more diffuse in the LGC-55 cation-II transgenic animals (S3B- S3D Fig). Synaptic markers also properly localized in tyramine-deficient, tdc-1 mutants and lgc-55 null mutants (Fig 3B and S3 Fig). The postsynaptic densities were expanded in tyramine-deficient animals, whereas presynaptic densities were enlarged in the tyramine receptor mutants. The RIM-AVB synaptic markers were slightly more diffuse in tyramine signaling mutants and the LGC-55 cation-II transgenic animals. Our data indicate that tyramine signaling and the sign of the synapse may affect the morphology of the synapse but does not change the formation of proper pre-and postsynaptic specializations.

A Switch in Ion Selectivity Reverses Behavior
To analyze the functional consequences of converting the ion selectivity of the LGC-55 channel, we compared the response of animals that expressed LGC-55 anion or LGC-55 cation under control of the native promoter to exogenous tyramine. LGC-55 is expressed in neck muscles, the RMD and SMD motor neurons that control foraging head movements, and the AVB premotor interneurons that drive forward locomotion (Fig 4). On plates containing exogenous tyramine, wild-type animals relax their neck and make long backward runs as a result of the activation of the LGC-55 anion receptor (S1 Movie). The animals eventually become immobilized in part through the subsequent activation of a tyramine G-protein coupled receptor SER-2 [19,26]. We have previously shown that the relaxation is mediated through hyperpolarization of the neck muscles and the cholinergic RMD and SMD head motor neurons that express LGC-55. Exogenous tyramine induced neck muscle relaxation and lengthening of the head in wild-type (LGC-55 anion) animals and lgc-55 null mutant animals that express a rescuing LGC-55 anion transgene (wild type: Δ head = 10 ± 6 μm, n = 68; Plgc-55:LGC-55(zfEx2): Δ head = 11 ± 4 μm, n = 75) (Fig 5A and 5B). Head movements persisted in lgc-55 mutants [19], with no significant change in head length (lgc-55(tm2913): Δ head = 3 ± 13 μm, n = 65). In contrast, transgenic animals that expressed the engineered LGC-55 cation-I or LGC-55 cation-II channel under control of the native promoter had a hypercontracted and shortened head length in response to exogenous tyramine (Plgc-55:: LGC-55 cation-I (zfEx8): Δ head = -15 ± 2 μm, n = 49; Plgc-55:: LGC-55 cation-II (zfEx40): Δ head = -28 ± 2 μm, n = 49) (Fig 5A and 5B).
doi:10.1371/journal.pbio.1002238.g005 body bends, n = 34). In sharp contrast, LGC-55 cation animals exhibit long forward runs (Plgc-55::LGC-55 cation-I (zfEx8): Δ fwd-bwd = 46.5 ± 8.4 body bends, n = 28; Plgc-55::LGC-55 cation-II (zfEx40): Δ fwd-bwd = 80.6 ± 9.5 body bends, n = 39), which continued for an extended period of time (Fig 5A, 5C, and 5D; S2 Movie). The forward runs and head contractions were more pronounced in LGC-55 cation-II than in LGC-55 cation-I transgenic animals, supporting the notion that the R to D substitution at the extracellular ring of the M2 domain increases the cation conductance (Fig 5C and 5D). However, we cannot exclude the possibility that the R to D substitution may also affect gating of the engineered LGC-55 cation-II channel. Animals that express the LGC-55 anion channel become immobilized more quickly than those expressing the LGC-55 cation channel. This suggests that the immobilization on exogenous tyramine is, in part, due to the inhibition of the forward premotor interneuron, AVB (Fig 5C).
The LGC-55 anion channel was shown to coordinate backward locomotion and the suppression of foraging head movements during the C. elegans escape response elicited by gentle anterior touch [19]. To test if the LGC-55 cation channel functions in response to endogenous tyramine release, we analyzed the escape response of transgenic LGC-55 cation animals (Fig 5). Laser ablation [27], genetic analysis [18,19], and calcium imaging experiments [28,29] support the following model for the circuit that controls the escape response (Fig 4): gentle anterior touch activates the mechanosensory ALM/AVM neurons that inhibit the PVC and AVB forward premotor interneurons and activate the AVD/AVA backward premotor neurons causing the animal to move backward. The tyraminergic motor neurons (RIM) are activated during the reversal through gap junctions with the AVA backward premotor interneurons [18]. Tyramine release promotes long backward runs and induces the suppression of head movements through activation of the LGC-55 anion channel in the AVB forward premotor interneurons and neck muscles, respectively (Fig 4) [19]. In response to touch, wild-type animals suppressed head movements by relaxing their head (Δ head = 5 ± 0.001 μm, n = 39) and reversed on average 3.14 +/-0.18 backward body bends (n = 100) (Fig 6, S4 Fig, S3 Movie). lgc-55 null mutant animals made shorter reversals than the wild type and fail to suppress the exploratory head movements during the reversal, with no significant change in head length (2.45 ± 0.15 backward body bends, n = 100). Strikingly, transgenic animals that expressed the LGC-55 cation channel variants contracted their neck muscles in response to touch (Plgc-55:: LGC-55 cation-I (zfEx8): Δ head = -11 ± 1.6 μm, n = 32; Plgc-55:: LGC-55 cation-II (zfEx40): Δ head = -14 ± 0.009 mm, n = 26), and the average reversal length was markedly reduced (Plgc-55::LGC-55 cation-I (zfEx8): 1.57 ± 0.1 body bends, n = 100; Plgc-55:: LGC-55 cation-II (zfEx40): 1.22 ± 0.1 body bends, n = 100) (Fig 6A and 6C). Furthermore, transgenic LGC-55 cation animals displayed ratchety backward locomotion, often pausing during their reversal (S4 Movie). In contrast to animals expressing the LGC-55 anion, which made long spontaneous reversals, LGC-55 cation animals predominantly make short reversals, and the number of spontaneous reversals is increased (S5 Fig). We previously proposed a model in which tyramine stimulates long reversals through the LGC-55 anion mediated hyperpolarization of the AVB forward premotor interneuron [19]. Our results strongly support this model in which the substitution of the LGC-55 anion with the LGC-55 cation induces depolarization of the AVB forward premotor interneuron during the reversal and the simultaneous activation of the forward and backward locomotion programs.
doi:10.1371/journal.pbio.1002238.g006 cation channel, LGC-55, in the postsynapstic muscle cells. Furthermore, these results indicate that the engineered LGC-55 cation channels are properly expressed and functional at the synapse within the neural circuit that modulates the C. elegans escape behavior.

Discussion
We have shown that the replacement of the M1-M2 loop of the C. elegans inhibitory tyramine receptor, LGC-55, with that of related cation channels changes the ion selectivity from anions to monovalent cations. We have demonstrated that these engineered receptors with switched ion selectivity properly localize to the synapse and are functional in vivo. Most strikingly, we show that behavioral outputs can be reversed by switching the sign of a synapse within a neural network. Mutations in the M1-M2 linker that change the ion selectivity can also lead to changes in the gating and desensitization kinetics of the channel [30]. While the LGC-55 cation channel may also exhibit kinetic differences compared to the native LGC-55 anion channel, the opposite phenotypes observed in animals that express the LGC-55 anion channel versus those that express the LGC-55-cation channel indicate that the difference in ionic selectivity is responsible for the reversal in behavioral outputs.
Previous studies in both vertebrates and invertebrates have shown that neurotransmitter release is not required for the initial development of neural circuits [31,32] and does not affect clustering of postsynaptic LGICs [33]. This indicates that changing the sign of the synapse does not affect the proper wiring of neural circuits or functional synaptic transmission. While homeostatic mechanisms that maintain the balance of excitatory and inhibitory are important for network stability, these mechanisms may only occur upon perturbation within the dynamic range of the response but not when the perturbation changes the sign of the synapse. Our results indicate that neural connectivity and the sign of synaptic connections represent independent modules of the nervous system that provide a degree of freedom in generating behavioral outputs. For example, in the developing brain, GABA's action switches from excitatory to inhibitory because of changes in the intracellular concentration of chloride [34]. Moreover, excitatory and inhibitory GABA signaling appears to coexist in the adult mammalian nervous system [35]. While in vertebrates acetylcholine (ACh) LGIC receptors are exclusively cation selective and GABA LGIC receptors are anion selective, this distinction is not as stringent in invertebrates. Molluscs have inhibitory anion-selective ACh receptors in addition to the typical excitatory cation-selective ACh LGICs [36], and C. elegans has both anion-and cation-selective ACh and GABA-gated LGICs [37][38][39]. Phylogenetic analysis of ion channel domains of the LGICs indicates that the C. elegans GABA-gated cation channels are more similar to the anionic GABA channels, and molluscan anion ACh channels are more closely related to the cationic ACh channels (Fig 7 and S6 Fig). This indicates that these nematode cationic GABA channels have evolved from their anionic ancestors through mutations in the ion selectivity domain, much like the engineered mutations causing the ionic switch in our engineered cation channel. The molluscan anionic ACh channels appear to have followed the opposite trajectory and changed the ion selectivity of their cationic ancestors [36]. Taken together with our results, this suggests that molecular changes in LGICs that result in a switch of the ions they flux provides an evolutionary mechanism to change behavior.
Our synaptic engineering of chemical synapses, together with the recent introduction of synthetic electrical synapses [41], indicates that the C. elegans connectome is remarkably stable. It will be interesting to see whether such manipulations are possible in neural circuits of other genetically tractable organisms. The engineering of ion selectivity of LGICs can be used as a general method to artificially change the sign of synapses in existing circuits. This synaptic engineering approach may have a broad range of applications in neuroscience, including LGICs.
LGIC phylogenetic comparison was performed on the ion channel domains of human and invertebrate LGICs. The neurotransmitter identities are indicated on the right. Blue shading indicates anionic channels, while red shading indicates cationic channels. Ce, C. elegans; Ls, Lymnaea stagnalis; Hs, Homo sapiens. Protein alignments were performed with ClustalW [40]. Phylogenetic analysis was performed using the neighbor joining method and midpoint rooted. Alignments and phylogenetic analyses were carried out using MacVector Software (Accelrys). GenBank accession number for the sequences used are as follows: LGC reprogramming neurotransmitter outputs and the ability to test neural circuit models, and may present a new avenue to change behavior.

Molecular Biology
Standard molecular biology techniques were used. An lgc-55 rescue construct was made by cloning an lgc-55 genomic fragment corresponding to nucleotide (nt) -2663 to +3895 relative to the translation start site into the EcoRV site in yk1072c7 [19]. To make the chimeric LGC-55 cation-I receptor, we performed DpnI site-directed mutagenesis on the lgc-55 rescuing construct using a primer that corresponded to the genomic sequence of the M1-M2 loop of the 5HT3a channel with 20 nt on either side homologous to the same region in LGC-55. The LGC-55 cation-II was made using DpnI site-directed mutagenesis with a primer that changed the codon at nts 1042-1044 relative to the translational start site, corresponding to a R to D substitution at the 20' position of the M2 loop. LGC-55 anion::GFP and LGC-55 cation-II::GFP translational fusion constructs were made by cloning GFP into an engineered AscI restriction site in the respective Plgc-55:: LGC-55 constructs in the sequence encoding the intracellular loop between TM3 and TM4. For muscle-specific expression of LGC-55 and LGC-55 cation-II, the full-length lgc-55 or lgc-55 cation-II cDNA was cloned into pPD95.86 behind the myo-3 promoter. For the identification of synapses in the RIM, we cloned a 1.1 kb cex-1 promoter fragment, which drives expression in the only RIM, upstream of the mCherry::RAB-3 fusion protein from the Gateway vector, pGH8, to produce the plasmid Pcex-1::mCherry::RAB-3. Transgenic strains were obtained by microinjection of plasmid DNA into the germline. At least DQ167344; AchE Ls, DQ167348; GABAθ Hs, NP_061028; GABAγ Hs, NP_775807; GABAα1 Hs, NP_000797; GABAρ Hs, NP_002033; GABAβ Hs, NP_000803; ACHα9 Hs, NP_060051; ACHα7 Hs, P36544; ACHδ Hs, NP_000742; ACHβ Hs, NP_000738; 5HT3A Hs, AAH04453; 5HT3B Hs, AAH46990.

Imaging
All strains were examined for colocalization of the presynaptic vesicle marker mCherry::RAB-3 in the RIM with the LGC-55 anion::GFP postsynaptic receptor using fluorescence confocal microscopy (Zeiss and Pascal imaging software). Images shown are compressed z-stacks formatted using ImageJ software.

Isolation and Culture of C. elegans Muscle Cells
Embryonic cells were isolated and cultured as described [42]. Briefly, adult animals expressing the Pmyo-3::LGC-55 anion or LGC-55 cation-II; Pmyo-3::GFP transgenes were exposed to an alkaline hypochlorite solution (0.5 M NaOH and 1% NaOCl). Eggs released were treated with 1.5 U/ml chitinase (Sigma-Aldrich, St. Louis, Missouri) for 30 to 40 min at room temperature. The embryonic cells were isolated by gently pipetting and filtered through a sterile 5 μm Durapore syringe filter (Millipore Corporation, Billerica, Massachusetts) to remove undissociated embryos and newly hatched larvae. Filtered cells were plated on glass coverslips coated with peanut lectin. Cultures were maintained at RT in a humidified incubator in L-15 medium (Hyclone, Logan, Utah) containing 10% fetal bovine serum. Complete differentiation to muscle cells was observed within 24 h. Electrophysiology experiments were performed 2 to 8 d after cell isolation. Muscle cells from transgenic animals were identified by GFP expression.
For the dilution-potential experiments, the intracellular and extracellular buffer composition were similar to those previously reported [43]. The intracellular solution for these experiments was (I2) 145 mM NaCl, 1mM CaCl 2 , 1 mM MgCl 2, 1mM EGTA and 10 mM HEPES, 10 mM glucose, pH 7.2. Control extracellular solution (1NaCl, symmetrical condition) contained 145 mM NaCl, 1mM CaCl 2, 1 mM MgCl 2 , and 10 mM HEPES (pH 7.2). The NaCl concentration were reduced to 72.5 and 36.25 mM in the extracellular buffers used in the dilution experiments (0.5 and 0.25 NaCl, respectively). Osmolarity was maintained by adding sucrose. P Cl /P Na permeability ratios were obtained by fitting shifts in the E rev to the GHK equation: E rev = (RT/F) ln {[P Na (a Na ) o + (a Cl ) i P Cl ] / [P Na (a Na ) i + (a Cl ) o P Cl ]}, where E rev is the potential where the current is zero, R is the gas constant, T is the temperature, F is the Faraday's constant, P ion is the permeability of the ion, and (a ion ) is the activity of the ion in the extracellular (subscript o) or intracellular (subscript i) solutions.
Data analyses were performed using Igor Pro software (Wavemetrics Inc, Lake Oswego, Oregon). Mean currents were fitted by a single exponential function: I (t) = I o exp (-t/τ d ) + I 1 , where I o is the current at the peak, I 1 is the current at the end of the recording, and τ d the current decay time constant. Data were normalized to I max , and the mean peak value in each condition was obtained after averaging three different traces (obtained not consecutively but in different voltage protocols in the same experiment). If the difference in current peak values was more than 80% for a given condition, the whole experiment was discarded. Reversal potential values are shown as mean ± standard error of 4-5 independent experiments for each extracellular solution. Curve fitting and statistical analysis were performed using Sigma Plot 11.0 (Systat Software.).

Behavioral Assays
All behavioral analysis was performed with young adult animals (18-24 h post L4) at room temperature (22 ºC); different genotypes were scored in parallel, with the researcher blinded to the genotype. Quantification of tyramine resistance and tyramine-induced reversals was performed as described [19]. To quantify body length on exogenous tyramine, animals were placed on agar plates supplemented with 30 mM tyramine. Still frames were taken at 5 min after exposure to tyramine, and animals were measured using ImageJ software. To quantify head length on exogenous tyramine, animals were placed on 30 mM tyramine plates. Still frames were taken at 5 min after exposure to exogenous tyramine. The neck was defined as the length from the anteriormost point of the buccal cavity to the posterior of the pharyngeal bulb (as illustrated in Fig 3B). Head contraction assays in response to touch and optogenetic activation of the RIM were performed in an unc-3(e151) mutant background. unc-3(e151) animals have normal head and neck movements but have defects in the specification ventral cord neurons that affect locomotion [43]. The unc-3(e151) genetic background was used in these assays to prevent backward locomotion in response to touch and to maintain the animal in the field of view at high magnification that would allow for accurate neck measurement. Head lengths were measured using ImageJ software. To quantify head lengths in response to touch, animals were filmed using a Sony SX910 camera and AstroII DC software for 10 s before and after a touch posterior to the pharyngeal bulb with an eyelash. Still frames were taken from the video just prior and just after the touch. Head lengths were measured from these still frames using ImageJ software. For optogentic experiments, L4 animals were transferred to assay plates that were seeded with either OP50 E. coli that was supplemented with or without all-trans retinal to a final concentration of 660 μM. Animals were raised overnight on plates with or without alltrans retinal. To quantify head lengths in response to optogenetic activation of the RIM, animals were filmed for 10 s before and after a 2-s blue light pulse. Still frames were taken from the video just prior to and during the blue light exposure.
LGC-55 is expressed in the AVB forward locomotion command neuron. In wild-type animals, spontaneous release of tyramine activates LGC-55 anion, causing a hyperpolarization of the AVB leading to a long reversal. In LGC-55 cation animals, spontaneous release of tyramine causes an activation of the AVB, leading to a shortened reversal length, and an increase in the number of short reversals made in 3 min. LGICs. Shown is the alignment of the ion pore and M2 region of invertebrate and human LGICs used in the phylogenetic analysis in Fig 6. The neurotransmitters are indicated on the right. Identities are highlighted in grey, and blue shading indicates anionic channels, while red shading indicates cationic channels. Ce: C. elegans, Ls: Lymnaea stagnalis, Hs: Homo sapiens. Protein alignments were performed with ClustalW [40] and were carried out using MacVector Software (Accelrys). See