Neurons communicate through chemical synapses and electrical synapses (gap junctions). Although these two types of synapses often coexist between neurons, little is known about whether they interact, and whether any interactions between them are important to controlling synaptic strength and circuit functions. By studying chemical and electrical synapses between premotor interneurons (AVA) and downstream motor neurons (A-MNs) in the Caenorhabditis elegans escape circuit, we found that disrupting either the chemical or electrical synapses causes defective escape response. Gap junctions between AVA and A-MNs only allow antidromic current, but, curiously, disrupting them inhibits chemical transmission. In contrast, disrupting chemical synapses has no effect on the electrical coupling. These results demonstrate that gap junctions may serve as an amplifier of chemical transmission between neurons with both electrical and chemical synapses. The use of antidromic-rectifying gap junctions to amplify chemical transmission is potentially a conserved mechanism in circuit functions.
Chemical synapses and gap junctions (also known as electrical synapses) are the structural bases for neurons to communicate and to form functional circuits. Synaptic transmission at chemical synapses occurs through the release of neurotransmitters and binding to postsynaptic receptors, whereas at electrical synapses it occurs through direct current flow. While chemical and electrical synapses often coexist between neurons, they are generally conceived as independent transmission modalities. However, emerging evidence suggests that chemical and electrical synapses may interact to control synaptic strength and circuit function (see Pereda1 for a review).
Our knowledge about interactions between chemical and electrical synapses has mostly come from studies of morphologically mixed synapses of invertebrates and lower vertebrates. At mixed synapses between auditory afferents and Mauthner cells in goldfish, excitatory chemical transmission is mediated by glutamate2 while electrical coupling results from a heterotypic gap junction that is more conductive to antidromic than orthodromic current3. It has been proposed that the antidromic junctional current may promote cooperativity among afferents3,4. At mixed synapses between interneurons and motor neurons in zebrafish, motor neurons may recruit the interneurons and regulate their glutamate release through apparently non-rectifying gap junctions5. In Drosophila, mixed synapses between the giant fibre and tergotrochanteral motor neurons use acetylcholine for chemical transmission6 and use heterotypic gap junctions formed by two different isoforms of the innexin Shaking-B for electrical coupling7. Analyses with the Xenopus oocyte heterologous expression system suggest that the gap junctions between giant fibre and tergotrochanteral motor neurons favour orthodromic current flow7. Mixed synapses are also abundant in mammalian brains8,9,10,11,12, but very little is known about their functional properties and potential interactions.
The rather limited amount of knowledge of interactions between chemical and electrical synapses is mainly due to technical difficulties with performing paired current- and voltage-clamp recordings from neurons in situ, and a lack of genetic means to disrupt either the electrical or chemical component specifically. In Caenorhabditis elegans escape circuit, a pair of command interneurons (AVA) contact downstream A-type cholinergic motor neurons (A-MNs) through both chemical synapses and gap junctions. Although these chemical and electrical synapses are not closely associated in space (WormWiring, http://wormwiring.org/), they may functionally interact because neurons in C. elegans are essentially isopotential due to a very high membrane resistance, and changes in membrane voltage are instantaneously experienced by the entire neuron13,14. In this study, we took advantage of our recent success in performing paired voltage- and current-clamp recordings with C. elegans neurons and the high genetic amenability of the worm to investigate how electrical and chemical synapses interact to control synaptic transmission and escape behaviour. We found that chemical transmission from AVA to A-MNs is mediated by acetylcholine and an LGC-46 postsynaptic receptor, and that gap junctions between AVA and A-MNs only allow antidromic current. Importantly, we found that the rectifying gap junctions serve to amplify the excitatory chemical transmission from AVA to A-MNs and are critical to the escape behaviour. These findings show for the first time how chemical synapses and gap junctions may interplay to control a specific behaviour. The mechanisms uncovered here are potentially relevant to some morphologically or functionally mixed synapses in other systems, including mammals.
Interneurons control A-MNs by producing PSC bursts
The C. elegans locomotion neural circuit consists of 5 pairs of command interneurons and 58 ventral cord motor neurons15. Three pairs of the command interneurons (AVA, AVD and AVE) form synapses with A-type cholinergic MNs (A-MNs) mediating backward locomotion whereas the remaining two pairs (AVB and PVC) form synapses with B-type cholinergic MNs (B-MNs) mediating forward locomotion. Among the command interneurons in the backward neural circuit, AVA contacts A-MNs through both chemical synapses and gap junctions whereas AVD and AVE only through chemical synapses (Fig. 1a). To understand how command interneurons control A-MNs through chemical synapses, we started by analysing spontaneous postsynaptic currents (PSCs) recorded from VA5, which was chosen as a representative of A-MNs because it could be more easily identified based on its anatomical location than other A-MNs (Fig. 1b). We observed spontaneous PSCs occurring both sporadically and in bursts (Fig. 1c). In a histogram of decay time distribution, spontaneous PSCs may be divided into two distinct populations with a huge difference in the mean decay time (0.92±0.03 ms versus 59.25±3.43 ms) (Fig. 1d). Spontaneous PSCs with a slow decay time (sPSCs) were also much larger in mean peak amplitude than those with a fast decay time (fPSCs) (Fig. 1d).
We defined PSC bursts as an apparent increase in PSC frequency accompanied by a persistent current lasting for ≥3 s because we previously found that C. elegans motor neurons control body-wall muscle cells by producing PSC bursts, and that a burst duration of 3 s is close to the minimum to cause detectable changes in muscle activity16. The duration of PSC bursts in VA5 ranged from 3 to 35 s with the majority of them (∼80%) lasting 5–15 s (Fig. 1e). The mean charge transfer rate during the burst was 6-fold higher than that outside the burst (Fig. 1e). Compared with extra-burst events, intra-burst PSCs showed significant increases in the frequency, mean amplitude and charge transfer but no change in the mean half width, 10–90% rise time, and decay time (Fig. 1f). These differences between extra- and intra-burst events are similar to those of spontaneous PSCs recorded from C. elegans body-wall muscle cells16.
We wondered whether the PSC bursts observed in VA5 are important to the control of A-MNs by presynaptic neurons, as are those produced by MNs to the control of body-wall muscles16. To answer this question, we analysed temporal relationships between PSC bursts in VA5 and either PSC bursts or Ca2+ transients in adjacent body-wall muscle cells, which are synchronous in activity17. Both PSC bursts and Ca2+ transients in muscle cells are temporally correlated with PSC bursts in VA5 (Fig. 2 and Supplementary Movie 1). These observations suggest that command interneurons control A-MNs by causing PSC bursts.
AVA interneurons control A-MNs by acetylcholine
To identify the neurotransmitter(s) used by the command interneurons in the backward circuit, we began by testing the effects of various exogenous neurotransmitters on VA5 whole-cell current. In each experiment, a candidate exogenous neurotransmitter was applied to VA5 by pressure-ejection through a glass pipette. An inward current was expected upon the activation of either a non-selective cation channel or a Cl− channel because their equilibrium potentials were more depolarized than the holding potential (−60 mV) under our experimental conditions. We tested the effects of nine different candidate neurotransmitters, including acetylcholine, γ-aminobutyric acid (GABA), glutamate, aspartate, ATP, dopamine, octopamine, serotonin and tyramine. Only acetylcholine and GABA were able to induce a current (Supplementary Fig. 1a), suggesting the existence of both acetylcholine and GABA receptors in VA5. Nicotine and levamisole, which are agonists for some C. elegans acetylcholine receptors (AChRs)18,19,20,21, were without an effect (Supplementary Fig. 1b). Choline, the precursor of acetylcholine, also showed no effect (Supplementary Fig. 1b).
Because chemical synaptic transmission from command interneurons to A-MNs presumably plays an excitatory role15,22 but GABA generally serves as an inhibitory neurotransmitter, we tested whether acetylcholine is a neurotransmitter used by the command interneurons. Indeed, d-tubocurarine, a competitive nicotinic AChR antagonist23, abolished both spontaneous PSCs and exogenous acetylcholine-induced current in VA5 (Supplementary Fig. 2a). Consistently, we found that CHA-1 (choline acetyltransferase) and UNC-17 (vesicular acetylcholine transporter), which are required for synthesizing acetylcholine and loading synaptic vesicles with acetylcholine, respectively24, are important to spontaneous PSCs in VA5. In the hypomorphic mutants cha-1(p1152) and unc-17(e245) (nulls are nonviable)24, the frequencies of sPSCs and fPSCs were greatly decreased (by 80–90%); and the remaining events of sPSCs but not fPSCs were also much smaller in mean amplitude and charge transfer compared with wild type (Supplementary Fig. 2b). These observations suggest that spontaneous PSCs in VA5 result from acetylcholine release.
To determine whether command interneurons are a source of acetylcholine acting on A-MNs, we assessed the effect of d-tubocurarine on optogenetically evoked PSCs in VA5 using a strain expressing an integrated channelrhodopsin-2 (ChR2) transgene in command interneurons16. Each photostimulus caused a PSC burst associated with a large initial transient, which could be eliminated by d-tubocurarine (Fig. 3a). This observation suggests that command interneurons use acetylcholine to control A-MNs. Because AVA forms more chemical synapses with A-MNs than AVD and AVE25,26 and is more important than the other two in backward locomotion22,27,28,29, we tested whether AVA is a source of the acetylcholine. We first checked whether AVA expresses UNC-17, which is a marker of cholinergic neurons, by creating a transgenic strain in which GFP was expressed in cholinergic neurons under the control of Punc-17 whereas mStrawberry specifically in the two AVA command interneurons using a Cre-LoxP recombination approach involving two promoters (Pflp-18 and Pgpa-14)30. In transgenic worms, two neurons in the head region were colabelled by GFP and mStrawberry (Fig. 3b), suggesting that AVA is cholinergic. We then determined whether spontaneous PSCs in VA5 depend on acetylcholine release from AVA, and found that knockdown of unc-17 specifically in AVA greatly inhibited PSC bursts and sPSCs but had no effect on fPSCs (Fig. 3c,d). The inhibition of sPSCs but not fPSCs in the AVA unc-17 knockdown strain suggests that the RNA interference did not spread to A-MNs, which are the source of acetylcholine for fPSCs as described later. Finally, we determined whether suppressing AVA activity would inhibit spontaneous PSCs in VA5 using a strain expressing the histamine-activated Cl− channel HisCl1 specifically in AVA31. We found that histamine administration abolished PSC bursts and nearly eliminated sPSCs but did not affect fPSCs (Fig. 3e). Collectively, these observations suggest that PSC bursts and sPSCs in A-MNs arise from acetylcholine release from AVA, whereas fPSCs in A-MNs are due to acetylcholine release from a different source.
LGC-46 AChR mediates AVA to A-MN transmission
To identify the receptor(s) mediating AVA to A-MN transmission, we began by analysing spontaneous PSCs in VA5 in mutants of selected genes of the Cys-loop superfamily of ligand-gated ion channels32. acr-5, acr-8, acr-14, acr-15, acr-18, acr-21, lgc-46 and lgc-55 were chosen for the analysis because they are known to be expressed in at least some of the ventral cord motor neurons (www.wormbase.org) but are not components of a previously identified motor neurons extrasynaptic receptor20,33. We found that mutations of lgc-46 (Fig. 4a) but not the other genes (Supplementary Fig. 3) caused a great reduction (∼70%) in the frequency of sPSCs with the remaining events having a smaller mean amplitude and less charge transfer compared with wild type. Knockdown of lgc-46 specifically in A-MNs also inhibited sPSCs (Fig. 4a). In contrast, fPSCs were unchanged in the mutants compared with wild type (Fig. 4a). The deficiencies of sPSCs in lgc-46 mutants could be rescued by expressing wild-type LGC-46 specifically in A-MNs (Fig. 4a). These observations suggest that LGC-46 is a key component of the AChR mediating sPSCs in A-MNs.
Although expression of lgc-46 (also known as Y71D11A.5) in ventral cord MNs has been reported34, A-MNs were not among those identified in the previous study. To confirm that lgc-46 is expressed in A-MNs, we created a transgenic strain coexpressing GFP and mStrawberry under the control of Plgc-46 and the A-MN-specific promoter Punc-4, respectively. In the transgenic worms, GFP expression was observed in ventral cord MNs, many head neurons, and some neurons in the tail (Fig. 4b). All mStrawberry-positive neurons were colabelled by GFP, suggesting that lgc-46 is expressed in A-MNs. In addition, GFP was observed in some ventral cord motor neurons not labelled by mStrawberry, which could be GABAergic MNs that reportedly express lgc-46 (ref. 34).
To confirm that LGC-46 may form an ionotropic AChR, we expressed LGC-46 in Xenopus oocytes and measured whole-cell current using the two-microelectrode voltage clamp technique. Acetylcholine caused inward current in a concentration-dependent manner with the half maximal effective concentration (EC50) being 50.6±2.6 μM (Fig. 4c). In contrast, none of the other tested agents, including levamisole, nicotine, choline, and glycine, showed any effect on the whole-cell current (Fig. 4d). These observations suggest that LGC-46 may form an ionotropic receptor that is specifically activated by acetylcholine.
ACR-14 autoreceptor in A-MNs facilitates transmitter release
To identify the AChR mediating fPSCs in VA5, we scrutinized recording traces of spontaneous PSCs from the analysed mutants. We found that acr-14 is important to fPSCs. In the loss-of-function mutant acr-14(ok1155)34, fPSCs were decreased by ∼80% in frequency with the remaining events having a smaller mean amplitude and less charge transfer compared with wild type (Fig. 5a). In contrast, sPSCs were normal in the mutant (Fig. 5a). The deficiencies of fPSCs could be rescued by expressing wild-type ACR-14 specifically in A-MNs (Fig. 5a). These observations suggest that ACR-14 is a key component of the AChR mediating fPSCs.
We then tried to identify the source of acetylcholine causing fPSCs in VA5. A previous study showed that GABA miniature PSCs (minis) in interneurons of the molecular layer of rat cerebellum include large-amplitude ‘ordinary minis’ and small-amplitude ‘preminis’ (presynaptic miniature currents); and that the latter result from the activation of autoreceptors35. We wondered whether fPSCs in VA5 result from a similar mechanism. To explore this possibility, we first examined the effect of unc-17 knockdown specifically in A-MNs, and observed a significant reduction in the frequency of fPSCs but not sPSCs (Fig. 5a), suggesting that fPSCs depend on acetylcholine release from A-MNs. The inhibition of fPSCs but not sPSCs in the A-MN-specific unc-17 knockdown strain suggests that the RNA interference did not spread to AVA, which is the source of acetylcholine for sPSCs as shown earlier (see Fig. 3). We then tested whether inhibiting neurotransmitter release from the same neuron (VA5) may inhibit fPSCs by including 5 mM BAPTA in the pipette solution used for whole-cell recording. BAPTA is a membrane impermeant fast Ca2+ chelator36, and is expected to inhibit neurotransmitter release from VA5 through depleting free Ca2+ in the cytosol. A great decrease in fPSC frequency was observed in the BAPTA-treated VA5 (Fig. 5a). Neither the A-MN-specific unc-17 knockdown nor the BAPTA treatment affected sPSCs (Fig. 5a). These observations suggest that fPSCs in A-MNs likely result from the activation of an autoreceptor containing ACR-14.
To confirm that acr-14 is expressed in A-MNs, We coexpressed GFP and mStrawberry under the control of Pacr-14 and the A-MN-specific promoter Punc-4 (ref. 37), respectively. In the transgenic worms, GFP was observed in all mStrawberry-labelled neurons plus some other neurons along the ventral nerve cord (Fig. 5b), suggesting that acr-14 is expressed in A-MNs as well as some other MNs, which could be D-MNs and B- and AS-type MNs reported in previous studies34,38,39. In addition, GFP expression was observed in several head and tail neurons (Fig. 5b).
What could be the role of the ACR-14 autoreceptor in motor neuron function? To answer this question, we compared the frequency and mean amplitude of spontaneous PSCs as well as the amplitude of evoked PSCs at the neuromuscular junction between wild type and acr-14(ok1155). The mutant showed ∼30% decrease in both the amplitude of evoked PSCs and the frequency of spontaneous PSCs compared with wild type without a change in the mean amplitude of spontaneous PSCs (Fig. 5c). These observations suggest that a physiological role of the ACR-14 autoreceptor is to facilitate spontaneous and evoked neurotransmitter release.
Gap junctions between AVA and A-MN are antidromic rectifiers
We analysed properties of the gap junctions between AVA interneurons and A-MNs by performing paired voltage- and current-clamp recordings with VA5 and an AVA (Fig. 6a). We found that gap junctions between them are strongly rectifying in the AVA direction. Significant junctional current (Ij) is observed only when the membrane voltage (Vm) of VA5 is more depolarized than that of AVA (Fig. 6b), indicating that Ij flows only in the AVA direction. Consistently, AVA showed large Vm changes in response to current injections into VA5 whereas VA5 showed little Vm change in response to current injections into AVA, as quantified by the coupling coefficient (CC) (Fig. 6c). Cell-specific rescue experiments for mutant locomotion defects have led to the suggestion that the gap junctions between AVA and A-MNs are heterotypic gap junctions formed by UNC-7 in AVA and UNC-9 in A-MNs22,40 but this conclusion remains to be substantiated by electrophysiological evidence. We found that, in both unc-7 and unc-9 mutants, Ij between AVA and VA5 was essentially absent, and CC was very small even when current was injected into VA5 (Fig. 6b,c), which confirmed the importance of UNC-7 and UNC-9 in the electrical coupling.
We also analysed electrical coupling between AVA and VA8, which is another A-MN innervating ventral muscles but is located far away from VA5 (Fig. 1b). VA8 was chosen because it was analysed in a previous study of electrical and chemical transmission between AVA and A-MNs22. We found that Ij is similarly rectifying in the AVA direction (Supplementary Fig. 4). The observations of rectifying gap junctions between AVA and two A-MNs from different locations suggest that gap junctions between AVA and A-MNs are likely antidromic rectifiers.
Electrical coupling amplifies chemical transmission
To determine how gap junctions might interact with chemical synapses between AVA and A-MNs, we compared spontaneous PSCs in VA5 between wild type and two strains with either unc-7 knockdown specifically in AVA or unc-9 knockdown specifically in A-MNs. Both knockdown strains showed little electrical coupling between AVA and VA5 (Fig. 7a,b). Remarkably, the frequency of PSC bursts was greatly reduced in the two knockdown strains compared with wild type (Fig. 7c), suggesting that the rectifying gap junctions normally serve as an amplifier of chemical transmission from AVA to A-MNs.
To test whether the observed inhibition of chemical synaptic transmission in the unc-7 knockdown strain resulted from a secondary developmental defect, we examined the effects of conditional UNC-7 depletion on synaptic transmission between AVA and A-MNs using an auxin-inducible degradation approach41. We created a strain in which UNC-7 is tagged with the degron at the C-terminus by Crispr/Cas9-based genome editing42 and the TIR1 F-box protein is expressed specifically in AVA from an extrachromosome array. TIR1 interacts with a number of endogenous proteins to form a functional E3 ubiquitin ligase and, in the presence of auxin, binds the degron-tagged protein and triggers its degradation. Auxin administration to this transgenic strain abolished Ij between AVA and VA5, and greatly inhibited PSC bursts in VA5 (Supplementary Fig. 5), which are similar to the effects of AVA-specific unc-7 knockdown. These observations reinforce the notion that the antidromic Ij between AVA and A-MNs facilitates chemical synaptic transmission.
We also analysed the effects of AVA-specific unc-7 knockdown on electrical coupling between AVA and DA4 (a representative A-MN innervating dorsal muscles), and on PSC bursts in DA4. We found that Ij is abolished, and PSC bursts are greatly inhibited in the knockdown strain (Supplementary Fig. 6), suggesting that gap junctions between AVA and A-MNs innervating dorsal muscles play a similar regulatory role in chemical transmission.
We then examined whether chemical transmission may regulate electrical coupling between AVA and A-MNs. To this end, we analysed Ij current between AVA and VA5 in the AVA-specific unc-17 knockdown strain described earlier (Fig. 3). We found that Ij was indistinguishable between wild type and the knockdown strain (Supplementary Fig. 7), suggesting that a deficiency of chemical transmission between AVA and A-MNs does not compromise electrical coupling.
Given that gap junctions and chemical synapses between AVA and VA5 are not physically close, the observed functional interactions between them suggest that these neurons likely have very high membrane resistance. We therefore measured input resistance, which almost completely results from the membrane resistance, in AVA and VA5 at holding voltages close to their resting membrane potentials (AVA −25 mV; VA5 −70 mV)43,44. The measured input resistance was 2.50±0.19 GΩ for AVA (n=9) and 3.47±0.18 GΩ for VA5 (n=22). These unusually high input resistance values suggest that a small Ij could have a major impact on the membrane voltage in these neurons, and that a change in membrane voltage may instantaneously spread to the entire neuron.
Escape depends on both chemical and electrical synapses
The identification of the key molecules for chemical and electrical transmission between AVA and A-MNs made it possible to determine the roles of these two synapses in C. elegans escape response. We specifically disrupted either the chemical or the electrical transmission, and analysed their effects on an escape response to a mechanical touch on the head. In response to the mechanical touch, worms move backward to escape. We counted the number of dorsoventral tail swings of each worm in response to the nose touch and used it as a measure of the backward locomotion. Compared with wild type, backward locomotion is severely compromised in the AVA-specific unc-17 knockdown strain, the lgc-46 mutants, and the A-MN-specific lgc-46 knockdown strain; and the lgc-46 mutant locomotion defect can be rescued completely by expressing wild-type LGC-46 under the control of its native promoter (Fig. 8a; Supplementary Movies 2–8), suggesting that chemical synaptic transmission between AVA and A-MNs is important to backward locomotion. Backward locomotion was also severely defective in the AVA-specific unc-7 knockdown and the A-MN-specific unc-9 knockdown strains (Fig. 8a; Supplementary Movies 9 and 10), which were used in our electrophysiological analyses (Fig. 7). Since ectopic gap junctions between AVA and B-MNs have been observed in unc-7 mutant45, we tested whether such ectopic gap junctions also occur in the AVA-specific unc-7 knockdown strain by analysing the electrical coupling between AVA and VB6, a representative of B-MNs. We did not observe any Ij in either wild type or the unc-7 knockdown strain (Supplementary Fig. 8). Taken together, we have demonstrated at both the synaptic and behavioural levels that gap junctions between AVA and A-MNs serve to amplify chemical synaptic transmission from AVA to A-MNs, as illustrated in the model diagram (Fig. 8b).
Our work with C. elegans shows that antidromic rectifying gap junctions can facilitate orthodromic chemical transmission between neurons with mixed transmission modalities, and that this functional interaction is important in controlling a specific behaviour. Our study also shows that motor neurons are not simply passive responders to commands but may modulate the activities of higher order neurons retrogradely.
Gap junctions are formed by connexins in vertebrates and innexins in invertebrates. Each gap junction consists of two hemichannels that are either identical or different in composition. Gap junctions of identical hemichannels are known as homotypic gap junctions while those of different hemichannels are known as heterotypic gap junctions. Among gap junctions in native tissues studied to date, the majority allow current flow in both directions equally, while only a small number pass current preferentially in one direction. Rectification is generally a property of heterotypic gap junctions formed by either two different gap junction proteins (connexins or innexins) or two different isoforms of one gap junction protein3,7,46. In this study, we showed that gap junctions between AVA and A-MNs allow current flow only in one direction, which is probably the strongest degree of rectification ever observed with gap junctions in native tissues. Our analyses suggest that this antidromically rectifying junctional current serves to amplify chemical synaptic transmission.
Among a total of 25 innexin genes in C. elegans, only mutants of unc-7 and unc-9 are associated with obvious locomotion defects. Cell-specific rescue experiments of the locomotion defects have led to the suggestion that the gap junctions between AVA and A-MNs are likely heterotypic gap junctions formed by UNC-7 in AVA and UNC-9 in A-MNs22,40. Our electrophysiological analyses confirmed this prediction. An earlier study found that AVA is hyperactive in unc-7 mutant, and that rescuing electrical coupling between AVA and A-MNs in unc-7 mutant facilitates forward instead of backward locomotion. Based on these observations, it was concluded that the electrical coupling between AVA and A-MNs normally inhibits backward locomotion by shunting excitatory current away from AVA into A-MNs22. In contrast, our results indicate that the gap junctions between AVA and A-MNs only allow current flow from A-MNs into AVA, and that a deficiency of the electrical coupling compromises rather than facilitates backward locomotion. Because ectopic gap junctions between AVA and B-MNs have been observed in unc-7 mutant45, and because B-MNs are important to forward locomotion15, it is unclear whether the observations of the previous study22 have been complicated by the ectopic gap junctions. Alternatively, other differences in experimental approaches could have contributed to the different conclusions. In the present study, we showed that AVA-specific removal of UNC-7 in adult worms produced similar results as AVA-specific unc-7 RNAi. Furthermore, we did not observe any electrical coupling between AVA and a representative B-MN in the unc-7 RNAi strain. Therefore, the effects of electrical coupling on chemical transmission and locomotion behaviour observed in the unc-7 RNAi strain are unlikely to be caused by either ectopic gap junctions or developmental defects.
The identity of neurotransmitter used by AVA has been uncertain. Studies based on either expressing GFP under the control of specific promoters or immunostaining against specific neuron markers have led to controversial conclusions. One study suggests that the neurotransmitter is acetylcholine47 whereas another suggests that it is glutamate48. The possibility of acetylcholine being the neurotransmitter is also disputed by results of two other studies33,49. Although the approaches used in the previous studies are generally effective in reporting gene expression patterns, there are instances where immunoreactivity is undetected because of either a low protein expression level or a low affinity of the antibody, and where the GFP reporter cannot accurately recapitulate the native expression pattern of a gene because of the absence of either introns or distant regulatory elements important to gene expression in the gfp transgene50,51,52. The present study confirms that acetylcholine is the neurotransmitter used by AVA. In addition, we identified LGC-46 as a key subunit of the postsynaptic AChR mediating synaptic transmission from AVA. The identification of both the neurotransmitter and a key component of the postsynaptic receptor for chemical transmission from AVA to A-MNs may serve as an important foundation for future studies on C. elegans locomotion neural circuit, which is commonly used as a model to understand how a small number of neurons may produce complex locomotion behaviours.
C. elegans neurons do not fire all-or-none overshooting action potentials13. It is generally thought that C. elegans neurons perform their physiological functions by producing graded potentials but it is not totally clear how graded potentials of presynaptic neurons may alter the activity of postsynaptic neurons. Our earlier study shows that C. elegans motor neurons control body-wall muscle cells by producing PSC bursts16. It is difficult to generalize this finding to other synapses without experimental evidence because body-wall muscle cells differ from neurons in that they fire all-or-none overshooting action potentials53,54. The findings that both AVA command interneurons and motor neurons control postsynaptic cells by producing PSC bursts suggest that PSC bursts might be generally used for signal transduction at chemical synapses in C. elegans.
ACR-14 has not been implicated in the function of A-MNs although it has been speculated as an autoreceptor39. Our results suggest that ACR-14 in A-MNs is indeed an autoreceptor. The fact that buffering intracellular Ca2+ by including BAPTA in the pipette solution eliminated the majority of fPSCs suggests that the ACR-14 autoreceptor is activated by acetylcholine release from the same neuron, and is likely located very close to the neurotransmitter release site. The inhibitory effects of acr-14 mutation on spontaneous and evoked PSCs at the neuromuscular junction suggest that a physiological role of the ACR-14 autoreceptor in A-MNs is to facilitate neurotransmitter release. This role of ACR-14 autoreceptor in C. elegans cholinergic motor neurons is reminiscent of the functions of presynaptic AChRs in rat brain55 and cultured Xenopus motor neurons56.
Neuronal gap junctions are mainly known for their roles in synchronizing cellular activities57,58,59. The existence of morphologically or functionally mixed synapses in escape neural circuits of various species2,15,60 suggests that potential interactions between electrical and chemical synapses may play major roles in escape responses. This study shows that gap junctions allowing only antidromic current act to amplify chemical transmission between neurons with functionally mixed synapses. This finding resembles that of a recent study, which shows that electrical coupling may regulate chemical synaptic transmission at mixed synapses between premotor interneurons and motor neurons in zebrafish5. This similarity between two evolutionarily distant species suggests that amplification of chemical synapses by electrical synapses is likely a conserved mechanism in circuit functions. Why would nature design such a mechanism to amplify chemical transmission? Perhaps this mechanism is employed to sustain synaptic strength between two neurons without requiring more excitatory inputs into the presynaptic neuron, which is potentially beneficial to the function of diverse neural circuits. For example, a brief noxious stimulus could cause a much longer escape response because of the amplifying effect of chemical transmission by electrical coupling. In the mammalian hippocampus, where mixed synapses exist9,12,61, enhancement of chemical transmission by electrical coupling at mixed synapses could be a basic and necessary mechanism for learning and memory. Whether mixed synapses do play such physiological roles is an interesting question to be answered by future studies.
C. elegans culture and strains
All worms were raised on agar plates with a layer of OP50 Escherichia coli at 21 °C inside an environmental chamber. The worm strains used in this study are listed in Table 1.
Gene expression pattern analyses
To confirm that lgc-46 and acr-14 are expressed in A-MNs, we coexpressed mStrawberry under the control of Punc-4 (A-MN-specific) and GFP under the control of either Plgc-46 or Pacr-14. To confirm that AVA interneurons are cholinergic neurons, we first created an integrated transgenic strain expressing GFP under the control of Punc-17 (ZW798), and then coinjected two plasmids, pNP259 (Pgpa-14::Cre) and wp1392 (Pflp-18::loxP::LacZ::STOP::loxP::mStrawberry) into the ZW798 strain. The use of Pgpa-14 and Pflp-18 in Cre-loxP recombination results in AVA-specific gene expression30. The expression patterns of GFP and mStrawberry were examined and imaged with an inverted microscope (TE-2000U, Nikon) equipped with specific fluorescence filters (HQ Texas Red/41004 and HQ FITC/41001, Chroma Technology Corp., Bellows Falls, VT, USA) and a CCD camera (F-View II, Olympus). Pflp-18 and Pgpa-14 were gifts from Alexander Gottschalk lab30. The other promoters, including Punc-4 (2.9 kb), Pacr-14 (3.8 kb), Plgc-46 (3.0 kb) and Punc-17(4.1 kb), were cloned using genomic DNA of the Bristol N2 strain as the template and the primers listed in Table 2.
lgc-46 and acr-14 mutants were rescued by expressing wild-type LGC-46 and ACR-14 under the control of either the native promoter or the A-MN-specific Punc-4. Full-length lgc-46 (Y71D11A.5) cDNA and acr-14 (T05C12.2) cDNA were cloned from a Bristol N2 cDNA library using the primers listed in Table 2.
Neuron-specific gene knockdown was achieved by coexpressing two plasmids encoding sense and corresponding antisense RNA fragment of a gene under the control of a specific promoter62. A-motor neuron-specific gene knockdown was achieved by using Punc-4, whereas AVA-specific gene knockdown through a Cre-LoxP approach using Pflp-18 and Pgpa-14 (ref. 30). The primers used for cloning the cDNA fragments of lgc-46 (585 bp), unc-17 (473 bp), unc-7 (485 bp) and unc-9 (397 bp) are listed in Table 2. One transgenic line of each kind was randomly chosen for electrophysiological and behavioural analyses.
Auxin-induced UNC-7 degradation
A DNA fragment encoding codon-optimized degron41 was synthesized (GenScript, Piscataway, NJ, USA) and inserted into wild-type genome using the Crispr/Cas9 approach42 to tag UNC-7 with degron at the C-terminus. The guide RNA sequence (5′-GGATGCGGAACACGGTCAA) targeting the last exon of unc-7 was inserted into pDD162 (Peft-3::Cas9+Empty sgRNA) (Addgene #47549) by site-directed mutagenesis. The resulting plasmid (wp1685) was coinjected with the transgenic marker Pmyo-2::mStrawberry (wp1613) into wild-type worms. Genome-edited worms were identified through PCR screen and DNA sequencing. The Cre-LoxP system was used to achieve AVA-specific degradation of UNC-7. The TIR1::mRuby sequence was amplified from pLZ31 (Peft-3::TIR1::mRuby::unc-54 3′UTR, pCFJ151) (Addgene #71720) and cloned into Pflp-18::loxP::LacZ::STOP::loxP::mStrawberry (wp1392) to generate Pflp-18::loxP::LacZ::STOP::loxP::TIR1::mRuby (wp1693). This plasmid was coinjected with Pgpa-14::Cre (pNP259) and Pflp-18::loxP::LacZ::STOP::loxP::mCherry::SL2::GFP (wp1383) into the UNC-7 degron strain. For auxin-induced UNC-7 degradation, L4 transgenic worms were transferred to NGM plates containing 1 mM auxin, and the adult worms were used for electrophysiological recording on the following day.
C. elegans electrophysiology
All electrophysiological experiments were performed with adult hermaphrodites. An animal was immobilized on a glass coverslip by applying Vetbond Tissue Adhesive (3M Company, St Paul, MN). Application of the glue was generally restricted to the dorsal anterior portion of the animal, allowing the tail to sway freely during the experiment. A short (∼300 μm) longitudinal incision was made along the glued region. After clearing the viscera by suction through a glass pipette, the cuticle flap was folded back and glued to the coverslip, exposing several ventral body-wall muscle cells, a small number of motor neurons anterior to the vulva, and some neurons in the head. The dissected worm preparation was treated with collagenase A (Roche Applied Science, catalogue number 10103578001, 0.5 mg ml−1) for 10–15 s and perfused with the extracellular solution for 5 to 10-fold of bath volume. Borosilicate glass pipettes were used as electrodes for voltage- and current-clamp recordings. Pipette tip resistance for recording from neurons was ∼20 MΩ whereas that for recording from body-wall muscle cells was 3–5 MΩ. Classical whole-cell configuration was obtained by applying a negative pressure to the recording pipette. Motor neurons were identified based on their anatomical locations whereas AVA based on GFP fluorescence from the expression of an integrated transgene, which was created using two promoters and the Cre-LoxP approach as described above, and crossed into various mutants. Spontaneous and evoked PSCs at the neuromuscular junction were recorded and analysed as previously described63. Voltage- and current-clamp experiments were performed with a Multiclamp 700B amplifier (Molecular Devices, Sunnyvale, CA, USA) and the Clampex software (version 10, Molecular Devices). Data were filtered at 2 kHz and sampled at 10 kHz. Postsynaptic currents and exogenous neurotransmitter/agonist-induced whole-cell currents were recorded at a holding voltage of −60 mV. Exogenous neurotransmitters and agonists were applied by pressure-ejection through a glass pipette using a Picospritzer III (Parker Hannifin, Hollis, NH) with pressure set at 10 pSi and pulse duration at 30 ms. The extracellular solution contained (in mM) NaCl 140, KCl 5, CaCl2 5, MgCl2 5, dextrose 11 and HEPES 5 (pH 7.2). The pipette solution contained (in mM) 120 KCl, 20 KOH, 5 Tris, 0.25 CaCl2, 4 MgCl2, 36 sucrose, 5 EGTA, and 4 Na2ATP (pH 7.2) except for that used for recording PSC bursts from body-wall muscle cells, which contained (in mM) 6.8 KCl, 113.2 Kgluconate, 20 KOH, 5 Tris, 0.25 CaCl2, 4 MgCl2, 36 sucrose, 5 EGTA and 4 Na2ATP (pH 7.2).
Xenopus oocyte expression
Oocytes were obtained from wild-type South African clawed frogs (Xenopus Express, Brooksville, FL) following a protocol approved by the Institutional Animal Care and Use Committee of UConn Health. Capped cRNA of lgc-46 was synthesized using a T3 mMessage mMachine Kit (Life Technologies, Carlsbad, CA, USA), and injected into defolliculated oocytes (∼50 nl per oocyte at 1.0 ng nl−1) using a Drummond Nanoject II injector (Drummond Scientific, Broomall, PA, USA). Two-electrode whole-cell recordings were performed 4–5 days after cRNA injection using an oocyte clamp amplifier (OC-725C, Warner Instruments, Hamden, CT, USA) and the Clampex software. The holding voltage was −60 mV. Data were filtered at 1 kHz and sampled at 10 kHz. In each experiment, a single oocyte was placed in an oocyte perfusion chamber (AutoMate Scientific, Berkeley, CA, USA) containing ND96 solution (compositions in mM: 96 NaCl, 2.0 KCl, 1.8 CaCl2, 1.0 MgCl2, 5.0 HEPES, pH 7.2). Candidate agonists were perfused into the oocyte perfusion chamber through a micro-manifold (AutoMate Scientific). An 8-channel perfusion controller (Valvelink8.2, AutoMate Scientific) was used for perfusion with its valves controlled by pClamp.
Young adult worms of an integrated strain expressing GCaMP2 in body-wall muscle cells54 were glued and filleted for imaging spontaneous fluorescence changes of body-wall muscle cells using an electron-multiplying CCD camera (iXonEM+885, Andor Technology, Belfast, Northern Ireland), a FITC filter set (59222, Chroma Technology Corp.), a light source (Lambda XL, Sutter Instrument, Novato, CA, USA), and the NIS-Elements software (Nikon). Images were acquired at 16 frames per second with 10–40 ms of exposure time (no binning) for 3 min. TTL signals from the camera were used to synchronize the recordings of Ca2+ transients and PSCs.
Worms of an integrated strain expressing channelrhodopsin-2 in command interneurons under the control of Pglr-1 (ref. 16) were grown to L1-L2 stage on standard worm culture plates, and then transferred to new plates either with or without (for negative control) all-trans retinal two days before the experiment. The retinal plates were prepared by spotting each plate (60-mm diameter with 10 ml agar) with 200 μl OP50 E. coli containing 2 mM retinal (R2500, Sigma-Aldrich). Photostimulation was applied through a 40X water immersion objective in 2-s pulses at 30-s intervals using a light source (Lambda XL with SmartShutter, Sutter Instrument) and a 470±20 nm excitation filter (59222, Chroma Technology Corp.). The measured light intensity at the specimen position was 6.7 mW mm−2, which was sufficient to cause maximal evoked peak responses16.
Measurements of backward locomotion
Backward locomotion in response to head touch by a platinum wire was assayed with L4-stage worms. The number of dorsoventral tail swings in response to each touch was counted by the experimenter with a stereomicroscope (SMZ-800, Nikon). One investigator (B.C.) prepared the worm strains whereas another (P.L.) performed the locomotion assay without knowing the strain genotypes.
Acetylcholine (AC159170050, ACROS Organics), GABA (AC103280250, ACROS Organics), aspartate (11240, Sigma-Aldrich), ATP (A3377, Sigma-Aldrich), dopamine (H8502, Sigma-Aldrich), glutamate (49621, Sigma-Aldrich), octopamine (O0250, Sigma-Aldrich), serotonin (H9523, Sigma-Aldrich), tyramine (T2879, Sigma-Aldrich), nicotine (AC181420050, ACROS Organics), levamisole (AC187870100, ACROS Organics), choline (AC110295000, ACROS Organics), glycine (G46-1, Fisher Scientific), d-tubocurarine (T2397, Sigma-Aldrich), gabazine (S106, Sigma-Aldrich), and histamine (H7250, Sigma-Aldrich) were first dissolved in water to make frozen stock solutions (10 to 100 mM), which were diluted to final concentrations using extracellular solutions before use.
The duration and charge transfer of PSC bursts were quantified with Clampfit software (version 10, Molecular Devices) as previously described16. The frequency of PSC bursts was manually counted. Amplitudes of junctional current and membrane voltage were quantified based on the mean amplitude during the last 100 ms at each voltage step (250 ms) or current injection step (1,000 ms) using pClamp. Calcium imaging data were analysed as previously described54. Data graphing and statistical analyses were performed with Origin Pro 2015 (OriginLab Corporation, Northampton, MA). Data are shown as mean±s.e. Either ANOVA (with Tukey’s HSD test) or unpaired t-test was used for statistical comparisons as specified in figure legends. P<0.05 is considered to be statistically significant. The sample size (n) equals to the number of cells or cell pairs recorded, or the number of worms used in locomotion analyses.
All data generated or analysed during this study are included in this published article and its Supplementary Information files.
How to cite this article: Liu, P. et al. Antidromic-rectifying gap junctions amplify chemical transmission at functionally mixed electrical-chemical synapses. Nat. Commun. 8, 14818 doi: 10.1038/ncomms14818 (2017).
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Pereda, A. E. Electrical synapses and their functional interactions with chemical synapses. Nat. Rev. Neurosci. 15, 250–263 (2014).
Korn, H. & Faber, D. S. The Mauthner cell half a century later: a neurobiological model for decision-making? Neuron 47, 13–28 (2005).
Rash, J. E. et al. Molecular and functional asymmetry at a vertebrate electrical synapse. Neuron 79, 957–969 (2013).
Pereda, A. E., Rash, J. E., Nagy, J. I. & Bennett, M. V. Dynamics of electrical transmission at club endings on the Mauthner cells. Brain Res. Brain Res. Rev. 47, 227–244 (2004).
Song, J., Ampatzis, K., Bjornfors, E. R. & El Manira, A. Motor neurons control locomotor circuit function retrogradely via gap junctions. Nature 529, 399–402 (2016).
Allen, M. J. & Murphey, R. K. The chemical component of the mixed GF-TTMn synapse in Drosophila melanogaster uses acetylcholine as its neurotransmitter. Eur. J. Neurosci. 26, 439–445 (2007).
Phelan, P. et al. Molecular mechanism of rectification at identified electrical synapses in the Drosophila giant fiber system. Curr. Biol. 18, 1955–1960 (2008).
Bautista, W., McCrea, D. A. & Nagy, J. I. Connexin36 identified at morphologically mixed chemical/electrical synapses on trigeminal motoneurons and at primary afferent terminals on spinal cord neurons in adult mouse and rat. Neuroscience 263, 159–180 (2014).
Hamzei-Sichani, F. et al. Mixed electrical-chemical synapses in adult rat hippocampus are primarily glutamatergic and coupled by connexin-36. Front. Neuroanat. 6, 13 (2012).
Nagy, J. I., Bautista, W., Blakley, B. & Rash, J. E. Morphologically mixed chemical-electrical synapses formed by primary afferents in rodent vestibular nuclei as revealed by immunofluorescence detection of connexin36 and vesicular glutamate transporter-1. Neuroscience 252, 468–488 (2013).
Rash, J. E. et al. Mixed synapses discovered and mapped throughout mammalian spinal cord. Proc. Natl Acad. Sci. USA 93, 4235–4239 (1996).
Vivar, C., Traub, R. D. & Gutierrez, R. Mixed electrical-chemical transmission between hippocampal mossy fibers and pyramidal cells. Eur. J. Neurosci. 35, 76–82 (2012).
Goodman, M. B., Hall, D. H., Avery, L. & Lockery, S. R. Active currents regulate sensitivity and dynamic range in C. elegans neurons. Neuron 20, 763–772 (1998).
Corsi, A. K., Wightman, B. & Chalfie, M. A transparent window into biology: a primer on Caenorhabditis elegans. Genetics 200, 387–407 (2015).
de Bono, M. & Maricq, A. V. Neuronal substrates of complex behaviors in C. elegans. Annu. Rev. Neurosci. 28, 451–501 (2005).
Liu, P., Chen, B. & Wang, Z. W. Postsynaptic current bursts instruct action potential firing at a graded synapse. Nat. Commun. 4, 1911 (2013).
Liu, P., Chen, B. & Wang, Z. W. Gap junctions synchronize action potentials and Ca2+ transients in Caenorhabditis elegans body wall muscle. J. Biol. Chem. 286, 44285–44293 (2011).
Richmond, J. E. & Jorgensen, E. M. One GABA and two acetylcholine receptors function at the C. elegans neuromuscular junction. Nat. Neurosci. 2, 791–797 (1999).
Fleming, J. T. et al. Caenorhabditis elegans levamisole resistance genes lev-1, unc-29, and unc-38 encode functional nicotinic acetylcholine receptor subunits. J. Neurosci. 17, 5843–5857 (1997).
Jospin, M. et al. A neuronal acetylcholine receptor regulates the balance of muscle excitation and inhibition in Caenorhabditis elegans. PLoS Biol. 7, e1000265 (2009).
Boulin, T. et al. Eight genes are required for functional reconstitution of the Caenorhabditis elegans levamisole-sensitive acetylcholine receptor. Proc. Natl Acad. Sci. USA 105, 18590–18595 (2008).
Kawano, T. et al. An imbalancing act: gap junctions reduce the backward motor circuit activity to bias C. elegans for forward locomotion. Neuron 72, 572–586 (2011).
Wenningmann, I. & Dilger, J. P. The kinetics of inhibition of nicotinic acetylcholine receptors by (+)-tubocurarine and pancuronium. Mol. Pharmacol. 60, 790–796 (2001).
Rand, J. B. in WormBook (ed. The C. elegans Research Community) http://dx.doi.org/10.1895/wormbook.1.131.1 (January 30, 2007).
White, J. G., Southgate, E., Thomson, J. N. & Brenner, S. The structure of the ventral nerve cord of Caenorhabditis elegans. Philos. Trans. R. Soc. Lond. B Biol. Sci. 275, 327–348 (1976).
Chen, B. L., Hall, D. H. & Chklovskii, D. B. Wiring optimization can relate neuronal structure and function. Proc. Natl Acad. Sci. USA 103, 4723–4728 (2006).
Piggott, B. J., Liu, J., Feng, Z., Wescott, S. A. & Xu, X. Z. The neural circuits and synaptic mechanisms underlying motor initiation in C. elegans. Cell 147, 922–933 (2011).
Chalfie, M. et al. The neural circuit for touch sensitivity in Caenorhabditis elegans. J. Neurosci. 5, 956–964 (1985).
Wicks, S. R. & Rankin, C. H. Integration of mechanosensory stimuli in Caenorhabditis elegans. J. Neurosci. 15, 2434–2444 (1995).
Schmitt, C., Schultheis, C., Husson, S. J., Liewald, J. F. & Gottschalk, A. Specific expression of channelrhodopsin-2 in single neurons of Caenorhabditis elegans. PLoS ONE 7, e43164 (2012).
Pokala, N., Liu, Q., Gordus, A. & Bargmann, C. I. Inducible and titratable silencing of Caenorhabditis elegans neurons in vivo with histamine-gated chloride channels. Proc. Natl Acad. Sci. USA 111, 2770–2775 (2014).
Jones, A. K. & Sattelle, D. B. The cys-loop ligand-gated ion channel gene superfamily of the nematode, Caenorhabditis elegans. Invert. Neurosci. 8, 41–47 (2008).
Barbagallo, B., Prescott, H. A., Boyle, P., Climer, J. & Francis, M. M. A dominant mutation in a neuronal acetylcholine receptor subunit leads to motor neuron degeneration in Caenorhabditis elegans. J. Neurosci. 30, 13932–13942 (2010).
Cinar, H., Keles, S. & Jin, Y. Expression profiling of GABAergic motor neurons in Caenorhabditis elegans. Curr. Biol. 15, 340–346 (2005).
Trigo, F. F. et al. Presynaptic miniature GABAergic currents in developing interneurons. Neuron 66, 235–247 (2010).
Tsien, R. Y. New calcium indicators and buffers with high selectivity against magnesium and protons: design, synthesis, and properties of prototype structures. Biochemistry 19, 2396–2404 (1980).
Miller, D. M. 3rd & Niemeyer, C. J. Expression of the unc-4 homeoprotein in Caenorhabditis elegans motor neurons specifies presynaptic input. Development 121, 2877–2886 (1995).
Fox, R. M. et al. A gene expression fingerprint of C. elegans embryonic motor neurons. BMC Genomics 6, 42 (2005).
Kratsios, P., Stolfi, A., Levine, M. & Hobert, O. Coordinated regulation of cholinergic motor neuron traits through a conserved terminal selector gene. Nat. Neurosci. 15, 205–214 (2012).
Starich, T. A., Xu, J., Skerrett, I. M., Nicholson, B. J. & Shaw, J. E. Interactions between innexins UNC-7 and UNC-9 mediate electrical synapse specificity in the Caenorhabditis elegans locomotory nervous system. Neural Dev. 4, 16 (2009).
Zhang, L., Ward, J. D., Cheng, Z. & Dernburg, A. F. The auxin-inducible degradation (AID) system enables versatile conditional protein depletion in C. elegans. Development 142, 4374–4384 (2015).
Dickinson, D. J., Ward, J. D., Reiner, D. J. & Goldstein, B. Engineering the Caenorhabditis elegans genome using Cas9-triggered homologous recombination. Nat. Methods 10, 1028–1034 (2013).
Liu, P., Chen, B. & Wang, Z. W. SLO-2 potassium channel is an important regulator of neurotransmitter release in Caenorhabditis elegans. Nat. Commun. 5, 5155 (2014).
Mellem, J. E., Brockie, P. J., Madsen, D. M. & Maricq, A. V. Action potentials contribute to neuronal signaling in C. elegans. Nat. Neurosci. 11, 865–867 (2008).
Starich, T. A., Herman, R. K. & Shaw, J. E. Molecular and genetic analysis of unc-7, a Caenorhabditis elegans gene required for coordinated locomotion. Genetics 133, 527–541 (1993).
Marder, E. Electrical synapses: rectification demystified. Curr. Biol. 19, R34–R35 (2009).
Pereira, L. et al. A cellular and regulatory map of the cholinergic nervous system of C. elegans. Elife 4, e12432 (2015).
Ohnishi, N., Kuhara, A., Nakamura, F., Okochi, Y. & Mori, I. Bidirectional regulation of thermotaxis by glutamate transmissions in Caenorhabditis elegans. EMBO J. 30, 1376–1388 (2011).
Duerr, J. S., Han, H. P., Fields, S. D. & Rand, J. B. Identification of major classes of cholinergic neurons in the nematode Caenorhabditis elegans. J. Comp. Neurol. 506, 398–408 (2008).
Chorev, M. Carmel L. The function of introns. Front. Genet. 3, 55 (2012).
Sanyal, A., Lajoie, B. R., Jain, G. & Dekker, J. The long-range interaction landscape of gene promoters. Nature 489, 109–113 (2012).
Dekker, J. Gene regulation in the third dimension. Science 319, 1793–1794 (2008).
Gao, S. & Zhen, M. Action potentials drive body wall muscle contractions in Caenorhabditis elegans. Proc. Natl Acad. Sci. USA 108, 2557–2562 (2011).
Liu, P. et al. Genetic dissection of ion currents underlying all-or-none action potentials in C. elegans body-wall muscle cells. J. Physiol. 589, 101–117 (2011).
Nakamura, M. & Jang, I. S. Presynaptic nicotinic acetylcholine receptors enhance GABAergic synaptic transmission in rat periaqueductal gray neurons. Eur. J. Pharmacol. 640, 178–184 (2010).
Fu, W. M., Liou, H. C. & Chen, Y. H. Nerve terminal currents induced by autoreception of acetylcholine release. J. Neurosci. 18, 9954–9961 (1998).
Sohl, G., Maxeiner, S. & Willecke, K. Expression and functions of neuronal gap junctions. Nat. Rev. Neurosci. 6, 191–200 (2005).
Connors, B. W. & Long, M. A. Electrical synapses in the mammalian brain. Annu. Rev. Neurosci. 27, 393–418 (2004).
Bennett, M. V. & Zukin, R. S. Electrical coupling and neuronal synchronization in the mammalian brain. Neuron 41, 495–511 (2004).
Allen, M. J., Godenschwege, T. A., Tanouye, M. A. & Phelan, P. Making an escape: development and function of the Drosophila giant fibre system. Semin. Cell Dev. Biol. 17, 31–41 (2006).
Nagy, J. I. Evidence for connexin36 localization at hippocampal mossy fiber terminals suggesting mixed chemical/electrical transmission by granule cells. Brain Res. 1487, 107–122 (2012).
Esposito, G., Di Schiavi, E., Bergamasco, C. & Bazzicalupo, P. Efficient and cell specific knock-down of gene function in targeted C. elegans neurons. Gene 395, 170–176 (2007).
Liu, Q., Chen, B., Ge, Q. & Wang, Z. W. Presynaptic Ca2+/calmodulin-dependent protein kinase II modulates neurotransmitter release by activating BK channels at Caenorhabditis elegans neuromuscular junction. J. Neurosci. 27, 10404–10413 (2007).
This work was supported by National Institute of Health (2R01MH085927 to Z.-W.W., and R01GM113004 to B.C.) and Air Force Office of Scientific Research (FA9550-15-1-0060 to R.M.). We thank Cori Bargmann for her insightful suggestions during the course of this study and for the HisCl1 strain. We thank Alexander Gottschalk for the pCoS10 and pNP259 plasmids, Abby F. Dernburg for the pLZ31 plasmid, Xiaoming Xia for an oocyte expression vector, and Caenorhabditis Genetics Center (USA) for mutant strains.
The authors have no competing interests as defined by Nature Publishing Group, or other interests that might be perceived to influence the results and/or discussion reported in this article.
Supplementary Figures. (PDF 1385 kb)
Supplementary Movie 1
Temporal correlation between postsynaptic current bursts in VA5 and calcium transients in nearby body-wall muscle cells. (WMV 1947 kb)
Supplementary Movie 2
Nose touch response of a wild-type worm. (WMV 1627 kb)
Supplementary Movie 3
Nose touch response of a worm with AVA-specific unc-17 knockdown. (WMV 1486 kb)
Supplementary Movie 4
Nose touch response of an lgc-46(ok2949) mutant worm. (WMV 744 kb)
Supplementary Movie 5
Nose touch response of an lgc-46(ok2900) mutant worm. (WMV 689 kb)
Supplementary Movie 6
Nose touch response of a worm with A-motoneuron-specific lgc-46 knockdown. (WMV 1470 kb)
Supplementary Movie 7
Nose touch response of a rescued lgc-46(ok2949) mutant worm. (WMV 1510 kb)
Supplementary Movie 8
Nose touch response of a rescued lgc-46(ok2900) mutant worm. (WMV 1681 kb)
Supplementary Movie 9
Nose touch response of a worm with AVA-specific unc-7 knockdown. (WMV 752 kb)
Supplementary Movie 10
Nose touch response of a worm with A-motoneuron-specific unc-9 knockdown. (WMV 830 kb)
Rights and permissions
This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
About this article
Cite this article
Liu, P., Chen, B., Mailler, R. et al. Antidromic-rectifying gap junctions amplify chemical transmission at functionally mixed electrical-chemical synapses. Nat Commun 8, 14818 (2017). https://doi.org/10.1038/ncomms14818
This article is cited by
C. elegans enteric motor neurons fire synchronized action potentials underlying the defecation motor program
Nature Communications (2022)
Evidence of Biorealistic Synaptic Behavior in Diffusive Li-based Two-terminal Resistive Switching Devices
Scientific Reports (2020)
NMDAR-mediated modulation of gap junction circuit regulates olfactory learning in C. elegans
Nature Communications (2020)
Active propagation of dendritic electrical signals in C. elegans
Scientific Reports (2019)
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.