Signal Decoding for Glutamate Modulating Egg Laying Oppositely in Caenorhabditiselegans under Varied Environmental Conditions

Summary

Animals' ability to sense environmental cues and to integrate this information to control fecundity is vital for continuing the species lineage. In this study, we observed that the sensory neurons Amphid neuron (ASHs and ADLs) differentially regulate egg-laying behavior in Caenorhabditis elegans under varied environmental conditions via distinct neuronal circuits. Under standard culture conditions, ASHs tonically release a small amount of glutamate and inhibit Hermaphrodite specific motor neuron (HSN) activities and egg laying via a highly sensitive Glutamate receptor (GLR)-5 receptor. In contrast, under Cu²⁺ stimulation, ASHs and ADLs may release a large amount of glutamate and inhibit Amphid interneuron (AIA) interneurons via low-sensitivity Glutamate-gated chloride channel (GLC)-3 receptor, thus removing the inhibitory roles of AIAs on HSN activity and egg laying. However, directly measuring the amount of glutamate released by sensory neurons under different conditions and assaying the binding kinetics of receptors with the neurotransmitter are still required to support this study directly.

Graphical Abstract

Highlights

• •Short-term exposure of CuSO₄ evokes hyperactive egg laying
• •ASHs inhibit HSNs and egg laying via GLR-5 receptor under no Cu²⁺ treatment
• •AIA interneurons suppress HSNs and thus egg laying through ACR-14 signaling
• •Under noxious Cu²⁺ treatment, ASHs and ADLs suppress AIAs and augment egg laying

Behavioral Neuroscience; Biological Sciences; Environmental Science

Introduction

The establishment of a succeeding generation is essential for the survival of all biological species. Therefore, it is of paramount importance for animals to breed, undergo pregnancy, and give birth to offspring at the right time and place. In this context, environmental conditions are of great importance since unsuitable environmental conditions cannot support the development of offspring. For this reason, animals must sense their environment, integrate sensory information, and translate the information into the reproductive process.

Remarkably, some animal species increase their reproduction rates in response to environmental stresses. For example, Acheta domesticus crickets increase egg laying in response to infections with bacteria and parasites (Adamo, 1999). Also, Drosophila nigrospiracula males infected experimentally by parasites courted females significantly more than unparasitized control males before dying, albeit they lived a shorter life than the control males. This higher courtship results in an increased mating rate and potentially greater reproductive success (Polak and Starmer, 1998). Culex pipiens mosquitoes infected with the parasite Plasmodium relictum bring forward their oviposition schedule (Vezilier et al., 2015). Freshwater Pomacea canaliculata snails produce eggs earlier in the presence of Trachemys scripta elegans predators (Guo et al., 2017). Intensive harvesting has caused pronounced shifts in growth rate, reproductive timing, and even genomic shifts in many commercially valuable stocks (Allendorf and Hard, 2009; Baskett and Barnett, 2015; Eikeset et al., 2016; Heino et al., 2015; Therkildsen et al., 2019). These observations and many others could be explained by the life-history theory, which argues that when organisms experience environmental conditions that can reduce their long-term reproductive success, they increase their present reproductive output through phenotypic plasticity (Dingle, 1990; Fox et al., 2019; Hirshfield and Tinkle, 1975; Kavanagh and Kahl, 2016; Neil and Ford, 1989). However, the molecular and neuronal mechanisms underlying this adaptive response have not been fully elucidated.

Both genes and the modalities of neuroinformation integration are conserved among animal genera. Experimentally tractable animal models are generally used to study the relationships among the genes, proteins, and neural circuits underlying behaviors. The nematode Caenorhabditis elegans (C. elegans) is a favored model for behavioral studies because of its compact nervous system and experimental tractability (de Bono and Maricq, 2005). The C. elegans hermaphrodite nervous system consists of merely 302 neurons and 7,446 synapses (Cook et al., 2019; https://www.wormatlas.org; White et al., 1986). Egg laying in C. elegans is a process by which hermaphrodites deposit developing embryos into the environment. A small network of motor neurons (two HSNs and six ventral nerve cord [VC] motor neurons) and specialized smooth muscle cells (four vulval vm2, four vulval vm1, and eight uterine muscle cells) control the laying of eggs. HSNs play a central role in egg-laying behavior. These neurons not only control vm2 vulval muscles and direct synaptic output to VC5 motor neurons but also receive synaptic input from the rest of the nervous system and feedback from muscle cells (Collins et al., 2016; Collins and Koelle, 2013; Schafer, 2005, 2006; Zhang et al., 2008, 2010). The tight regulation of egg laying results in adaptive egg laying and coordination with other behaviors (de Bono and Maricq, 2005; Schafer, 2005). Egg laying is regulated by HSNs through a number of neurotransmitters: serotonin (Greenwald and Horvitz, 1982), acetylcholine (Duerr et al., 2001), and at least three neuropeptides encoded by flp-19, nlp-3, and nlp-15 (Brewer et al., 2019; Kim and Li, 2004; Nathoo et al., 2001; Schafer, 2006). Serotonin stimulates vulval muscle activity by increasing the frequency of spontaneous calcium transients (Shyn et al., 2003). Two serotonin receptors, SER-1 and SER-7, appear to be involved in activating vulval muscle activity (Dempsey et al., 2005; Hobson et al., 2006). HSNs also receive humoral modulation from endocrine uv1 cells, which release tyramine (Alkema et al., 2005), and the neuropeptides FMRFamide-related neuropeptides (FLP-11 and FLP-22) (Kim and Li, 2004).

Many environmental factors affect the rate and temporal pattern of egg laying. Food supply affects egg-laying behavior. C. elegans lays more eggs in the presence of abundant food than in the absence of food (Trent et al., 1983). Gentle touch inhibits egg laying (Sawin, 1996; Zhang et al., 2008) via sensation in the touch receptor nterior lateral microtubule cell (ALM) and Posterior lateral microtubule cell (PLM) neurons (Zhang et al., 2008). High and low osmolarity enhances and suppresses egg-laying and spike frequency in the HSN and VC motor neurons, respectively (Zhang et al., 2008). High glucose reduces the egg-laying rate (Teshiba et al., 2016). High environmental CO₂ (5%) inhibits egg laying and activity in HSNs by activating CO₂-sensitive Amphid wing cell (AWC) neurons (and possibly other sensory neurons) via guanylate cyclase-9 (GCY-9)-TAX-2/TAX-4 signaling (Fenk and de Bono, 2015). Predator-secreted sulfolipids suppress egg laying for many minutes after exposure, as mediated by TAX-4 signaling in sensory Amphid neurons (ASI and ASJ) (Liu et al., 2018). C. elegans responds to environmental stresses to augment its egg laying, similar to other animals. Approximately 48 hr of exposure to teratogens (such as nicotine, cadmium, and tributyltin) at low doses increases the egg-laying rate (Killeen and Marin de Evsikova, 2016). However, possibly due to toxicity, long-term exposure to high levels (100 and 500 μM) of CuSO₄ decreased worm brood size and led to internal hatching of offspring inside their mothers (bagging) (Calafato et al., 2008). Conversely, short-term (<6 hr) exposure to cadmium ions increased the worm's egg-laying rate (Killeen and Marin de Evsikova, 2016). However, the neural mechanisms, i.e., map and neurosignal integration in neural circuits governing the regulation of egg laying by environmental cues, have not been fully elucidated.

The polymodal ASH sensory neurons are primary nociceptor neurons. These neurons are involved in sensing a variety of aversive stimuli and mediating the avoidance of high osmotic, mechanical, and chemical stimuli (Bargmann et al., 1990; Bargmann, 2006; Guo et al., 2015; Hart et al., 1995; Hilliard et al., 2005; Kaplan and Horvitz, 1993). The sensory neurons ADLs, Amphid neuron (ASKs, and ASEs ) play minor roles in the sensation of aversive cues and avoidance behavior from heavy-metal ions Cd²⁺ and Cu²⁺, odors (octanol), high osmotic strength, and Sodium dodecyl sulfate (SDS) (Bargmann et al., 1990; de Bono and Maricq, 2005; Hilliard et al., 2002; Sambongi et al., 1999; Troemel et al., 1995). ASHs and ADLs play primary and secondary roles in sensing and avoiding heavy metals (Sambongi et al., 1999). These neurons sense Cu²⁺ via OSM-9 and capsaicin receptor-related (OCR-2)/osmotic avoidance abnormal (OSM-9) signaling. The transient receptor potential cation channel, subfamily V (TRPV) proteins OCR-2 and OSM-9, which may assemble into a single channel complex (Tobin et al., 2002), mediate the nociception of Cu²⁺ (Guo et al., 2018; Hilliard et al., 2005; Kahn-Kirby and Bargmann, 2006; Sambongi et al., 1999; Wang et al., 2015). ASHs and ADLs modulate Cu²⁺ avoidance primarily via glutamatergic signaling (Choi et al., 2015; Mellem et al., 2002). Glutamate is an essential neurotransmitter in the vertebrate and invertebrate nervous systems. Glutamate signaling modulates a broad range of behaviors mediated by diverse receptors in C. elegans (Whittaker and Sternberg, 2004). Glutamate is involved in the regulation of a broad spectrum of behaviors. Examples include salt chemotaxis (Chalasani et al., 2007), habituation in tap-withdrawal response (Rankin and Wicks, 2000), local search (Hills et al., 2004), cryophilic response (Clark et al., 2007), and pumping inhibition by and initiation of avoidance of quinine (Zou et al., 2018). The functions of glutamate in the regulation of egg laying have not been determined.

In this study, we utilized C. elegans to explore the effects of metal ions on egg laying in C. elegans. We found that short-term (30 min) treatments with the metal ions tested, that is, Cu²⁺, Ni²⁺, Mn²⁺, and Fe³⁺, induced hyperactive egg laying. Using genetic analyses, behavioral tests, genetic and chemogenetic manipulation of neurons, neuronal calcium imaging, and pharmacological tests, we dissected the molecular and neurocircuit mechanisms underlying the effect of Cu²⁺. We also analyzed the mechanisms by which sensory information is encoded, decoded, and translated into behavior.

Results

Noxious Stimulation of CuSO₄ Evokes Hyperactive Egg Laying and HSN Calcium Signals via the OCR-2/OSM-9 Channel Complex in ASH and ADL Sensory Neurons in C. elegans

Heavy-metal ions, such as Cu²⁺, Ni²⁺, Mn²⁺, and Fe³⁺, are common environmental contaminants. These ions are harmful to animals at high contents. We examined the C. elegans egg-laying rate under short-term exposure to Cu²⁺ (CuSO₄), Ni²⁺ (NiCl₂), Mn²⁺ (MnCl₂), and Fe³⁺ (Fe₂(SO₄)₃) ions at 100 μM. A 30-min exposure of Cu²⁺, Ni²⁺, Mn²⁺, and Fe³⁺ at 100 μM significantly increased the number of eggs laid in C. elegans (Figure S1A). We used 30 min of administration of these metal ions based on our time course assay of Cu²⁺. The number of eggs laid significantly increased upon exposure to 100 μM CuSO₄ for 10 min, 20 min, and 30 min and then decreased thereafter (Figure 1A). CuSO₄ (Cu²⁺) is a commonly used noxious chemical in studies of nociception and avoidance in C. elegans. Thus, we used 30 min of treatment with Cu²⁺ as stimulation (subsequent short-term Cu²⁺ treatment) in our following tests. The egg-laying rate responded dose-dependently to the short-term challenges of Cu²⁺ at various concentrations, which was accurately fitted by a Hill function with concentration for 50% of maximal effect (EC₅₀) = 4.68 μM, Figure 1B). Adult hermaphrodites generally have a store of 10–15 eggs in their uterus at any given time (Schafer, 2005). Augmented egg laying increases the eggs laid at early developmental stages (Alkema et al., 2005; Ringstad and Horvitz, 2008). Short-term Cu²⁺ treatment notably increased the number of eggs at the 1–8 and 9–20 cell stages (Figures 1C and 1D). We summed the ratios of the eggs at these two embryonic stages and indicated them as early eggs. Egg laying is controlled by HSN command motor neurons (de Bono and Maricq, 2005; Schafer, 2005; Collins et al., 2016; Desai and Horvitz, 1989; Trent, 1983; Zhang et al., 2008, 2010). Cu²⁺ treatment should excite HSNs. Thus, we employed calcium imaging to examine HSN activity in response to the acute application of Cu²⁺ at various concentrations using immobilized worms glued onto a Polydimethylsiloxane (PDMS) pad attached to a cover glass, as previously described (Zhang et al., 2008). As shown in Figures 1E and 1F, Ca²⁺ signals in HSNs dose-dependently increased in response to CuSO₄ stimuli (well fitted by the Hill function with EC₅₀ = 2.46 μM). Interestingly, the HSN Ca²⁺ signals and the egg-laying rates correlated linearly (Pearson r = 0.9845, Figure 1G).

Figure 1.

Cu²⁺ Evokes Hyperactive Egg Laying and Augments Calcium Signals in HSN Motor Neurons in C. elegans

(A) The ratio of the eggs laid (the egg number in Cu²⁺-treated worms/that in Cu²⁺-untreated animals) under treatment of 100 μM Cu²⁺ for various durations (5 min, 10 min, 20 min, 30 min, 45 min, 1 hr, 2 hr, 3 hr, 4 hr, 5 hr, and 6 hr) in wild-type N2 worms. Data are expressed as means ± SEM. Statistical significance of difference was analyzed by one-way ANOVA with the post hoc test of Dunnett's multiple comparison correction in comparison of a hypothesized ratio (= 1) in Cu²⁺-untreated worms and indicated as follows: ns = not significant, ∗∗∗p < 0.001 and ∗∗∗∗p < 0.0001.

(B) The number of the eggs laid per worm per 30 min (egg-laying rate) under application of CuSO₄ (Cu²⁺) at various concentrations (in μM: 0, 0.01, 0.1, 1, 10, 50, 100, and 500) in WT worms. The solid blue line indicated the Hill-function fitting. Data are expressed as means ± SEM.

(C) Cumulative percent of the eggs at different embryo stages (1–8 cells, 9–20 cells, 21 + cells, comma, two folds, and three folds+) in WT worms treated without versus with 100 μM Cu²⁺. Data are expressed as means ± SEM. Statistical significance of difference was analyzed by two-way ANOVA with post hoc Sidak's multiple comparisons test in the comparison between the data from Cu²⁺-untreated and Cu²⁺-treated worms, and indicated as follows: ns = not significant and ∗∗∗∗p < 0.0001.

(D) Percent of the eggs at early developmental stages (stages of 1–8 cells and 9–20 cells) in WT worms treated without and with 100 μM CuSO₄. Data are displayed in box plots, with each dot representing the data from each individual tested animal. Statistical significance of difference was analyzed by unpaired t test.

(E) Heat maps of percent changes in somal calcium (Ca²⁺) transients in HSN motor neurons in WT worms treated with CuSO₄ (Cu²⁺) at various concentrations for 5 min ΔF = F - F_B. F, the average fluorescence intensity of the region of interest (ROI) of an HSN soma in each frame; F_B, a background signal defined as the average fluorescence intensity of ROI adjacent to the HSN soma in all frames. The label on the left y axis indicates the number of tested worms.

(F) Hill plot of the somal HSN Ca²⁺ signals in WT worms treated with CuSO₄ at various concentrations shown in (E). Data are expressed as means ± SEM.

(G) Linear correlation between the egg-laying rates and the HSN somal Ca²⁺ signals in WT worms treated with Cu²⁺ at varied concentrations (in μM: 0, 0.01, 0.1, 1, 10, 50, 100, and 500).

See also Figure S1 and Table S1.

Nociceptor ASHs and ADLs sense Cu²⁺ and mediate avoidance of this ion via the OCR-2/OSM-9 complex. These neurons and molecules are likely involved in the augmentation of egg laying by Cu²⁺. Next, we used ocr-2 and osm-9 mutants and genetically rescued worms to test the functional roles of this channel complex in the modulation of egg laying and HSN activity. No ASH-specific promoter was identified. Therefore, we used an FLP-out (a recombinase-catalyzed intramolecular excision of spacer DNA between tandemly oriented transcriptional terminator flanked by FLP recognition [FRT] sites) system to drive specific expression of TeTx in ASH neurons. The system is based on the FLP-out of a transcriptional terminator (Davis et al., 2008; Guo et al., 2015; Macosko et al., 2009; Voutev and Hubbard, 2008). When worms expressed both sra-6p::flp::unc-54 3′UTR and gpa-11p::FRT::stop::FRT::ocr-2(or osm-9)::sl2::GFP constructs, FLP recombinase driven by the sra-6 promoter excised the FRT-flanked transcriptional terminator and drove ocr-2 or osm-9 directed by gpa-11p to express specifically in ASHs (Figure S1B). Our results showed that the ocr-2(ak47) and osm-9(tm5418) mutants and all transgenic worms with neuron-specific genetic rescue (reconstitution) of ocr-2 or osm-9 had no defects in egg laying or HSN Ca²⁺ signals under normal culture conditions. However, the mutants lost the responses of egg-laying behavior and HSN Ca²⁺ transients to the short-term and acute treatments of Cu²⁺ at 100 μM (subsequent acute Cu²⁺ treatment), respectively. The reconstitution of both genes in ASHs or ADLs alone or in both neurons, although not in other neurons, restored wild-type phenotypes in both activities (Figures 2A–2F and S1C–S1F). In addition, ADLs displayed a mild increase in Ca²⁺ transients in response to stimulation with 100 μM Cu²⁺ (Figure S1C). These results indicate that primary ASH and secondary ADL sensory neurons sense Cu²⁺ via OCR-2/OSM-9 signaling and upregulate egg laying in C. elegans.

Figure 2.

Sensory ASH and ADL Neurons Response to Cu²⁺ Treatment and Regulate Egg Laying Activity in HSN Motor Neurons through OSM-9/OCR-2 Signaling

(A–F) The number of the eggs laid per worm per 30 min (egg-laying rate, A and D), percent of the eggs at early developmental stages (both 1–8 cells and 9–20 cells, B and E), and the somal HSN calcium signals (C and F) in the worms of indicated genotypes treated without (in black) or with (in red) 100 μM CuSO₄. ΔF = F - F_B. F, the average fluorescence intensity of the region of interest (ROI) of an HSN soma in each frame; F_B, a background signal defined as the average fluorescence intensity of ROI adjacent to the HSN soma in all frames.

All Data are displayed as box plots with each dot representing the data from each individual tested worm. Statistical significance of difference was analyzed by two-way ANOVA analysis with the post hoc test of Tukey's multiple comparison correction and indicated as follows: ns = not significant, ∗∗∗∗p < 0.0001, and in different colors for varied comparisons. Black or red, a comparison of tested worms with the control (wild type N2) under Cu²⁺-untreated or Cu²⁺-treated conditions; blue, that of Cu²⁺-untreated with Cu²⁺-treated animals of the same genotype.

See also Figure S1 and Table S1.

ASH and ADL Sensory Neurons Regulate Egg Laying and HSN Activity Differently under Standard Culture Conditions and Cu²⁺ Treatment

ASHs and ADLs play roles in avoidance of heavy metals, primarily via glutamatergic signaling (Choi et al., 2015; Mellem et al., 2002; Sambongi et al., 1999). ASHs and HSNs but not ADLs and HSNs connect with each other via chemosynapses (Cook et al., 2019; https://www.wormatlas.org; White et al., 1986). The vesicular glutamate transporter EAT-4 is essential for glutamate filling in synaptic vesicles (Bellocchio et al., 2000; Lee et al., 1999). Thus, we employed an eat-4 mutant to test glutamate regulation of egg laying and HSN activity. Interestingly, the eat-4(ky5) mutant exhibited the maximum egg-laying rate and HSN Ca²⁺ signals under both standard culture conditions and Cu²⁺ treatments. Genetic rescuing of eat-4 in ASHs alone but not in ADLs restored the wild-type phenotypes, including egg-laying and HSN activity in the transgenes under both conditions. The eat-4 rescue in ADLs even augmented both Cu²⁺-elicited activities (Figures 3A–3C and S3A). However, this observation may be attributable to EAT-4 overexpression. The reconstitution of eat-4 in other eat-4-expressing sensory neurons, such as ADFs, ASKs, ASJs, and AWCs, had no impact under either state (S2A and S2B). Taken together, this set of results suggests that ASHs and ADLs modulate egg-laying behavior differently. Next, we set out to identify the glutamatergic receptor(s) involved in egg-laying regulation. Among the mutants that we tested, glr-5(tm3506) displayed defects in egg laying under standard culture conditions and short-term Cu²⁺ treatment, while glc-3(ok321) showed behavioral abnormalities only under Cu²⁺ treatment (Figures S2C and S2D). We first focused on the glr-5 gene, which encodes a GLR-5 glutamate receptor. GLR-5 is a kainate (non-NMDA)-type ionotropic receptor subunit (Brockie et al., 2001; Brockie and Maricq, 2003). GLR-5 is expressed in interneurons and motoneurons, including HSNs and VCs (https://wormbase.org). We employed mutants of glr-5(tm3506) single mutant, eat-4(ky5); glr-5(tm3506) double mutant, and eat-4 and glr-5 rescued worms to further test the regulatory roles of glutamatergic signaling in modulating egg laying and HSN activity. Expressing glr-5 in all glr-5-expressing cells (under its promoter region) or HSNs directed by the egl-6a promoter (Emtage et al., 2012) restored the wild-type phenotypes (Figures 3D–3F and S3A). The eat-4; glr-5 double mutant displayed similar phenotypes to eat-4(ky5). However, this mutant showed different phenotypes than glr-5(tm3506). Only dual reconstitution of both eat-4 in ASHs and glr-5 in HSNs (in worms of ZXW1425 strain, its genotype showed in Table S1), neither single re-expression of eat-4 in ASHs nor glr-5 in HSNs, restored wild-type phenotypes (Figures 3G–3I and S3B).

Figure 3.

Sensory ASH Neurons Differently Modulate Egg-Laying and HSN Activity under Standard Culture Conditions and Cu²⁺ Treatment via Glutamatergic GLR-5

(A–L) The number of the eggs laid per worm per 30 min (egg-laying rate, A, D, G, and J), percent of the eggs at early developmental stages (both 1–8 cells and 9–20 cells, B, E, H, and K), and the HSN somal calcium signals (C, F, I, and L) in the worms of indicated genotypes treated without (in black) or with (in red) 100 μM CuSO₄. ΔF = F - F_B. F, the average fluorescence intensity of the region of interest (ROI) of an HSN soma in each frame; F_B, a background signal defined as the average fluorescence intensity of ROI adjacent to the HSN soma in all frames.

All Data are displayed as box plots with each dot representing the data from each individual tested worm. Statistical significance of difference was analyzed by two-way ANOVA analysis with the post hoc test of Tukey's multiple comparison correction and indicated as follows: ns = not significant, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, and in different colors for varied comparisons. Black or red, a comparison of tested worms with the control (wild type N2) under Cu²⁺-untreated or Cu²⁺-treated conditions; blue, that of Cu²⁺-untreated with Cu²⁺-treated worms of the same genotype.

See also Figures S2–S4 and Table S1.

To specifically block neurotransmission in ASHs, we used the tetanus toxin system (TeTx). TeTx is a specific protease of synaptobrevin (Schiavo et al., 1992) that blocks vesicle fusion with the plasma membrane and thus neurotransmission. The TeTx system has been used successfully in neurobiological studies to inhibit chemical synaptic transmission in tested neurons (subsequent neuron::TeTx in short) in C. elegans (Guo et al., 2015, 2018; Liu et al., 2019; Macosko et al., 2009; Wang et al., 2016). As expected, ASH::TeTx resulted in defects in egg-laying and HSN Ca²⁺ signals in both Cu²⁺-untreated and Cu²⁺-treated worms. However, ADL::TeTx resulted in normal egg laying and HSN activity under standard culture conditions and loss of the responses of both activities to Cu²⁺ treatments (Figure 3J–3L and S3C). However, the expression of TeTx in other sensory neurons, except ASIs, did not significantly affect egg-laying behavior (Figures S4A and S4B). Given that ASIs reciprocally inhibit ASHs (Guo et al., 2015), the effects of ASI::TeTx are likely a result of the functional interaction between two sensory neurons.

Continual inhibition of neurotransmission by TeTx at embryonic periods may interfere with the development of the nervous system. Therefore, spatiotemporally optogenetic and chemogenetic manipulation of neurons is used favorably in neuroscience studies. We further employed chemogenetic activation of ASHs and ADLs as further evidence to support our results of TeTx manipulation. We expressed TRPV1 in ASHs and ADLs and employed capsaicin to excite tested neurons under standard culture conditions (Guo et al., 2015, 2018; Liu et al., 2019; Tobin et al., 2002). The chemogenetic activation of ADLs alone did not affect egg laying. In contrast, the activation of ASHs or both ASHs and ADLs (i.e., ASHs) and even TRPV1 expression per se in ASHs significantly inhibited spontaneous egg laying (Figure S4C). This result suggests that our chemogenetic activation of ASHs and ADLs was not strong enough to mimic Cu²⁺ stimulation, and the TRPV1 channel may have displayed leaky activity.

In summary, our results show that primary ASHs and secondary ADLs regulate egg laying and HSN activity differently. ASHs modulate egg laying and HSN activity under both standard culture conditions and Cu²⁺ treatments, while ADLs regulate both activities only under Cu²⁺ exposure. Thus, there may be two different glutamate effect sites or two receptors.

ASH/ADL, AIA, and HSN Disinhibitory Circuits Initiate Cu²⁺-Induced Hyperactive Egg Laying

glc-3(ok321) mutants do not enhance egg laying in the presence of Cu²⁺ (Figures S2C and S2D). glc-3 is expressed in many interneurons, including AIAs and AIYs, and motor neurons (Blazie et al., 2017; https://wormbase.org; Shinkai et al., 2011; Wenick and Hobert, 2004). AIAs are postsynaptic to ASHs and ADLs (Cook et al., 2019; https://www.wormatlas.org; White et al., 1986). Thus, GLC-3 signaling very likely mediates the regulation of egg laying by ASHs and ADLs under Cu²⁺ treatment. The GLC-3 protein is an L-glutamate-gated chloride channel subunit (Horoszok et al., 2001). This protein mediates the inhibition of AIAs by ASHs in the regulation of behavioral choice on stimulation of attractive diacetyl and noxious Cu²⁺ (Shinkai et al., 2011). We first used neuron::TeTx transgenes to screen glc-3-expressing interneurons that may function in egg-laying regulation. The AIA::TeTx transgene displayed maximal egg laying under standard culture conditions and exhibited no behavioral response to short-term Cu²⁺ treatment (Figures S4A and S4B). Therefore, we decided to focus on AIA neurons. We used glc-3 mutant, glc-3 genetically rescued worms, and AIA::TeTx transgene to further test AIA function in the regulation of HSN activity and egg laying. As expected, the AIA::TeTx transgene also displayed the same phenotype in HSN activity as egg laying (Figures 4A–4C). However, the glc-3(ok321) mutant did not phenocopy the AIA::TeTx transgene. The mutant displayed only hypoactive egg laying and HSN activity under Cu²⁺ treatments. The genetic rescue of glc-3 in GLC-3-expressing cells (driven by glc-3p) or AIA alone (directed by gcy-28(d)p) but not in AIY interneurons (directed by T19C4.5p [Chalasani et al., 2007]) fully restored wild-type phenotypes (Figures 4A–4C and S5A). This result indicated that GLC-3 signaling in AIAs may only regulate egg laying and HSN activity under Cu²⁺ treatment; however, AIAs per se regulate both activities in C. elegans under both states. AIAs may tonically inhibit egg laying in normal cultured worms. To test the regulatory function of the ASH/ADL and AIA circuits, we next used the eat-4(ky5); glc-3(ok321) double mutant and reconstituted eat-4, glc-3, and both genes in the mutant. Interestingly, eat-4(ky5); glc-3(ok321) fully phenocopied the AIA::TeTx transgene. Only the dual reconstitution of both eat-4 in ASHs or ASHs/ADLs and glc-3 in AIAs restored wild-type phenotypes in both Cu²⁺-untreated and Cu²⁺-treated worms. Dual reconstitution of eat-4 in ADLs and glc-3 in AIAs was enough to restore wild-type phenotypes in Cu²⁺-treated but not in Cu²⁺-untreated animals (Figures 4D–4F and S5B). The abovementioned results suggest that GLC-3 signaling in AIAs is involved in egg-laying regulation under Cu²⁺ treatments and that glutamate from ASHs modulates egg-laying and HSN activity under both states.

Figure 4.

AIA Interneurons Mediate Cu²⁺-evoked Augment of Egg-Laying and HSN Activity via GLC-3 Signaling

(A–F) Egg-laying rates (A and D), percent of eggs at early development stages (B and E), and percent changes in the HSN calcium signals (C and F) in the worms of indicated genotypes without (in black) or with (in red) the treatment of 100 μM CuSO₄. Data are displayed as box plots with each dot representing the data from each individual tested worm. Statistical significance of difference was analyzed by two-way ANOVA analysis with the post hoc test of Tukey's multiple comparison correction, and indicated as follows: ns = not significant, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, and in different colors for varied comparisons. Black or red, a comparison of tested worms with the control (wild type N2, WT) under Cu²⁺-untreated or Cu²⁺-treated conditions; blue, that of Cu²⁺-untreated with Cu²⁺-treated worms of the same genotype.

(G) Curves (Left) and Hill plot (Right) of AIA axonal calcium responses to administration of CuSO₄ at various concentrations (in μM: 0, 0.01, 0.1, 1, 10, 50, 100, and 500). Data are presented as means ± SEM, as indicated by solid traces or dots ± gray shading or error bars, n ≥ 12. ΔF = F – F₀. F, an average background signal-subtracted fluorescence intensity in an AIA axon in each frame (in gray shading); F₀, the average background signal-subtracted fluorescence intensity of an AIA axon within the initial 60 s before stimulation.

(H) Linear correlation of the egg-laying rates and the Cu²⁺-elicited changes in the AIA axonal Ca²⁺ transients in WT worms treated with Cu²⁺ at various concentrations. Data on the egg-laying rates share with those in Figure 1B.

(I) Linear correlation of the Cu²⁺-elicited changes in axonal Ca²⁺ transients in AIAs with the HSN Ca²⁺ signals in WT worms treated with Cu²⁺ at various concentrations (in μM: 0, 0.01, 0.1, 1, 10, 50, 100, and 500). Data on the HSN Ca²⁺ signals share with those in Figure 1F.

(J–M) Changes in the axonal Ca²⁺ signals evoked by 100 μM Cu²⁺ in AIA interneurons in WT, neuron blocked worms (J), eat-4 mutant and its transgenetic worms (K), glc-3 mutant and its transgenetic worms (L), and eat-4; glc-3 mutant and its transgenetic worms (M). Data are displayed as box plots with each dot representing the data from each individual tested animal. Data are displayed as box plots with each dot representing the data from each individual tested animal. Statistical significance of difference was analyzed by one-way ANOVA with the post hoc test of Dunnett's multiple comparison correction compared with WT worms or as indicated, and denoted as follows: ns = not significant, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001.

See also Figure S4–S6 and Table S1.

Both egg-laying rates and intensities of HSN Ca²⁺ transients displayed dose-dependent responses to Cu²⁺ stimuli, which were well fitted by the Hill function with EC₅₀ = 4.68 μM and 2.46 μM, respectively (Figures 1B and 1F). Next, we examined the relation of AIA activities with Cu²⁺ doses using Ca²⁺ imaging combined with a microfluidic device. The Ca²⁺ transients in AIAs in wild-type N2 worms dose-dependently decreased upon treatment with Cu²⁺ at various concentrations. The Ca²⁺ responses were well fitted by the Hill function with a half-maximal inhibitory concentration (IC₅₀) = 2.28 μM (Figures 4G and S6A). Although the conditions for behavioral tests and Ca²⁺ imaging were different, the intensities of AIA Ca²⁺ transients negatively and linearly correlated with egg-laying rates and levels of HSN Ca²⁺ transients (Pearson r = −0.9689 and −0.9859, respectively, Figures 4H and 4I). We further examined AIA Ca²⁺ signals in response to Cu²⁺ challenge under genetic and neuronal manipulations as follows. Blocking neurotransmission by TeTx in ASHs, ADLs, and both types of neurons almost equally eliminated the inhibition of Cu²⁺-induced Ca²⁺ transients in AIAs by both sensory neurons (Figures 4J and S6B). Loss-of-function mutation of eat-4 or glc-3 similarly augmented the AIA Ca²⁺ response to 100 μM Cu²⁺. The eat-4 reconstitution in ASHs, ADLs, or both neurons in eat-4(ky5) and glc-3 in AIAs in glc-3(ok321) restored the wild-type phenotype (Figures 4K, 4L, and S6C). The eat-4; glc-3 double mutant phenocopied with a single mutant of eat-4 or glc-3. As expected, the dual rescue of eat-4 in each type of sensory neuron and glc-3 in AIAs, neither eat-4 in ASHs nor ADLs nor glc-3 in AIAs, restored wild-type Cu²⁺-induced AIA Ca²⁺ signals (Figures 4M and S6D). The above results suggest that sensory ASH and ADL neurons release the neurotransmitter glutamate in a graded manner to inhibit AIA interneurons in response to Cu²⁺ stimulation. Thus, the removal of AIA inhibition (disinhibition) initiates Cu²⁺-evoked hyperactive egg laying and HSN activity.

AIAs and HSNs chemosynaptically connect with each other (Cook et al., 2019; https://www.wormatlas.org; White et al., 1986). AIAs release acetylcholine, a classic neurotransmitter, and neuropeptides may include FLP-1, FLP-2, and INS-1 (Altun-Gultekin et al., 2001; Chalasani et al., 2010; Li and Kim, 2008). HSNs express two acetylcholine receptors: a non-alpha subunit ACR-14 and a G protein-linked receptor GAR-2 (https://wormbase.org; https://www.wormatlas.org; Rand, 2007). However, it is unknown whether HSNs express neuropeptide receptor(s). Therefore, we focused on acetylcholine signaling using mutants of acetylcholine biosynthesis and acetylcholine receptors. CHO-1 is the only high-affinity vesicular choline transporter in C. elegans (Matthies et al., 2006; Okuda et al., 2000). The cho-1(tm373) fully phenocopied the AIA::TeTx transgene phenotypes (Figures 5A–5C and S7A). Both strains displayed maximum egg-laying and HSN Ca²⁺ signals under standard culture conditions. The Cu²⁺ challenges did not further augment either activity. The genetic rescue of cho-1 in all of its expression neurons and AIAs (with the gcy-28(d) promoter), not in other single types of cho-1-expressing neurons, fully restored wild-type phenotypes (Figures 5A–5C, S7A, and S8). These results indicate that AIAs inhibit HSN activity, suppress egg laying, and mediate Cu²⁺-induced augmentation of egg laying via acetylcholine signaling.

Figure 5.

Acetylcholine Receptor ACR-14 Mediates AIA Inhibition on HSN Activity and Egg-Laying in Worms Treated without and with Cu²⁺

(A–L) The number of the eggs laid per worm per 30 min (egg-laying rate, A, D, G, and J), percent of the eggs at early developmental stages (both 1–8 cells and 9–20 cells, B, E, H, and K) and HSN somal calcium signals (C, F, I, and L) in the worms of indicated genotypes treated without (in black) and with (in red) 100 μM CuSO₄. ΔF = F - F_B. F, the average fluorescence intensity of the region of interest (ROI) of an HSN soma in each frame; F_B, a background signal defined as the average fluorescence intensity of ROI adjacent to the HSN soma in all frames.

All Data are displayed as box plots with each dot representing the data from each individual tested worm. Statistical significance of difference was analyzed by two-way ANOVA analysis with the post hoc test of Tukey's multiple comparison correction and indicated as follows: ns = not significant, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001, and in different colors for varied comparisons. Black, a comparison of tested worms with the control (wild type N2 worm) under standard culture conditions; red, that of tested animals with the control under the Cu²⁺ exposure; blue, that of Cu²⁺-untreated with Cu²⁺-treated animals of the same genotype; green, that as indicated.

See also Figure S7, S8, and Table S1.

An acetylcholine receptor mutant of gar-2 showed a partial defect in the inhibition of egg laying (Bany et al., 2003). We next identified cholinergic receptor(s) that may be involved in the regulation of egg laying and HSN activity. Our results showed that the acetylcholine receptor ACR-14 mutant acr-14(ok1155) displayed similar phenotypes to those in cho-1(tm373), while gar-2(ok520) exhibited wild-type phenotypes (Figure S8). Genetically rescuing acr-14 in all its expression neurons and HSNs (with egl-6(a) promoter) but not in other single types of acr-14-expressing neurons restored wild-type phenotypes (Figures 5D–5F, S7B, and S8). Moreover, the cho-1(tm373); acr-14(ok1155) double mutant exhibited very similar phenotypes to those in the single mutant of cho-1(tm373) or acr-14(ok1155), and the phenotypic defects were eliminated only by double genetic rescue of both cho-1 in AIAs and acr-14 in HSNs (Figures 5G–5I and S7C). These results indicate that AIAs inhibit HSNs and thus inhibit egg laying via the acetylcholinergic ACR-14 receptor.

We further generated glr-5(tm3506); acr-14(ok1155) double mutants to test the functions of GLR-5 and ACR-14 signaling in regulating HSN activity and egg laying under different conditions. The glr-5; acr-14 double mutant differed in phenotypes from cho-1(tm373), acr-14(ok1155), and cho-1(tm373); acr-14(ok1155) mutants (Figure 5J–5L and S7D). The latter three mutants almost entirely phenocopied with each other. The glr-5; acr-14 mutant displayed augmented but not maximal activities in egg-laying and HSNs under standard culture conditions. The Cu²⁺ treatments further increased both events. In genetically rescued transgenes, the reconstitution of either glr-5 or acr-14 alone or both genes in HSNs was enough to restore WT phenotypes under standard culture conditions. However, the dual rescue of both glr-5 and acr-14 in HSNs was necessary and sufficient to recover wild-type activities under Cu²⁺ treatments. Notably, HSN::acr-14; glr-5; acr-14 transgenic worms displayed maximal responses of egg laying and HSN activity to Cu²⁺ treatments, as did glr-5; acr-14 double mutant, while HSN::glr-5; glr-5; acr-14 transgenic animals lost behavioral and cellular responses to Cu²⁺ stimulation (Figures 5J–5L and S7D).

The above results support the following conclusions. Under standard culture conditions, ASHs and AIAs inhibit both HSN activity and egg laying redundantly via the GLR-5 and ACR-14 signaling pathways. In contrast, upon Cu²⁺ exposure, ASHs release glutamate at high levels to inhibit HSNs and AIAs via GLR-5 and GLC-3 receptors. ADLs only act on AIAs via high-concentration glutamate and the GLC-3 receptor. Therefore, ASHs/ADLs remove AIA inhibition on egg laying and HSN activity. Thus, ASHs/ADLs, AIAs, and HSNs form a disinhibitory circuit. Disinhibition, a model of neurosignal integration, provides the basis for the function of this circuit.

Glutamatergic Receptors GLR-5 and GLC-3 with Distinct Sensitivities to Ligand Decode Intensities of Sensory Signals from ASHs and ADLs under Different Conditions

Our above results indicate that sensory ASH and ADL neurons modulate egg-laying behavior differently. ASHs downregulate and upregulate egg laying under standard culture conditions and Cu²⁺ exposure, respectively. In contrast, ADLs only regulate egg laying under Cu²⁺ exposure. Two glutamatergic receptors and two circuits are involved in the regulation of egg laying and HSN activity, specifically the GLR-5 and GLC-3 receptors, ASHs to HSNs inhibitory connection via GLR-5, and the disinhibitory circuitry consisting of ASHs/ADLs, AIAs, and HSNs. Previous studies report that two different receptors of a neurotransmitter decode the intensity of sensory signals and direct varied behaviors under different environmental conditions. Two glutamate receptors, GLR-1 and GLR-5, on AIB interneurons decode the intensity of sensory information and yield different behavioral outputs. GLR-1, exhibiting a low activation threshold (high sensitivity) and fast kinetics, functions in reversal initiation induced by quinine at low concentrations. GLR-5, which possesses a higher activation threshold (low sensitivity) and more sustained kinetics, acts in pumping inhibition by quinine at high levels (Zou et al., 2018). The octopaminergic receptors SER-3 and SER-6 (with high and low sensitivities; without direct evidence) function in two distinct neurons and circuits (Liu et al., 2019). These receptors augment pharyngeal pumping and maintain basal pumping and 5-HT production under the conditions of food supply and food deprivation (Liu et al., 2019).

The results of the present study show that the glutamatergic GLR-5 receptor functions to inhibit HSN activity and egg laying under standard culture conditions. Under this state, ASHs release low levels of glutamate. In contrast, the GLC-3 receptor primarily serves to inhibit AIA activity and augment HSN activity and egg laying under Cu²⁺ challenge that stimulates ASH and ADL sensory neurons to release high-level glutamate. Based on these results, it is reasonable to deduce that the two receptors may have different sensitivities to glutamate. GLR-5 and GLC-3 should be strongly and weakly sensitive, respectively. We subsequently used robust egg-laying and HSN activity assays to (indirectly) test this working hypothesis.

First, we challenged wild-type N2 animals with glutamate at various concentrations under standard culture conditions. Our results show that the egg-laying rates and HSN Ca²⁺ transients in WT worms dose-dependently increased in response to exogenous glutamate of different doses, which were accurately fitted by the Hill function with EC₅₀ = 29.13 μM and 23.66 μM, respectively (Figures 6A, 6B, and S9A). The Hill responses suggest that there is only one glutamate effector, one glutaminergic receptor, or one functional site under this state. The egg-laying rates linearly correlated with HSN Ca²⁺ signals (Pearson r = 0.9818); however, we performed two tests under distinct conditions. In addition, in WT worms, the intensities of AIA Ca²⁺ transients dose-dependently decreased upon stimulation with glutamate at various concentrations (accurately fitted by the Hill function with IC₅₀ = 23.53 μM, Figures 6C and S10A). The egg-laying rates and HSN Ca²⁺ signals negatively and linearly correlated with AIA Ca²⁺ signals (Pearson r = −0.9722 and −0.9962, respectively). Moreover, Ca²⁺ transients in AIAs decreased upon intensive chemogenetic activation of ASHs (with capsaicin at 1 μM, Figure S11).

Figure 6.

Bidirectional Regulation of Egg-Laying and HSN Activity in C. elegans by the Neurotransmitter Glutamate

(A and B) Hill plots of the egg-laying rates (A) and the HSN somal Ca²⁺ signals (B) in wild type N2 worms treated with exogenous glutamate at various concentrations (0, 0.1, 0.5, 5, 20, 50, 100, and 300, in μM).

(C) Curves and Hill plot of the AIA axonal Ca²⁺ signals in WT worms treated with exogenous glutamate at various concentrations (in μM: 0, 0.1, 0.5, 5, 20, 50, 100, and 300).

(D and E) Two-binding-sites Hill plots of the egg-laying rates (D) and the HSN somal Ca²⁺ signals (E) in eat-4 mutant worms treated with exogenous glutamate at various concentrations (in μM: 0, 0.005, 0.05, 0.1, 0.3, 0.5, 1, 5, 10, 20, 50, 100, 200, and 500).

(F) Curves and Hill plot of the axonal AIA Ca²⁺ signals in eat-4 mutant worms treated with exogenous glutamate at various concentrations (in μM: 0, 0.05, 0.1, 1, 5, 10, 20, 50, 100, 200, and 500).

(G and H) Hill plots of the egg-laying rates (G) and the HSN somal Ca²⁺ signals (H) in the indicated worms treated with exogenous glutamate at various concentrations (in μM: 0, 0.001, 0.01, 0.05, 0.1, 0.5, 1, 5, and 10).

(I and J) Hill plots of the egg-laying rates (I) and the HSN somal Ca²⁺ signals (J) in the indicated worms treated with exogenous glutamate at various concentrations (in μM: 0, 0.1, 0.5, 5, 20, 50, 100, and 300).

(K and L) Hill plots of the egg-laying rates (K) and the HSN somal Ca²⁺ signals (L) in acr-14 mutant worms treated with exogenous glutamate at various concentrations (in μM: 0, 0.001, 0.01, 0.05, 0.1, 0.5, 1, 5, and 10).

All data are represented as means ± SEM, as indicated by dots or solid traces ± error bars or gray shading. See also Figures S9–S11 and Table S1.

Second, we used eat-4(ky5) to perform the same set of tests. Interestingly, in contrast to WT worms, the mutant displayed double-Hill responses of egg-laying rates (Figure 6D) and HSN Ca²⁺ signals (Figures 6E and S9B) to exogenous glutamate treatments, i.e., inhibitory and excitatory reactions in ranges of low and high concentrations, respectively. Two responses were well fitted by a two-binding-site Hill function with IC₅₀ and EC₅₀ values of 0.17 and 34.32 for egg-laying rates and 0.11 and 28.1 μM for the HSN Ca²⁺ transients, respectively, suggesting two effectors (two receptors or functional sites) of glutamate. The egg-laying rates linearly correlated with HSN Ca²⁺ signals (Pearson r = 0.9835). However, in this mutant, AIA Ca²⁺ signals were only well fitted by the single-binding-site Hill function, with IC₅₀ = 27.48 μM (Figures 6F and S10B). Notably, the IC₅₀ for AIA responses was close to the EC₅₀ for HSN reactions (28.1 μM). The EC₅₀ and IC₅₀ values in the eat-4 mutant worms were higher than those in WT worms. A reasonable explanation is that the glutamate levels in wild-type N2 animals under standard culture conditions should be a few micromolar (approximately 4.5 μM). AIA Ca²⁺ signals correlated negatively and linearly with egg-laying rates and HSN Ca²⁺ signals (Pearson r = −0.9674 and −0.9838, respectively). The difference in the dependence on glutamate between WT worms and the eat-4 mutant, combined with previous results, suggests that GLR-5 and GLC-3 not only function differently but are also distinctly sensitive to the neurotransmitter glutamate. GLR-5 is highly sensitive to glutamate and may be saturated by low-level glutamate, even under standard culture conditions, in wild-type N2 worms.

Third, we employed the eat-4; glr-5 double mutant and transgenes to further evaluate the kinetics of GLR-5 and GLC-3 receptors indirectly, as we did previously. As expected, the double mutant lost the responses to glutamate challenges in egg laying and HSN Ca²⁺ signals. GLR-5 reconstitution in the HSN::glr-5; eat-4; glr-5 transgene enabled worms to respond to glutamate with an inhibitory Hill relationship of IC₅₀ = 0.22 μM and 0.17 μM for egg laying and HSN activity, respectively (Figures 6G and 6H). The IC₅₀ values were comparable to those in the eat-4 mutant (0.17 μM and 0.11 μM, Figures 6D and 6E). Two actions were linearly correlated (Pearson r = 0.9840) in this transgene.

Fourth, we reconstituted glc-3 in the glc-3(ok321) mutant for the next test. This mutant also lost responses to glutamate treatments in egg-laying and HSN Ca²⁺ transients. GLC-3 expression in AIAs restored both reactions in the transgene. The EC₅₀ values of Hill responses in egg laying and HSN activity were 32.05 μM and 25.75 μM, respectively, and comparable to those in N2 worms (29.13 μM and 23.66 μM, Figures 6A and 6B). Two reactions linearly correlated (Pearson r = 0.9844) in this transgene (Figures 6I and 6J, and S9D).

Fifth, as shown in Figures 6K, 6L and S9E, the acr-14(ok1155) mutant displayed sole inhibitory Hill responses to glutamate treatments, with IC₅₀ values of 0.13 and 0.08 μM for egg-laying rates and HSN Ca²⁺ responses, respectively, which were also linearly correlated (Pearson r = 0.982). Interestingly, among eat-4(ky5), HSN::glr-5; eat-4; glr-5 and acr-14(ok1155) worms, the acr-14 mutant possessed the lowest IC₅₀ values, suggesting that AIAs may inhibit HSNs and egg laying tonically.

In summary, our experiments suggest that glutamatergic receptors GLR-5 and GLC-3 have different affinities to glutamate, although biochemical tests for binding kinetics are still needed. The apparent IC₅₀ and EC₅₀ values of these receptors to glutamate are approximately 0.1 μM and 30 μM (estimated by an indirect method), respectively. The sensitivity difference of the two receptors enables decoding the intensities of sensory signals glutamate and translating them into different behavioral outputs through a single neurotransmitter.

Discussion

In C. elegans, egg-laying behavior is among the best-characterized behaviors and has been used not only for behavioral studies but also for the genetic analysis of nervous system function and development (Schafer, 2005, 2006; Zhang et al., 2008). Worms maintain a relatively constant number of unlaid eggs in the uterus (McCarter et al., 1999). Egg laying occurs through the contraction of vulval muscles. Electrically coupled vm2s, which are the only egg-laying muscles receiving significant synaptic input from two HSN and six VC motor neurons (Cook et al., 2019; White et al., 1986), are essential for opening the vulva (Schafer, 2006). Most egg laying occurs in short bursts lasting approximately 1–2 min, which represents an active phase. A more extended quiescent period averaging approximately 20 min in duration separates the active phases (Collins et al., 2016; Collins and Koelle, 2013; Schafer, 2005, 2006; Trent et al., 1983; Waggoner et al., 1998; Weinshenker et al., 1995). Serotonin, released primarily from the HSNs and, to a lesser extent, from VC4 and VC5, facilitates the onset of the active phase, while HSNs and VCs release acetylcholine to trigger individual muscle contractions within the active phase (Waggoner et al., 1998). HSN neurons release serotonin and the neuropeptide NLP-3, which act as partially redundant cotransmitters to initiate egg laying (Brewer et al., 2019). HSNs are postsynaptic to sensory neurons (ASHs, ASIs, ASJs, AIMs, and PLMs), interneurons (AIAs, AIYs, AVBs, AVFs, AVJs, BDUs, PVQs, PVNs, RIFs, and RIR), and VC motor neurons (Cook et al., 2019; https://www.wormatlas.org; White et al., 1986). HSNs also receive humoral regulation from vulval muscle cells in modulating the development of the two-state patterned activity, i.e., alternate active and quiescent periods (Ravi et al., 2018; Schafer, 2005, 2006). Sensory information from the touch receptor ALM and PLM neurons (Zhang et al., 2008), chemosensory ASIs, ASJs (Liu et al., 2018), and AWCs (Fenk and de Bono, 2015) inhibits HSN activity and egg laying. The results of this study help to elucidate the regulation of egg laying by environmental stress.

In the present study, we found that short-term noxious exposure to metal ions, such as Cu²⁺, Ni²⁺, Mn²⁺, and Fe³⁺, evokes hyperactive egg laying in C. elegans. We have dissected the molecular and neurocircuit mechanisms underlying ASH and ADL modulation of HSN activity and egg laying under varied environmental conditions. Our results suggest a working model in which sensory ASH and ADL neurons sense Cu²⁺ stimulation through OCR-2/OSM-9 signaling (Figure 7). Under standard culture conditions, ASHs tonically release a small amount of neurotransmitter glutamate to inhibit HSN activity and egg-laying behavior via GLR-5 signaling in HSNs. Under Cu²⁺ treatment, ASHs and ADLs release a large amount of glutamate to regulate egg laying via two circuits: the ASH-HSN inhibitory circuit via the highly sensitive GLR-5 receptor and the disinhibitory ASH/ADL-AIA-HSN circuit. In the latter circuitry, ASHs and ADLs release high levels of glutamate and thus inhibit AIA interneurons via the low sensitivity of the GLC-3 receptor, consequently relieving AIA inhibition on HSNs via ACR-14 signaling. This circuitry predominates over the former circuitry in this activity (Figure 7). Our study indirectly supports that the graded release of neurotransmitter glutamate encodes sensory signals from ASHs and ADLs. Two receptors, GLR-5 and GLC-3, with distinct sensitivities to ligands, encode sensory signals and induce different behavioral outputs under varied environmental conditions.

Figure 7.

The Molecular and Neurocircuital Mechanisms Underlying the Opposite Modulation of Egg-Laying under Standard Culture Conditions and Exposure to Cu²⁺ in C. elegans

Under standard culture conditions, sensory neuron ASHs tonically release a small amount of glutamate (red dots) to inhibit (red line) HSN motor neurons via GLR-5 signaling. In contrast, AIA interneurons release high-level acetylcholine (magenta dots) to suppress HSNs via ACR-14 signaling. Under noxious stimulation of Cu²⁺ (blue dots), ASHs and ADLs release a large amount of glutamate to inhibit AIAs via GLC-3 signaling and thus remove (green lines with arrows) HSN depression by AIAs (disinhibition). ASHs, but not ADLs (no chemosynaptic connection with HSNs), suppress activity in HSNs via GLR-5 receptor under the Cu²⁺ challenge. However, the action of disinhibitory ASH/ADL-AIA-HSN predominates over ASH inhibition on HSNs. The azure arrows indicate chemosynapses. The width of arrows and lines shows the intensities of synapses and actions, including inhibition and excitation. The different colors indicate the power of cellular activity and actions: undertone and deep colors, weak and robust activities, respectively; gray, no evident activity or action.

Copper (Cu) is an essential element for animals and plants; however, increased uptake caused deleterious effects in the normal functioning of organisms, with a narrow difference amid the indispensable and detrimental concentrations. The typical values of Cu in uncontaminated soils varied from 2 to 109 mg kg⁻¹. The human industrial and agricultural activities, including mining, refinery, fossil fuel combustion, waste incineration, traffic, fertilizers, and soil amendments, cause copper contamination in the environment (Kumar et al., 2020; Malhotra et al., 2020). Exposure of even low-level Cu (down to 0.05 μg/L) caused a hatching delay in Zebrafish Brachydanio rerio (Dave and Xiu, 1991). The long-term exposure of excessive Cu decreases successful hatching and suppresses the development of eggs in pulmonate snail, Physa acuta (Gao et al., 2017), and reduces C. elegans brood size and lead to internal hatching of offspring inside their mothers (bagging) (Calafato et al., 2008). High level of Cu deteriorates hatching success and hatched juvenile survival in antarctic terrestrial nematode Plectus murrayi (Brown et al., 2020). Here, we find that short-term Cu exposure augments egg laying in C. elegans. This observation argues for the life-history theory, which argues that when organisms experience environmental conditions that can reduce their long-term reproductive success, they increase their present reproductive output.

AIA interneurons inhibit HSN activity and egg laying. Nociceptive ASHs and ADLs inhibit AIA activity under noxious exposure to Cu²⁺ ions (this study). AIAs are first-layer amphid interneurons; they are postsynaptic to 12 cells: sensory ASK, ASH, ASG, ADL, ASE, AWC, and ASI; and they form gap junctions with ASI, AWA, and ADF neurons (https://www.wormatlas.org). AIAs integrate information from amphid sensory neurons and regulate multiple behaviors. The ablation of AIAs shortens the duration of forward locomotion (Gray et al., 2005; Tsalik and Hobert, 2003; Wakabayashi et al., 2004). AIAs control foraging behavior (Calhoun et al., 2015; Lopez-Cruz et al., 2019). These neurons integrate sensory information of pathogen P. aeruginosa from ASHs/ASIs and mediate avoidance and immune response to the pathogen (Cao et al., 2017). AIAs integrate signals from ASK (via inhibitory synapses) and other sensory neurons and play essential roles in attraction to pheromone ascarosides and innate social behavior (Macosko et al., 2009; Srinivasan et al., 2012). AIAs receive excitatory and inhibitory inputs from AWAs and ASHs/ADLs (via GLC-3 signaling), respectively, and regulate the behavioral choice between conflicting alternatives when worms are facing conflicting sensory cues, such as diacetyl and Cu²⁺ (Shinkai et al., 2011). AIAs receive excitatory inputs from ASER and provide feedback to ASER to generate chemotaxis learning (Tomioka et al., 2006). However, AIAs are unlikely to be essential for egg-laying suppression by 5% CO₂ (Fenk and de Bono, 2015).

The results of the present study indicate that AIAs inhibit HSN activity and egg-laying behavior via acetylcholine/ACR-14 signaling. Based on the synaptic connections and functions of AIA interneurons, we suggest that AIAs function as a primary hub for integrating multiple sensory cues to modulate egg laying. ACR-14, a non-α subunit of the ACR-16 group, is categorized by homologs (Jones and Sattelle, 2004; Mongan et al., 1998; Nassel, 2018); however, to the best of our knowledge, ACR-14 is not involved in nicotine responses in C. elegans, including nicotine approach (Sellings et al., 2013), withdrawal, and sensitization (Feng et al., 2006). The signal transduction pathway employed by ACR-14 warrants further study.

Sensory signal encoding, decoding, and behavioral translation are fundamental issues in neuroscience. The nervous system encodes environmental stimuli through the type (vision, audition, taste, smell, balance, touch, proprioception, temperature sense, pain, or itch) and the quality (spatial, position, intensity, or frequency) of a stimulus. Sensory neurons may release different types of neurotransmitters or graded amounts of a neurotransmitter in response to different stimulation conditions to encode sensory signals (see reviews by Burnstock, 2004; Nassel, 2018; Nusbaum et al., 2017). The receptors of neurotransmitters or neuromodulators decode sensory cues, and neural circuits integrate neuroinformation and then translate it into behavioral outputs.

In C. elegans, different neurotransmitter receptors decode sensory information from polymodal ASH sensory neurons and drive different avoidant behaviors (Hart et al., 1995; Maricq et al., 1995). Moreover, distinct receptors of a single neurotransmitter decode sensory intensity encoding by the amount of a single neurotransmitter. For instance, both glutamatergic NMDA (NMR-1) and non-NMDA (GLR-1 and GLR-2) receptor subunits mediate the osmotic avoidance response, while non-NMDA receptors mediate the response to mechanical stimuli (Mellem et al., 2002). The neuromodulator serotonin acts directly on vulval muscles via G_qα signaling to increase the frequency of Ca²⁺ transients and worm egg-laying rate. However, serotonin inhibits spontaneous activity in HSNs via G_oα signaling (Shyn et al., 2003). Two glutamate receptors, GLR-1 and GLR-5, encode sensory signals of quinine at different concentrations. GLR-1 is highly sensitive and is rapidly inactivated; in contrast, GLR-5 is less sensitive and is slowly inactivated. These two receptors on AIB interneurons decode low and high concentrations of quinine and translate into reversal initiation and feeding suppression, respectively (Zou et al., 2018). SER-3 and SER-6, strongly and weakly sensitive receptors, possibly decode varied neuroinformation (graded octopamine release) from Ring interneuron (RICs) under food supply and food deprivation. SER-3 responds to octopamine at low levels to augment pharyngeal pumping via an ADF–RIC–SIA feedforward circuit under abundant supply. SER-6 mainly works at high levels of octopamine via a disexcitatory ADF-RIC-AWB-ADF feedback circuit to maintain basal pumping and serotonin production in ADFs under food deprivation (Liu et al., 2019).

In this study, we indirectly detected that the graded release of neurotransmitter glutamate encodes sensory information from ASHs and ADLs. Highly sensitive GLR-5 receptors on HSNs decode low-intensity sensory signals from ASHs under standard culture conditions. These receptors translate neuronal information into the inhibition of egg laying by directly inhibiting HSN activity. However, low-sensitivity GLC-3 receptors on AIAs decode high-intensity signals from ASHs and ADLs under noxious Cu²⁺ administration and translate neuroinformation into behavioral augmentation by subtracting AIA suppression on HSNs. At the same time, ASHs still inhibit HSNs directly. Using behavioral assays, Ca²⁺ imaging, and pharmacological tests, we indirectly estimated the apparent dose-dependent IC₅₀ and EC₅₀ of the two receptors. GLR-5 and GLC-3 are strongly and weakly sensitive to glutamate, respectively.

Limitations of the Study

The results in the study support that varied amounts of neurotransmitter glutamate are released under different conditions, and two glutamatergic receptors have different sensitivities to the neurotransmitter. However, direct measurement of neurotransmitters released and assays of the binding kinetics of receptors with the ligand are required to support the conclusions directly.

Resource Availability

Lead Contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Zheng-Xing Wu (ibbwuzx@mail.hust.edu.cn).

Materials Availability

All worm strains and plasmids used in this study are available to scientist's community. Please email the Lead Contact.

Data and Code Availability

The published article includes all data generated or analyzed during this study.

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.

Published: October 23, 2020