Nicotine increases reversal frequency in C. elegans

To investigate the behavioral effects of nicotine, we developed a brief chronic nicotine exposure assay in which young adult hermaphrodites were incubated with 1 mM nicotine for 3 h (Fig 1A; Materials and methods). Following exposure, animals were transferred to nicotine-free Nematode Growth Medium (NGM) agar plates seeded with a thin layer of Escherichia coli OP50 for behavioral analysis. Under these conditions, worms exhibited a robust nicotine-dependent behavioral response [24].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Nicotine increased the coupling of DMP and reversals.

(A) An assay diagram of response to nicotine exposure and withdrawal in Caenorhabditis elegans. Young adult hermaphrodites were used. (B) Representative 3-min locomotion trajectories of two animals on withdrawal plates without (− nic) and with (+ nic) nicotine exposure. (C) Coupling time diagrams for reversal and defecation motor program (DMP). Upper, representative speed traces labeled with the moment of DMP setup. Gray bar: DMP not couple with reversal; Orange bar: DMP couples with reversal. Bottom, raster plots of 10 consecutive DMP cycles that coupled with or without reversal in 10 animals. (D) Quantification of the DMP-reversal coupling proportion. Nicotine, but not glutamate, significantly enables the coupling of DMP to reversal. Two-way ANOVA was performed (interaction: F_{(1, 37)} = 225.9, P < 0.0001). (E) 100 segmented DMP-associated movements, each shown for 30 s. The 0-s time point represents the initiation of the DMP and serves as the alignment reference. The speed of 10 s before and 20 s after DMP initiation were used. The figure below shows the average speed of the 100 segments. (F) The dose dependence of nicotine (3 h) for the coupling of DMP and reversal. (G) The time dependence of nicotine (1 mM) for the coupling of DMP and reversal. (H) The failure time dependence of nicotine. n ≥ 10 animals. ns, no significance, *** p < 0.001 by Two-way ANOVA analysis. Error bars, SEM. The data underlying this figure can be found in S1 Data.

https://doi.org/10.1371/journal.pbio.3003423.g001

Compared with nicotine-naïve controls (− nic), nicotine-exposed worms (+ nic) displayed disrupted forward locomotion, characterized by discontinuous crawling trajectories, markedly increased reversal frequency, and elevated locomotion speed (Fig 1B). The most prominent difference was the substantial increase in reversal frequency and movement velocity. Whereas spontaneous backward movements in control worms were brief and sporadic, nicotine-exposed animals exhibited reversals that were highly rhythmic, significantly prolonged, and executed at faster speeds (S1A Fig). Consequently, the proportion of time spent moving backward was markedly greater in the nicotine-treated group (S1B Fig). Quantitatively, reversal frequency increased nearly 3-fold (− nic, 3.3 ± 0.34/3 min, n = 22; + nic, 9.3 ± 0.46/3 min, n = 22, p < 0.001) (S1C Fig), reversal duration was extended more than 4-fold (− nic, 1 ± 0.08 s, n = 22; + nic, 4.7 ± 0.27 s, n = 22, p < 0.001) (S1D Fig), and reversal speed was significantly elevated (− nic, 0.09 ± 0.003 mm/s, n = 22; + nic, 0.12 ± 0.003 mm/s, n = 22, p < 0.001) (S1E Fig). Meanwhile, forward locomotion speed was also enhanced (− nic, 0.11 ± 0.003 mm/s, n = 22; + nic, 0.13 ± 0.004 mm/s, n = 22, p < 0.001) (S1F Fig). These high-frequency, prolonged, high-speed reversals resemble nicotine-induced anxious or aversive behaviors in mammals [25,26].

Nicotine-enhanced reversals are temporally coupled with the defecation motor program

Nicotine-induced behavior changes have been reported across multiple model organisms [26,27]. Here, we observed that nicotine-evoked high-frequency reversals in C. elegans exhibited a striking rhythmicity. These reversals occurred approximately every 45–50 s, prompting us to consider whether they might be linked to an intrinsic rhythmic process. We analyzed locomotion over 10 consecutive cycles of the DMP in wild-type animals. In controls, ~ 20% of DMP cycles were followed by short backward movements (Fig 1C and 1D and S1 Movie), consistent with previous reports of DMP-locomotion coupling [28]. By contrast, nicotine drove nearly 100% of DMP cycles to trigger reversals immediately following DMP onset, indicating robust DMP-reversal coupling (Fig 1C and 1D and S2 Movie). This suggested that nicotine’s effect arises from modulation of intrinsic rhythmic signaling, rather than simply producing general excitatory drive.

To distinguish between these possibilities, we examined the effect of glutamate—another neurotransmitter implicated in reward regulation [29] and nicotine dependence [10]—under identical conditions. Glutamate exposure did not alter the proportion of DMP-reversal coupling (Fig 1D). Furthermore, although nicotine affects gastrointestinal motility in mammals [30], the DMP cycle length in C. elegans remained unchanged after nicotine exposure (S2A–S2C Fig). These results suggest that nicotine-induced motor coupling does not stem from altered gut rhythmicity or nonspecific excitation.

Interestingly, nicotine-enhanced couplings showed <5 s delays between DMP onset and reversal initiation, in contrast to previous observations in which reversals typically occurred ~10 s before DMP [28]. The DMP consists of three sequential phases: posterior body wall muscle contraction (pBoc), followed by anterior body wall muscle contraction (aBoc), and the final enteric muscle expulsion (Exp) [31] (S2A Fig). To determine whether nicotine alters the coupling temporal relationship between DMP phases and reversal, we categorized reversal onset times relative to these phases and found that nicotine-driven reversals occurred predominantly during pBoc (S2D Fig).

We next assessed coupling in mutants with phase-specific DMP defect. In egl-8 mutants (defective pBoc), nicotine-enhanced coupling was strongly reduced, whereas nlp-40 and aex-2 mutants (impaired Exp) [31,32] showed no significant change (S2E Fig). aex-3 mutants, defective in both aBoc and Exp [31,33], exhibited partial reduction, but less than egl-8. These results indicate that intact pBoc activity is critical for nicotine-enhanced motor coupling, consistent with nicotine amplifying an interoceptive response to body wall contraction rather than altering DMP timing. In agreement, the delay from DMP onset to reversal initiation was unchanged by nicotine (S2F Fig).

To more precisely quantify the timing, we aligned all DMP events to the moment of minimal body length (time 0) and extracted locomotion speed traces spanning 10 s before to 20 s after this point (Fig 1Ei). This alignment allowed us to precisely synchronize locomotion speed at the moment of DMP, facilitating a comprehensive analysis of the locomotion characteristics associated with 100 DMP events across 10 worms. By aligning the data in this way, we ensured that variations in DMP cycle timing did not obscure the underlying coupling patterns, thereby improving the accuracy and reliability of our analysis. We found that nicotine induced a pronounced backward movement immediately following DMP compared to controls (Fig 1Eii).

Nicotine-enhanced DMP-reversal coupling was both concentration and exposure-time-dependent. With a fixed 3-h exposure, coupling increased with nicotine concentration, yielding a half-maximal effective concentration (EC₅₀) of 0.34 mM (Fig 1F). At a fixed saturated dose (1 mM), coupling increased with exposure duration, with a half-maximal effect at ~15.2 min (Fig 1G). Furthermore, similar to how nicotine induces a reversible aversive reaction in humans [34], nicotine-enhanced coupling was reversible: after transfer to nicotine-free plates, coupling gradually returned to baseline (~20%) with a half-life of ~1.2 h (Fig 1H).

Given that DMP is an intrinsic bodily rhythm, these findings suggest that nicotine acts as an interoceptive modulator, selectively enhancing sensory-motor linkage between internal rhythmic contractions and reversal behavior [35–37]. This prompted us to investigate the molecular and circuit-level mechanisms underlying this phenomenon.

Nicotine-induced abnormal motor coupling requires the nicotinic acetylcholine receptor ACR-16

Previous studies have shown that nicotine-dependent behaviors require nAChRs [5,38], which mediate responses to the neurotransmitter acetylcholine (ACh) [39,40]. To test whether nAChRs mediate nicotine’s effects in our assay, we pharmacologically blocked these receptors using DHβE, a competitive antagonist of nAChRs [40]. DHβE treatment significantly suppressed nicotine-enhanced DMP-reversal coupling (Fig 2A). Consistently, other nicotine-induced motor behaviors—including increased reversal frequency (S3A Fig), prolonged reversal duration (S3B Fig), and enhanced reversal and forward speeds (S3C and S3D Fig)—were also inhibited. In contrast, a cocktail of glutamate receptor antagonists, DNQX and MK-801, which block ionotropic excitatory glutamate receptors [41,42], did not affect the coupling proportion (Fig 2B). These pharmacological results provide strong evidence that nicotine’s action in C. elegans depends on nAChRs rather than glutamate receptors.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Nicotine enhanced DMP-reversal coupling requires ACR-16 in AVA.

(A) Suppression of nicotine enhanced DMP-reversal coupling by nAChR antagonist DHβE (20 µM). Two-way ANOVA was performed (interaction: F_{(1, 39)} = 275.6, P < 0.0001). (B) The GluR antagonists DNQX (500 µM) and MK-801 (50 µM) do not affect nicotine-enhanced DMP reverse coupling. Two-way ANOVA was performed (interaction: F_{(1, 36)} = 1.619, P = 0.2114). (C) Dendrogram of Caenorhabditis elegans nAChRs. A fraction of nAChRs from Drosophila, mouse, and humans were included, such as mouse α7. (D) Genetic screening identified a specific nAChR gene, acr-16, disrupts nicotine-enhanced DMP-reversal coupling. Two-way ANOVA was performed (interaction: F_{(21, 437)} = 12.95, P < 0.0001). (E) Expression pattern of acr-16. Up, endogenous acr-16 promoter driven GFP was observed in AVA command interneuron and body wall muscles. Scale bar, 20 µm. Bottom, the co-localization of Pacr-16::GFP and Ptwk-40s::RFP in AVA neuron. Scale bar, 10 μm. Worm orientation, A anterior, D dorsal. (F) Quantification of the rescue of DMP-reversal coupling by expression of ACR-16 in different tissues. AVA, AVE, etc. command interneurons, but not motor neurons or muscles, are the critical functional cells for ACR-16. Expression of mouse α7 nAChR restored the DMP-reversal coupling proportion in acr-16(ok789) mutants. Two-way ANOVA was performed (interaction: F_{(6, 129)} = 34.81, P < 0.0001). (G) Overexpression of ACR-16 in wild-type N2 increases the DMP-reversal coupling proportion even without nicotine exposure. Two-way ANOVA was performed (interaction: F_{(1, 70)} = 22.05, P < 0.0001). n ≥ 10 animals. ns, no significance, *** p < 0.001 by Two-way ANOVA analysis. # p < 0.05 by Student t test. Error bars, SEM. The data underlying this figure can be found in S1 Data.

https://doi.org/10.1371/journal.pbio.3003423.g002

To further validate this, we performed a genetic screen focusing on nAChR mutants. The C. elegans genome encodes over 29 nAChR subunits [43], many with available loss-of-function alleles (Fig 2C) [16,44,45]. Among these, only acr-16 mutants completely lost the nicotine-enhanced DMP-reversal coupling, with nicotine failing to induce motor coupling (Fig 2D and S3 and S4 Movies). Other uncoordinated mutants such as unc-29, unc-38, and unc-63 showed reduced coupling proportions (S3E Fig), although their severe locomotion defects may nonspecifically affect the behavior [46,47]. In agreement with the pharmacological data, mutants lacking ionotropic glutamate receptors like glr-1, glr-2, glr-4, glr-5, nmr-1, and nmr-2, displayed normal DMP-reversal coupling (S3F Fig).

Together, these findings demonstrate that nicotine-induced abnormal motor coupling specifically requires the nAChR subunit ACR-16 in C. elegans, highlighting its pivotal role in mediating nicotine’s modulatory effects on sensory-motor integration.

ACR-16 acts in command interneurons to mediate nicotine-induced motor responses

acr-16 encodes an α7-like nAChR broadly expressed in both the nervous system and body wall muscles [44,45,48]. To assess how acr-16 contributes to nicotine-induced motor behavioral changes in C. elegans, we examined its expression pattern and functional site of action. Using a fluorescence reporter driven by the endogenous acr-16 promoter, we detected expression in both neuronal and muscular tissues (Fig 2E). Within the nervous system, acr-16 is predominantly expressed in AVA and AVB command interneurons and in a subset of motoneurons, all of which are critical for regulating motor behaviors [45,49,50]. In mammals, nicotine’s behavioral effects are primarily mediated by neuronal, rather than muscle, nAChRs [40]. To pinpoint the neuronal site of acr-16 action, we performed cell-specific rescue experiments in acr-16 mutants. Restoring ACR-16 expression in the backward command interneurons AVA and AVE using the nmr-1 promoter fully rescued the nicotine-enhanced DMP-reversal coupling (Fig 2F). In contrast, expression in body wall muscles (myo-3 promoter), backward motor neurons (unc-4 promoter), or forward command interneurons such as AVB and PVC (sra-11 promoter) failed to restore nicotine responsiveness. These results indicate that ACR-16 acts primarily in the backward command interneurons to mediate nicotine-induced motor coupling.

Interestingly, overexpression of ACR-16 in AVA not only rescued the nicotine response in acr-16 mutants but also elevated baseline DMP-reversal coupling in the absence of nicotine (Fig 2F). This suggests that the expression level of ACR-16 in AVA may set the threshold for coupling. To test this, we overexpressed ACR-16 in AVA using the short twk-40s promoter [51] in wild-type N2 worms. Indeed, even without nicotine, these animals exhibited increased baseline coupling, and nicotine exposure drove the coupling proportion to 100% (Fig 2G).

Together, these findings demonstrate that ACR-16 mediates nicotine’s behavioral effects primarily via backward command interneurons, and that the strength of this response is strongly dependent on ACR-16 expression levels in these neurons.

The Mouse nAChR α7 subunit can functionally substitute for worm ACR-16

Given the essential role of ACR-16 in nicotine-dependent behavioral response in C. elegans, we asked whether the mammalian homolog could serve a similar function. ACR-16 shares high sequence similarity with the vertebrate α7 subunit of nAChRs (Fig 2C) [44,45]. To test cross-species functional compatibility, we expressed the mouse α7 nAChR in acr-16 mutant worms under the control of the endogenous acr-16 promoter. Remarkably, mouse α7 expression restored nicotine-enhanced DMP-reversal coupling in the mutants (Fig 2F). However, unlike worm ACR-16, the mouse α7 receptor did not increase baseline coupling proportion in the absence of nicotine (Fig 2F). These findings suggest that α7/acr-16-dependent nicotine responses are conserved between mammals and C. elegans.

Nicotine enhances AVA calcium activity linked to the DMP

To establish the physiological role of AVA interneuron in nicotine’s interoceptive effects, we measured real-time AVA activity in freely moving animals using the genetically encoded calcium (Ca²⁺) indicator GCaMP6s, with wCherry as a ratiometric reference [52] (Materials and methods). These AVA::GCaMP transgenic animals responded to nicotine similarly to wild-type N2 worms in terms of reversal frequency, duration, and locomotion speed (wild-type, S4 Fig). In parallel, we used an independent infrared imaging system to monitor the rhythmic defecation behavior in real time [53] (Materials and methods). AVA Ca²⁺ activity was recorded across five consecutive DMP cycles to assess their synchrony.

Following nicotine exposure, AVA displayed significantly elevated Ca²⁺ oscillation compared to nicotine-free conditions (Fig 3A). Both the frequency and amplitude of Ca²⁺ transients were markedly increased in the wild-type transgenic worms (Fig 3A—3D), indicating that nicotine enhances AVA excitability. This nicotine-induced AVA activation was strongly temporally coupled to DMP events. To precisely define this relationship, we aligned AVA Ca²⁺ traces to the shortest body length point of each DMP cycle, extracting activity from 10 s before to 20 s after this reference point (Fig 3A). This alignment allowed us to synchronize AVA activity with 100 DMP events across 20 worms. Under nicotine treatment, nearly every DMP event was accompanied by a robust AVA Ca²⁺ transient (Fig 3B and 3C), whereas in the absence of nicotine, DMP initiation elicited minimal AVA activity (Fig 3B). These results parallel the behavioral observation that nicotine enhances DMP-triggered reversals, suggesting that elevated AVA activity drives this coupling.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Nicotine enhances calcium activity in AVA.

(A) Left, A representative calcium (Ca²⁺) transient of AVA neurons over five consecutive DMP cycles from a freely behaving animal without nicotine treatment. Only one DMP triggers Ca²⁺ activity in AVA. Right, Nicotine significantly enhances spontaneous Ca²⁺ transient activity in AVA, with all five DMP cycles initiating Ca²⁺ activity in AVA. Green boxes and arrows indicate tight coupling between DMP and AVA Ca²⁺ transients, while gray boxes and arrows represent the absence of coupling. Dashed boxes highlight 30 s segments of AVA Ca²⁺ activity, with 10 s before and 20 s after DMP initiation. Arrows indicate the initiation moments of DMP. (B) The AVA Ca²⁺ activity from five consecutive DMP cycles, recorded from 20 individual worms, is shown for both nicotine-free (− nic) and nicotine-exposed (+ nic) conditions across different genotypes (n = 20 animals per genotype). The Ca²⁺ transients are aligned based on the DMP initiation moments (indicated by arrows and dashed lines). Numbers 1 through 5 correspond to the five consecutive DMP cycles. The average AVA Ca²⁺ response for each DMP is displayed below. (C) Quantitative analysis of the coupling percentage of AVA Ca²⁺ transient responses to DMP. Nicotine enhanced coupling proportion of DMP-AVA Ca²⁺ transients is dependent on ACR-16. Two-way ANOVA was performed (interaction: F_{(2, 114)} = 83.67, P < 0.0001). (D) Quantification of the peak amplitude of AVA Ca²⁺ transients that coupled with DMP. Nicotine increased the amplitude of AVA Ca²⁺ transients, which is dependent on ACR-16. Two-way ANOVA was performed (interaction: F_{(2, 277)} = 26.60, P < 0.0001). n = 20 animals in each genotype. ns, no significance, *** p < 0.001 by Two-way ANOVA analysis. Error bars, SEM. The data underlying this figure can be found in S1 Data.

https://doi.org/10.1371/journal.pbio.3003423.g003

Consistent with behavioral requirements, AVA Ca²⁺ activity was nearly absent in acr-16 mutants under nicotine exposure (Fig 3B–3D). In these mutants, nicotine failed to increase either the response probability or the peak amplitude of AVA transients. Cell-specific rescue of ACR-16 in AVA using the short twk-40s promoter fully restored both the frequency and amplitude of DMP-evoked Ca²⁺ transients (Fig 3B—3D). Notably, ACR-16 rescue also modestly increased baseline AVA activity in the absence of nicotine, mirroring the enhanced baseline DMP-reversal coupling seen in ACR-16-overexpressing worms. Furthermore, the ACR-16 dependence of nicotine-induced increases in reversal frequency, duration, and locomotion speed was confirmed in these rescue animals (S4 Fig). Together, these findings indicate that nicotine enhances motor coupling by elevating AVA activity in an ACR-16-dependent manner.

Nicotine upregulates functional expression of ACR-16

Having established that enhanced AVA activity is crucial for nicotine’s action and that this regulation depends on ACR-16, we next aimed to identify how ACR-16 mediates this activity. In mammals, chronic nicotine exposure is well known to up-regulate nAChRs, contributing to addiction and withdrawal responses [54,55]. Given the specific requirement of ACR-16 for both the behavioral and Ca²⁺ responses we observed, we hypothesized that nicotine might increase the functional expression of ACR-16 in AVA, thereby driving the enhanced coupling.

To test this, we generated a transgenic C. elegans strain expressing functional ACR-16 cDNA tagged with a fluorescent marker (SL2d::GFP) in AVA neuron in wild-type N2 worms. Quantification of GFP fluorescence intensity before and after nicotine treatment revealed a significant increase following nicotine exposure (S5A and 5B Fig). To avoid potential artifacts arising from overexpression or non-specific promoters, we employed CRISPR-Cas9 genome editing to generate an AVA-specific ACR-16::GFP knock-in strain (EN9002) that preserves endogenous expression levels [50,56,57]. In this strain, an spGFP11 tag was inserted into the acr-16 locus, spGFP1–10 was expressed specifically in AVA using a Cre-loxP-based promoter combination (Materials and methods). ACR-16::GFP signal was enriched in the AVA soma, with punctate labeling along the neurite (Fig 4A). Consistently, nicotine exposure significantly enhanced ACR-16::GFP fluorescence in both soma and neurite compartments (Fig 4A—4C), indicating that nicotine upregulates ACR-16 expression in AVA.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Upregulation of ACR-16 is responsible for the nicotine-enhanced DMP-reversal coupling.

(A) Nicotine up-regulates ACR-16 expression in AVA. Using fluorescently-tagged ACR-16 based on split GFP reconstitution, we inserted three copies of the spGFP11 (split GFP11) sequence into the acr-16::aid::Scarlet locus via CRISPR/Cas9 (ACR-16-AID-Scarlet-spGFP11) and expressed the complementary spGFP1-10 (split GFP1-10) moiety in the AVA to generate AVA specific ACR-16::GFP transgenic line. Scale bar, 10 µm. (B, C) After nicotine exposure, the fluorescence intensity in both AVA soma and neurite are significantly increased. (D) Up, A schematic diagram of in situ whole-cell recording of Caenorhabditis elegans AVA under voltage clamping configuration (V _Clamp). Bottom, Representative currents evoked by acetylcholine (1 mM) (I _ACh) in different genotypes without (− nic) or with (+ nic) nicotine exposure. The purple arrows indicate the dosing time. AVA was hold at −60 mV. (E) Quantification of peak currents across different genotypes. Two-way ANOVA was performed (interaction: F_{(2, 42)} = 4.105, P = 0.0235). n ≥ 6 animals. ns, no significance, * p < 0.05, ** p < 0.01, *** p < 0.001 by Two-way ANOVA analysis. Error bars, SEM. The data underlying this figure can be found in S1 Data.

https://doi.org/10.1371/journal.pbio.3003423.g004

The behavioral, Ca²⁺ imaging, and expression results all point to a potential mechanism in which nicotine upregulates ACR-16, thereby enhancing AVA activity and promoting reversals. To directly test this idea, we performed in situ whole-cell patch-clamp recordings from dissected AVA neurons to measure ACh-evoked currents (Fig 4D). In untreated wild-type animals, ACh (1 mM) elicited robust inward currents (Fig 4D), confirming the presence of functional AChRs in AVA. Strikingly, nicotine-pretreated animals displayed significantly larger ACh-evoked currents (Fig 4D and 4E). Application of the nAChR-selective antagonist DHβE strongly attenuated this nicotine-induced potentiation (S5C and S5D Fig). These results indicate that AVA not only expresses functional ionotropic AChRs [49], but that their expression may also be enhanced following nicotine exposure. In light of our previous findings identifying a specific ACR-16 dependence, we next examined whether ACR-16 was required for this potentiation. In acr-16 mutants, nicotine exposure failed to increase ACh-evoked currents (Fig 4D and 4E). Reintroducing ACR-16 in AVA restored the nicotine-induced enhancement (Fig 4D and 4E). These findings suggest that nicotine-enhanced ACh current in AVA is also mediated by ACR-16.

Notably, in nicotine-free conditions, acr-16 mutants exhibited smaller ACh currents than wild-type worms, and this deficit was rescued by AVA-specific ACR-16 expression—consistent with the reduced DMP-reversal coupling observed in these mutants in the absence of nicotine (Fig 4D and 4E). Importantly, neither nicotine exposure, acr-16 mutation, nor DHβE application significantly altered AVA resting membrane potential or voltage-dependent conductance (S6 Fig), suggesting that the enhanced Ca²⁺ activity arises from increased synaptic input rather than changes in intrinsic excitability.

Touch receptor neurons mediate nicotine-induced abnormal motor coupling

What signaling pathway links the DMP-reversal coupling with enhanced AVA Ca²⁺ activity? The C. elegans DMP involves rhythmic release of transmitters (e.g., H⁺) and peptides (e.g., NLP-40) from the intestine, which trigger strong contractions of body wall and enteric muscles [32,58]. One possibility is that these signaling molecules directly stimulate AVA, leading to enhanced activity. However, the receptors for H⁺ and NLP-40 are neither expressed nor functional in AVA [32,59], making direct activation unlikely. An alternative hypothesis is that the forceful muscle contractions during DMP generate mechanical stimuli that activate mechanosensory, proprioceptive, or nociceptive neurons, which in turn influence AVA activity. To test this, we systematically screened a range of sensory-deficient mutants. Worms lacking sensory receptors in nociceptive neurons (ASH, OLQ, ALA, IL1), including deg-1, tmc-1, trpa-1, and osm-9 [60–62], displayed normal nicotine-enhanced DMP-reversal coupling (Fig 5A). Similarly, mutants deficient in major proprioceptive receptors (trp-1, trp-2, trp-4) expressed in DVA [16,63] or SMD neurons [63] also retained robust coupling in response to nicotine (Fig 5A). These results suggest that nociceptive and proprioceptive neurons are not required for the effect.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Exteroceptive TRNs are essential for nicotine enhanced DMP-reversal coupling.

(A) Suppression of nicotine enhanced DMP-reversal coupling in mec-2, mec-4, mec-6 mutation worms. Neuronal specific expression of MEC-6 restored nicotine enhanced DMP-reversal coupling in mec-6 mutants. The coupling proportion was unaffected by mutations in genes essential for harsh touch, proprioception, and nociceptive functions. Two-way ANOVA was performed (interaction: F_{(12, 217)} = 19.72, P < 0.0001). (B) Up, Blue light irradiation of TRNs expressing GtACR2 completely blocked nicotine-enhanced DMP-reversal coupling. The numbers 1 through 3 refer to three consecutive DMP cycles. Gray bar: DMP not couple with reversal; Orange bar: DMP couples with reversal. Bottom, Quantification of the proportion in transgenic TRN::GtACR2 animals without and with blue light after nicotine exposure. (C) Up, A schematic diagram of ablation of TRNs by miniSOG. Bottom, TRNs injury abolished nicotine enhanced DMP-reversal coupling. Two-way ANOVA was performed (interaction: F_{(1, 25)} = 87.52, P < 0.0001). n ≥ 5 animals. ns, no significance, *** p < 0.001 by Two-way ANOVA analysis. Error bars, SEM. The data underlying this figure can be found in S1 Data.

https://doi.org/10.1371/journal.pbio.3003423.g005

Strikingly, mutants lacking key components of the gentle-touch mechanosensory channel complex—mec-2, mec-4, or mec-6—exhibited a profound loss of nicotine sensitivity, such that nicotine exposure no longer induces significant DMP-reversal coupling (Fig 5A). These mec genes encode essential mechanosensory channel complexes in TRNs (ALM, AVM, PLM, and PVM) [64] and did not significantly affect the DMP cycle length (S7A Fig). Furthermore, analysis of mec-4(e1611) mutants showed no significant changes in reversal frequency under either on-food or off-food conditions (S7B Fig) [65], indicating no locomotory defects that would interfere with the analysis of motor coupling. In mec-6 mutants, reintroducing MEC-6 specifically into TRNs fully restored nicotine-enhanced coupling (Fig 5A). By contrast, other mechanosensory neurons, such as PVD and FLP, which mediate harsh-touch sensation and are affected by mec-10 mutations, showed no changes in coupling (Fig 5A). Together, these results indicate that mechanosensory input from TRNs is essential for nicotine’s effect on abnormal motor coupling.

To further confirm this, we performed optogenetic inhibition of TRNs using GtACR2, a light-gated inhibitory chloride channel [66,67]. Light illumination completely suppressed nicotine-enhanced DMP-reversal coupling (Fig 5B). Likewise, targeted ablation of TRNs using neuron-specific expression of miniSOG (mini singlet oxygen generator) [68] (S7C Fig)—which induces acute functional loss and subsequent neuronal death upon photoactivation—abolished the nicotine effect (Fig 5C). In contrast, ablation of the proprioceptive neuron DVA, previously reported to sense the body contraction [69], had no effect on the coupling (S7D Fig). These results demonstrate that TRNs are indispensable for mediating nicotine’s action.

Nicotine enhances the reversal sensitivity to gentle touch

TRNs are well-characterized mechanosensory neurons responsible for detecting exteroceptive gentle touch stimuli [70]. The finding that TRNs mediate nicotine-enhanced interoceptive response raises the question of whether nicotine might also affect exteroceptive sensitivity. To examine this, we applied gentle-touch stimulation to the anterior body using an eyelash probe [71]. Wild-type animals exhibited rapid and sustained reversals to touch, with an average reversal speed of 0.104 ± 0.007 mm/s and a reversal duration of 4.9 ± 0.4 s (Fig 6A—6C). Remarkably, in nicotine-treated worms, the touch elicited a faster reversal speed (+ nic, 0.136 ± 0.006 mm/s) (Fig 6A and 6B), and a prolonged reversal duration (+ nic, 6.4 ± 0.5 s) (Fig 6A and 6C).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. Nicotine facilitates the signal strength from TRNs to AVA.

(A) Upper, a brief diagram shows standard gentle touch evoked reversal for exteroceptive sensitivity examination. Bottom, raster plots showing individual speed responses of different genotypic animals to gentle touch on the anterior body, with and without nicotine exposure (− nic/ + nic) (n = 20 animals per group). Arrows denote gentle touch. (B, C) Quantification of the gentle touch evoked reversal speed (B) and duration (C). Two-way ANOVA was performed (interaction: F_{(2, 114)} = 3.557, P = 0.0317 for (B); F_{(2, 114)} = 2.581, P = 0.0801 for (C)). (D) Left, schematic diagram of in situ whole cell recording of dissected AVA neuron by optogenetically stimulating TRNs under current clamp configuration (I _Clamp). Right, light-evoked excitatory post-synaptic potentials (EPSPs) under different pharmacological conditions in wild-type animals. (E) Quantification of the peak amplitude of the EPSPs. Nicotine significantly increased the signal strength in TRN-AVA circuit. ns, no significance, * p < 0.05, *** p < 0.001 by One-way ANOVA analysis (F_{(3, 20)} = 23.87, P < 0.0001). Error bars, SEM. (F) TRNs evoked EPSPs in acr-16 mutants and rescue lines. (G) Quantification of the peak amplitude of the EPSPs in different genotypes. Nicotine increased EPSPs were reduced in acr-16 mutant animals and could be restored by expression ACR-16 back to AVA. Two-way ANOVA was performed (interaction: F_{(2, 30)} = 3.067, P = 0.0614). n ≥ 6 animals. ns, no significance, * p < 0.05, ** p < 0.01, *** p < 0.001 by Two-way ANOVA analysis. # p < 0.05 by Student t test. Error bars, SEM. The data underlying this figure can be found in S1 Data.

https://doi.org/10.1371/journal.pbio.3003423.g006

Consistently, when we examined the gentle touch-evoked reversal response in acr-16 mutants, nicotine exposure failed to enhance reversal speed (− nic, 0.096 ± 0.006 mm/s; + nic, 0.098 ± 0.005 mm/s) or duration (− nic, 4.3 ± 0.3 s; + nic, 4.6 ± 0.3 s) (Fig 6A—6C). Restoring acr-16 expression in AVA rescued the nicotine-induced enhancement of reversal responses (Fig 6A—6C). These results demonstrate that nicotine not only induces an anxiety/aversion-like motor coupling behavior but also heightens escape responses to mechanical stimulation via ACR-16 in AVA. Both behaviors are regulated by TRNs, highlighting the critical role of TRNs in integrating nicotine’s effects on abnormal sensory-motor coupling.

Nicotine enhances neural activity from TRNs to AVA via ACR-16

We next asked how TRNs regulate nicotine-modulated behaviors. Connectome analyses indicate that TRNs form functional neural connections with command interneurons to mediate escape responses [72]. To assess the impact of nicotine on the neural activity between TRNs and AVA, we combined optogenetics together with in vivo electrophysiological recordings [73–75]. Light activation of TRNs (TRN::ChR2) with expressing the excitatory Channelrhodopsin-2 (ChR2) [76], evoked excitatory postsynaptic potentials (EPSPs) in AVA (Fig 6D). Nicotine exposure significantly increased EPSP amplitude (Fig 6D and 6E). Notably, this increase was blocked by the nicotinic receptor antagonist DHβE (Fig 6D and 6E). These results demonstrate that nicotine enhances the signal strength from TRNs to AVA.

To determine whether ACR-16 is required for nicotine-induced signal potentiation between TRNs and AVA, we analyzed the light-evoked EPSPs in acr-16 mutants. Nicotine failed to further enhance EPSPs in acr-16 mutants (Fig 6F and 6G). Reintroducing acr-16 into AVA restored nicotine-enhanced EPSP potentiation. These results denote that nicotine-potentiated EPSPs depend on ACR-16 (Fig 6F and 6G). Interestingly, in the absence of nicotine exposure, EPSPs were reduced in acr-16 mutants compared to wild-type animals (Fig 6G), suggesting that ACR-16 contributes intrinsically, at least in part, to basic neural signal from TRNs to AVA. This is consistent with our observation that ACh perfusion-evoked currents also require ACR-16.

In summary, our behavioral and in vivo electrophysiological data demonstrate that nicotine strengthens TRNs and AVA synaptic signal via ACR-16. This synaptic enhancement provides a neuronal integration mechanism for nicotine’s modulation of both interoceptive and exteroceptive responses.

Maintenance and generation of C. elegans strains

All strains were maintained on standard NGM plates seeded with E. coli OP50, which were allowed to grow for at least 16 h prior to experimentation, and kept at 22 °C under standard conditions. Unless otherwise stated, “wild-type” refers to the Bristol N2 strain. For optogenetic experiments, strains were maintained in darkness on NGM plates supplemented with 0.5 mM all-trans retinal (ATR), with control groups reared on plates lacking ATR. Genetic mutants used to generate transgenic lines and compound mutants were obtained from the Caenorhabditis Genetics Center (CGC; http://www.cbs.umn.edu/CGC/) or the National Bio-Resources Project (NBRP; http://www.shigen.nig.ac.jp/c.elegans/indices.jsp).

Transgenic animals carrying non-integrated, extra-chromosomal arrays (gaaEx) were produced by co-injecting an injection marker along with one or more DNA constructs at 5–30 ng/µL. Stable integrated lines (gaaIs) were obtained from gaaEx strains via UV irradiation, followed by at least four rounds of outcrossing with N2. A complete list of strains used in this study is provided in S1 Table.

Molecular biology

All expression plasmids were generated using the Multisite Gateway system (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA). Three entry clones—containing the promoter, target gene, and fluorescent marker (designated as slot1, slot2, and slot3, respectively)—were recombined into the pDEST R4-R3 Vector II via an LR reaction [100]. Genomic DNA from wild-type C. elegans was used to amplify acr-16, twk-40s, nmr-1, mec-4, myo-3, sra-11, and unc-4. The promoter lengths of Pacr-16, Ptwk-40s, Pnmr-1, Pmec-4, Pmyo-3, Psra-11, and Punc-4 were 3.5, 1.5, 5.0, 1.5, 2.3, 2.0, and 3.0 kb, respectively.

Genes of interest—including endogenous proteins (e.g., ACR-16) and exogenous functional proteins (e.g., GCaMP6s, GtACR2, ChR2, and miniSOG)—were inserted into pDONR221 to generate slot2 constructs. Fluorescent proteins or the unc-54-3′UTR sequences were cloned into the pDONR-P2R-P3 vector to generate slot3 constructs. The Pacr-16::GFP reporter was generated using the Gateway recombination system (Invitrogen) in combination with In-Fusion cloning. To make the entry vectors, a 3.5 kb of sequence upstream of the acr-16 gene was PCR amplified from C. elegans N2 genomic DNA using gene-specific promoter primers containing attB1 and attB2 sequences. The gene-specific sequences for Pacr-16 are as follows:

1. Forward: 5′-AGTTGCTCCCTCGAGACATTCACAACATATACGGTG-3′
2. Reverse: 5′-CTTGCACGACTAGTGGAGTGAATGTACATCCG-3′

The amplified promoter fragment was inserted into a linearized Gateway entry vector (Entry Clone A) using In-Fusion cloning, generating an entry clone containing the acr-16 promoter flanked by attL1 and attL2 recombination sites. An LR recombination reaction was subsequently performed between the entry clone and a Gateway destination vector harboring GFP and the unc-54–3′UTR, resulting in the final expression construct, Pacr-16::GFP::unc-54-3′UTR.

To visualize endogenous ACR-16 expression, three tandem copies of the spGFP11 sequence were inserted into the acr-16::aid::Scarlet locus using CRISPR/Cas9, generating acr-16::aid::Scarlet::spGFP11. Complementary spGFP1-10 was expressed specifically in AVA neurons. Scarlet fluorescence, visible with NeuroSIL, labeled all ACR-16 clusters, while GFP fluorescence was reconstituted exclusively in AVA neurons expressing spGFP1-10. AVA-specific expression was achieved using a Cre-loxP recombination strategy with Pflp-18 and Pgpa-14 promoters.

The strain driven by the Ptwk-40s promoter expressed ACR-16 cDNA tagged with GFP, resulting in predominant expression of functional ACR-16 in AVA neurons, as confirmed by GFP fluorescence. The cDNA sequence was amplified via 2× Phanta Max Master Mix (Dye Plus) Taq DNA Polymerase (Vazyme Biotech Co.). Detailed plasmid and primers information are provided in S2 and S3 Tables.

Neuronal manipulation

To investigate the roles of TRNs and the proprioceptive neuron DVA in regulating DMP-reversal coupling, we expressed the genetically encoded photosensitizer miniSOG in these neurons driven by Pmec-4 and Pnlp-12 promoters, respectively. Neuronal ablation was achieved by exposing animals to 470 nm blue LED light (8.3 mW/cm²) for 45 min. To verify the specificity and efficiency of ablation, cytoplasmic wCherry was co-expressed with miniSOG under the same promoter, enabling visualization of target neurons. Ablations were performed at the L2–L3 larval stage, and wCherry fluorescence was examined 24 h later to confirm loss of target neurons.

For optogenetic manipulation of TRN neurons, we used the neuron-specific promoter Pmec-4 to drive the expression of the GtACR2 or ChR2 in TRN neurons. Neuronal activity was inhibited or activated using blue light during behavioral assays. In GtACR2 experiments, TRN inhibition was used to evaluate the contribution of these neurons to nicotine-induced modulation of DMP behavior. Worms were subjected to three consecutive DMP trials: in the first and third trials (no blue light), TRNs were active, and nicotine-treated worms exhibited reversals following DMP. In the second trial, blue light was applied to inhibit TRNs, which abolished reversals after the DMP, indicating that TRN activity is required for nicotine-enhanced DMP-reversal coupling.

Behavioral analysis

Behavioral assays were conducted at 22 °C. Late L4 stage or early young adults (12 h post L4 stage) hermaphrodites were used. 4.86 μL of nicotine (Shanghai PureOne Biotechnology, l-Nicotine, HPLC purity ≥ 98%) was added into 1 mL of an E. coli OP50 suspension to create a 30 mM nicotine-OP50 mixture after vortexing for 1 min. 100 μL of this mixture was then dropped and spread onto the surface of a standard 3 mL NGM agar plate, and the NGM was shaken to ensure that the mixed solution evenly covered the entire surface. These nicotine plates were left at 22 °C for 16 h to allow diffusion of nicotine into the agar and OP50 bacterial lawn formation [101]. The final nicotine concentration in these plates is approximately 1 mM. NGM plates were prepared both with and without nicotine, with DHβE added to a final concentration of 20 µM. DNQX (500 µM) and MK-801 (50 µM) were simultaneously included in both nicotine-present and nicotine-absent NGM plates to serve as the treatment conditions for C. elegans. Worms were transferred to NGM plates with OP50 containing nicotine and cultured for 3 h prior to the experiment.

Defecation behavior was measured on NGM plates under standard conditions, as previously described [31]. The nematodes’ defecation behavior was monitored for 10 consecutive cycles. The backward coupling rate was then calculated by dividing the number of backward events by the total number of defecations. The time window for assessing motor coupling was defined as the 5-s period following DMP initiation, during which reversals and calcium transients were recorded. Defecation phenotypes were scored under a modified stereomicroscope (Axio Zoom V16, Zeiss) equipped with a digital camera (acA2500-60 um, Basler).

For locomotion analysis, hermaphroditic worms, cultured under standard conditions, were transferred to imaging plates seeded with a thin layer of OP50 bacteria. One minute after the transfer, a 3-min video of the nematodes’ locomotion was recorded. The eyelash touch assay was performed on food. Post-imaging analysis was conducted using an in-house developed MATLAB script, with the centerline used for tracking. The speed analysis of each animal was performed by dividing the recorded images into 33 segments, and the midpoint of each segment was used to calculate the velocity [100]. All images were captured using a 12× objective at a frame rate of 10 Hz.

Confocal fluorescence microscopy

L4 stage transgenic animals expressing fluorescence markers were picked to imaging. Live worms were immobilized by 2.5 mM levamisole (Sigma-Aldrich) in M9 buffer. SGA741 gaaIs55 N2; [Ptwk-40s::ACR-16::GFP] fluorescence signals were captured using a Plan-Apochromatic 60× objective on a confocal microscope (FV3000, Olympus). EN9002 krSi81 [Pmyo-3::TIR1::bfp] I; bab535 [acr-16::aid::scarlet::spgfp11x3] V; krEx1404 [Pflp-18::lox::STOP::lox::spgfp1-10; Pgap-14::cre; Pmyo-2::mCherry] fluorescence signals were performed using an Andor Revolution XD laser confocal microscope system (Andor Technology plc., Springvale Business Park, Belfast, UK) based on a spinning-disk confocal scanning head CSU-X1 (Yokogawa Electric Corporation, Musashino-shi, Tokyo, Japan). The confocal system was mounted on an Olympus IX-71 inverted microscope (Olympus, Tokyo, Japan) and controlled by Andor IQ 1.91 software. The neurons and muscles labeled with GFP or RFP have been imaged with an excitation wavelength laser at 488 or 561 nm, respectively. The images were displayed using Image J 1.52o software (Wayne Rasband, National Institutes of Health, USA).

Calcium imaging

The free-tracking Ca²⁺ imaging setup consists of two integrated modules: a behavior tracking system and a fluorescence recording system. Worm behavior was imaged under dark-field illumination in the near-infrared range. Individual worms were picked from 6 cm NGM plates and placed on a THOPLABS SPW602 XY motorized stage for automated control. Imaging plates were prepared using the same agar composition as behavioral assay plates.

Behavioral tracking was performed using a 10× inverted objective (Olympus, Japan) and captured with a CCD camera (Point Gray Research, CM3-U3-13S2M, Canada). Custom real-time computer vision software kept the worm centered in the field of view by tracing the centerline of the worm. The fluorescence recording system records neuronal Ca²⁺ activity using an sCMOS digital camera (Hamamatsu, Japan) equipped with a 10× objective lens (Olympus, Japan). To simultaneously capture images of wCherry and GCaMP6 side by side, we utilized a two-channel imager (W-VIEW GEMINI, Japan). Red and green channel images were simultaneously recorded at 10 frames per second (fps) with exposure times of 10 and 60 ms, respectively, using HCImage software (Hamamatsu).

Neural activity was quantified as the normalized change in the fluorescence ratio between GCaMP6s and wCherry, calculated as: ΔR/R₀ = (R − R₀)/R₀, where R = I_GCaMP6s/I_wCherry, and R₀ was defined as the minimum R value within the recording period. Intensities I_GCaMP6s and I_wCherry were extracted as the pixel intensity in the green and red channels, respectively, using custom MATLAB scripts. Analysis focused on AVA neuronal activity during freely moving worms performing the DMP.

A calcium response was defined as an increase in fluorescence intensity of ≥10% that occurred with a reversal behavior immediately following the pBoc phase of the DMP. Each such event was counted as one calcium response. We then calculated the calcium response percentage by dividing the number of these response events by five—the total number of DMP cycles analyzed per animal. This quantification yields discrete values (e.g., 0, 20%, 40%, …, 100%) corresponding to the number of DMP cycles that showed a coupled calcium response. To minimize signal drift and motion artifacts, we limited our analysis to five consecutive DMP cycles per worm.

Electrophysiology

Whole-cell patch-clamp recordings were performed following established protocols with minor modifications [102,103]. Briefly, 1–2-day-old hermaphrodite adults were immobilized on a Sylgard-coated (Sylgard 184, Dow Corning) coverslip using Histoacryl Blue adhesive (Braun) under a stereomicroscope (M50, Leica) and submerged in bath solution. Viscera were removed by gentle suction through a glass pipette, and the resulting cuticle flap was folded back and secured with WORMGLU adhesive (GluStitch) to expose the neuromuscular system.

Body wall muscle cells and neurons were patched using borosilicate glass pipettes (1B100F-4, World Precision Instruments) with tip resistances of 4–6 MΩ and 15–20 MΩ, respectively. Pipettes were pulled by micropipette puller P-1000 (Sutter) and fire-polished using an MF-830 microforge (Narishige). Recordings were obtained in the whole-cell configuration using an EPC9 amplifier (HEKA) controlled by Pulse software, with data processed using Igor Pro (WaveMetrics) and Clampfit 10 (Molecular Devices). Membrane currents were recorded at a holding potential of −60 mV, while membrane potentials were recorded at 0 pA. Signals were digitized at 10 kHz and low-pass filtered at 2.6 kHz.

The pipette solution contains (in mM): K-gluconate 115; KCl 25; CaCl₂ 0.1; MgCl₂ 5; BAPTA 1; HEPES 10; Na₂ATP 5; Na₂GTP 0.5; cAMP 0.5; cGMP 0.5, pH 7.2 with KOH, ∼320 mOsm. cAMP and cGMP were included to maintain the activity and longevity of the preparation. The bath solution consists of (in mM): NaCl 150; KCl 5; CaCl₂ 5; MgCl₂ 1; glucose 10; sucrose 5; HEPES 15, pH 7.3 with NaOH, ∼330 mOsm. All chemicals were purchased from Sigma unless otherwise specified. Experiments were conducted at 20–22 °C. To test AVA receptor function, ACh (1 mM in bath solution) was directly perfused onto patched neurons. For pharmacological blockade, recordings were performed in bath solution containing 20 μM DHβE. Optogenetic activation of TRNs during recordings was achieved by blue light illumination (470 nm, 13.75 mW/cm²).

Statistical analysis and display

Statistical analyses were conducted via Student t test, One-way ANOVA, or Two-way ANOVA analysis, two-sample Z test, as appropriate. Data are presented as mean ± SEM, and significance thresholds were set at p < 0.05 (* p < 0.05, ** p < 0.01, *** p < 0.001). Calcium imaging data were visualized as heat maps (MATLAB, MathWorks), time-course curves (Igor Pro, WaveMetrics), and scatter plots (GraphPad Prism 9, GraphPad Software), with each point representing a single worm. Additional graphing software included Clampfit (Molecular Devices), ImageJ (NIH), and MEGA 6.60 (for phylogenetic analyses). For behavioral and fluorescence imaging experiments, unless otherwise noted, each recording trace represented an independent animal.