FLP-18 Functions through the G-Protein-Coupled Receptors NPR-1 and NPR-4 to Modulate Reversal Length in Caenorhabditis elegans

Animal behavior is critically dependent on the activity of neuropeptides. Reversals, one of the most conspicuous behaviors in Caenorhabditis elegans, plays an important role in determining the navigation strategy of the animal. Our experiments on hermaphrodite C. elegans show the involvement of a neuropeptide FLP-18 in modulating reversal length in these hermaphrodites. We show that FLP-18 controls the reversal length by regulating the activity of AVA interneurons through the G-protein-coupled neuropeptide receptors, NPR-4 and NPR-1. We go on to show that the site of action of these receptors is the AVA interneuron for NPR-4 and the ASE sensory neurons for NPR-1. We further show that mutants in the neuropeptide, flp-18, and its receptors show increased reversal lengths. Consistent with the behavioral data, calcium levels in the AVA neuron of freely reversing C. elegans were significantly higher and persisted for longer durations in flp-18, npr-1, npr-4, and npr-1 npr-4 genetic backgrounds compared with wild-type control animals. Finally, we show that increasing FLP-18 levels through genetic and physiological manipulations causes shorter reversal lengths. Together, our analysis suggests that the FLP-18/NPR-1/NPR-4 signaling is a pivotal point in the regulation of reversal length under varied genetic and environmental conditions. SIGNIFICANCE STATEMENT In this study, we elucidate the circuit and molecular machinery required for normal reversal behavior in hermaphrodite Caenorhabditis elegans. We delineate the circuit and the neuropeptide receptors required for maintaining reversal length in C. elegans. Our work sheds light on the importance of a single neuropeptide, FLP-18, and how change in levels in this one peptide could allow the animal to change the length of its reversal, thereby modulating how the C. elegans explores its environment. We also go on to show that FLP-18 functions to maintain reversal length through the neuropeptide receptors NPR-4 and NPR-1. Our study will allow for a better understanding of the complete repertoire of behaviors shown by freely moving animals as they explore their environment.


Introduction
Neuropeptides are used by neurons to communicate with each other and to modulate behavior. They modulate synaptic activity through synaptic and extrasynaptic neuropeptide receptors. Neurons generally signal through only one neurotransmitter, but they can secrete multiple neuropeptides. One difficulty in studying neuropeptide function is that a single neuropeptide can signal through multiple receptors (Li and Kim, 2008).
Locomotion is the basis for most behaviors in Caenorhabditis elegans, including feeding, mating, sleeping, avoidance, and escape behaviors. The animal's movement is mediated by multiple sets of motor neurons along the ventral nerve cord innervating the dorsal and ventral body-wall muscles (White et al., 1976). Muscle contractions are mediated by cholinergic A-and B-type motor neurons that innervate these muscles (White et al., 1986;Alfonso et al., 1993), whereas muscle relaxations are mediated by the GABAergic D-type motor neurons (White et al., 1986;McIntire et al., 1993a, b). The multiple sets of motor neurons that innervate the body-wall muscles receive synaptic inputs from interneurons, which are in turn stimulated by sensory neurons (Goodman, 2006). Together, these neurons coordinate the activity of the ventral and dorsal muscles to generate the sinusoidal waves for forward locomotion, interspersed by short bouts of backward movement or reversals.
Reversals are critical determinants of the C. elegans navigation strategy as they increase the probability of change in direction. Reversal behavior can be parsed into two components: reversal initiation, which determines the frequency of reversals; and reversal termination, which determines the reversal length. Shorter reversals allow for change in directions at angles Ͻ90 degrees, whereas longer reversals are almost always followed by turns that lead to an ϳ180 degree change in direction of locomotion. Laser ablation of the AVA command interneuron completely eliminated longer reversals, but the animal could still initiate reversals, suggesting that reversal initiation (frequency) and regulation of reversal length might have a different cellular basis .
Although a number of studies have highlighted the neurons, molecules, and circuits that are required for the reversal behavior in C. elegans (Chalfie et al., 1985;Kaplan and Horvitz, 1993;Hart et al., 1995;Maricq et al., 1995;Y. Zheng et al., 1999;Brockie et al., 2001;Zhao et al., 2003;Alkema et al., 2005;Cohen et al., 2009;Piggott et al., 2011;Campbell et al., 2016), most of these studies were focused on the reversal frequency. The role of neurons in regulating reversal length has to date been elaborated in a single previous study involving laser ablation of neurons . A very recent study has shown that the AVA interneurons control reversal state by affecting the motor neurons and that the P-/Q-type calcium channel UNC-2 is required for maintaining the duration of reversals through its effect on motor neuron oscillations. The authors further showed that enhancing UNC-2 activity shows an increase in reversal duration and velocity (Gao et al., 2018). However, although the neurons required for maintaining reversal length and the role of UNC-2 in this process have been studied, the molecular mechanisms that regulate the length of reversals remain largely unknown.
Studies have shown that the neuropeptide FLP-18 is required for modulating spontaneous reversal frequency and swimming rates in C. elegans (Kim and Li, 2004;Cohen et al., 2009;Y.J. Chang et al., 2015). However, the effects of FLP-18 on reversal length have not been studied. We show that FLP-18 controls reversal length by regulating the function of AVA interneurons and ASE sensory neurons through the G-proteincoupled neuropeptide receptors NPR-4 and NPR-1, respectively. Our experiments suggest a possible correlation between increased reversal length and increased activity of the AVA neuron in flp-18, npr-1, and npr-4 genetic backgrounds in freely moving animals. We further show that FLP-18 expression is regulated by the transcription factor CREB1/CRH-1 as well as by starvation and that both loss of creb1/crh-1 and starvation cause decrease in reversal length through the FLP-18 pathway. This indicates that FLP-18 could act as a behavioral switch from local search to global search behavior under different genetic conditions as well as during starvation. Our work identifies the FLP-18/NPR-1/4 signaling path-way as a pivotal point in defining an aspect of C. elegans locomotion under varied environmental and genetic conditions.

Materials and Methods
Strains. All the strains used in this work were grown at 20°C under standard conditions using nematode growth medium (NGM) plates. The N2 strain was used as the reference wild-type (WT) strain. The list of strains used in this study is given in Table 1. All experiments were performed with hermaphrodite C. elegans.
A list of all the primers used to make constructs and genotype the C. elegans in this work is given in Table 2, and the plasmids used in this work are indicated in Table 3.
Behavioral assays. Well-fed young adult animals were used to conduct all the behavioral studies, except for the starvation experiments. The reversal assays were performed after transferring the C. elegans to food-free NGM plates. Observation time started 1 min after transfer. The C. elegans were transferred using halocarbon oil or eyelash picks to avoid any injury to the animal. Spontaneous reversals were scored for 5 min as previously described (Zhao et al., 2003). The reversal length is reported throughout this work as body bends per reversal. We defined one body bend as equal to one-third the length of the C. elegans. For all reversal assays, a 5 min video was made and the number of body bends per reversal calculated for every reversal made during the 5 min that was recorded, the number of body bends was then averaged over every reversal made during the recording and was plotted as a single dot in the scatter plot. More than 20 such recordings were analyzed for each genotype. The results were plotted as mean Ϯ SEM and evaluated using the standard Student's t test.
For the starvation assays, animals were starved for 24 h at the L4 stage (54 h including hatching time) on peptone free plates (to avoid contamination). After 24 h, C. elegans were assayed for reversal behavior. Again, reversal length was determined for each spontaneous reversal.
Optogenetic assays. To stimulate a specific neuron, channelrhodopsin-2 (ChR2) was used as previously described (Husson et al., 2013). The sra-6 promoter was used to drive ASH neuron-specific expression of ChR2 as previously reported in the SRS85 strain (sraIs49 V; lite-1 (ce314 ) X; sraEx80) (Guo et al., 2009). Animals expressing ChR2 were grown on NGM agar plates seeded with OP50-containing 400 M All-trans retinal. The C. elegans were grown in the dark until late L4/early adult stages. The assay was performed on freshly seeded NGM plates. During the assay, ChR2 was excited by blue light (460 -490 nm) sourced from an epifluorescence unit (U-HGLGPS, Olympus) attached to the Nikon SMZ2000 microscope. This experiment was performed under low illumination to avoid the preactivation of ChR2. Blue light was illuminated for 3 s, and reversal length was quantified as number of body bends in response to illumination. This assay was done with Ͼ20 animals per genotype, and each dot in the scatter plot indicates the number of body bends in a single reversal brought about by stimulation of ASH. The results were plotted as mean Ϯ SEM and evaluated using the standard Student's t test. The experimenter was blind to the genotypes of the strains while performing these experiments.
Calcium imaging. The genetically encoded Ca 2ϩ indicator GCaMPexpressing strain Prig-3::GCaMP5 was used to visualize Ca 2ϩ transients in the AVA command interneuron (Larsch et al., 2013). Calcium transients were recorded in freely navigating C. elegans as previously described (Faumont et al., 2011) using an Olympus IX73 inverted microscope fitted with the ASI-based worm-tracker. Imaging was done at 40ϫ objective with 0.6 NA. Images were acquired through QImaging camera using the ImageJ software. Videos were recorded at 10 frames per second with 100 ms exposure.
The AVA neuron activity was analyzed according to the protocol described previously (Kerr, 2006). The analysis was done using the FIJI ImageJ software. The ROI was drawn as a 25 ϫ 25 pixel circle over the AVA cell body-expressing GCaMP. The measured average pixel value from the ROI, F meas , includes fluorescence from sample and background fluorescence (F bkg ). Then the fluorescence (F) from the given ROI was estimated by subtracting background fluorescence from the measured fluorescence value (i.e., F ϭ F meas Ϫ F bkg ). The fluorescence value was estimated for each frame after 100 ms by manual repositioning of the ROI. Calcium transients were plotted as ⌬F/F o , where ⌬F is the change in the fluorescence value (F) from its baseline fluorescence (F o ). Calcium levels were estimated as ⌬F/F o max, which indicated the maximum ⌬F/F o value for each animal. Calcium-raising duration in AVA was calculated from the frame when the animal initiated the reversal to the end of the reversal. The standard Student's t test was used to compare the ⌬F/F o, duration of AVA and the calcium-raising duration (⌬t) values between the WT and mutant strains.
Microscopy. For all fluorescence confocal microscopy-based experiments, except the experiments involving starvation assays, late L4/early adult C. elegans were imaged. The animals were paralyzed using 2,3butanedione monoxime (30 mg/ml) for imaging pflp-18::FLP-18::sl2:: GFP in the neurons of the head region (Cohen et al., 2009). Images were acquired using a Leica TCS SP8 confocal microscope. Imaging of starvation experiments was done 24 h following the L4 stage. The fluorescence intensity from the ROI was quantified as the integrated intensity using ImageJ. The data are expressed as mean Ϯ SEM. To perform neuronspecific imaging, the neuron of interest was identified by expressing the mCherry marker under a neuron-specific promoter. The Pnmr-1:: mCherry and Pttx-3::mcherry promoters were used as markers to identify AVA, RIM, and AIY. GFP fluorescence quantification of Pflp-18:: FLP-18::sl2::GFP was performed from a single neuron for each animal using ImageJ. Imaging results were evaluated using standard Student's t test.
Statistical analysis. The Student's t test (unpaired t test with Welch's correction) was performed using Prism 6 (GraphPad). All the statistical values for unpaired t test were provided as ( p, t, df ). The level of significance was set as p Ͻ 0.05  npr-1 npr-4; Pflp-18::NPR-4a (indEx501) 1 BAB1505
To get a cleaner readout of the reverse body bends in the triple mutants, we decided to combine optogenetics with behavioral studies. We used a transgenic line expressing ChR2 specifically in the ASH sensory neurons using the sra-6 promoter. We induced reversals optogenetically by stimulating the ASH neuron, which forms functional connections with the AVA interneuron (Guo et al., 2009;Lindsay et al., 2011). We noted that ASH stimulation caused WT animals to undergo reversals (illustrated in Fig. 1C), even though the optogenetically activated neural circuit is different from the neural circuitry that activates reversals during exploratory behavior in C. elegans Pirri and Alkema, 2012). ChR2-mediated stimulation of the ASH neurons in flp-18 mutants resulted in a significant increase in the number of body bends per reversal compared with WT control animals ( p Ͻ 0.0001, t ϭ 4.739, df ϭ 32.95; Fig. 1D). A similar number of increased reversal body bends upon ChR2 activation were seen in npr-1 npr-4 double mutants ( p ϭ 0.0003, t ϭ 4.222, df ϭ 25.48), which were significantly rescued by expressing NPR-4 in the AVA command interneuron ( p ϭ 0.0003, t ϭ 4.132, df ϭ 27.82). Interestingly, expressing NPR-4 in just the AVA command interneurons could completely rescue the double mutant phenotype, indicating that overexpressing NPR-4 in the double mutant was enough to reduce reversal length to WT levels ( p ϭ 0.629, t ϭ 0.49, df ϭ 25.21) in optogenetically induced reversals. Further, we found that the flp-18 npr-1 npr-4 triple mutants showed a number of body bends per reversal that was similar to that seen in flp-18 mutants ( p ϭ 0.078, t ϭ 1.824, df ϭ 30.31) or in npr-1 npr-4 double mutants ( p ϭ 0.166, t ϭ 1.432, df ϭ 23.26; Fig. 1D). These data together indicated that FLP-18 functions through the G-protein coupled receptors NPR-1 and NPR-4 to control reversal length.
NPR-1 and the Brugia malayi homolog of NPR-4 have been shown to have inhibitory functions through FLP-18 (Frooninckx et al., 2012;Anderson et al., 2014). Hence, it is likely that FLP-18 functions through the NPR-4 and NPR-1 receptors and allows for modulation of the AVA and ASE neurons. Together, these results indicate that FLP-18 could modulate the activity of AVA and ASE sensory neuron through the NPR-4 and NPR-1 receptors and this modulation allows the animal to control reversal lengths. Although our experiments point to the fact that NPR-4 is functioning in the AVA neuron and NPR-1 functions in the ASE neurons, the fact that expressing NPR-1 in other sensory neurons and/or RMG neurons (with gpa-3 and flp-5 promoters) also allows for rescue of the increased reversal phenotype seen in the npr-1 mutants indicates the possibility of NPR-1 functioning in other neurons to maintain reversal length. Further analysis at the level of single neurons could allow one to understand whether NPR-1 has a much more complex role in the process of regulating reversal length. Further,  have shown that C. elegans with laserablated AVA neurons do not show longer reversals during local search behavior. This information, along with our previously described results where NPR-4 expression in AVA rescues the npr-4 mutant phenotype, indicates a possible role of AVA in controlling the length of backward movement. How activity of AVA is related to length of reversals is not clear. We reasoned that calciumraising duration and/or calcium levels of AVA during reversals might have some relation with the length of reversals. To delineate the relation of AVA activity with length of reversals, we used C. elegans with GCaMP5.0 expressed in the AVA interneuron using the rig-3 promoter (Larsch et al., 2013) and measured calcium levels in freely navigating animals during spontaneous reversals. Consistent with our previous results, flp-18 mutants showed increased reversal length compared with WT animals (Fig. 2). Furthermore, very strikingly, flp-18 mutants showed an increase in calcium levels ( p Ͻ 0.0001, t ϭ 8.336, df ϭ 16.13) and in the calcium-raising duration for AVA during reversals ( p Ͻ 0.0001, t ϭ 5.464, df ϭ 17.22; Fig. 2A-C; https://www.amazon.com/ clouddrive/share/oEQEG4UJTiZxidtYTldE2AKcgppXtZgp5luQ d7RJYWD and https://www.amazon.com/clouddrive/share/2xZ pGLfjhSBcfbe4GAFZVDdQM6TRdo1CrZ4NvbUeoMp).

Mutants in
Our results show that FLP-18 functions through NPR-4 in AVA and NPR-1 in ASE to control reversal length ( Fig. 1 A, B). Both npr-4 and npr-1 mutants showed an increase in calcium levels: WT and npr-1 ( p Ͻ 0.0001, t ϭ 4.656, df ϭ 20.78), WT and npr-4 ( p ϭ 0.0015, t ϭ 3.712, df ϭ 19.11), and calcium-raising duration of AVA during reversal: WT and npr-1 ( p ϭ 0.0054, t ϭ 3.085, df ϭ 22.06), WT and npr-4 ( p ϭ 0.0036, t ϭ 3.198, df ϭ 25.8), compared with WT control animals (Fig. 2). Increased AVA activity in npr-4 mutants suggests that FLP-18 released from AVA and other neurons controls the activity of AVA through the NPR-4 receptor. The function of FLP-18 through NPR-4 in AVA provides a novel example of self-regulation of neuronal activity by neuropeptide signaling. NPR-4 has been reported to be involved in various functions in C. elegans, but the signaling mechanism underlying NPR-4 function is not completely clear. Increased activity of AVA in npr-4 mutants during reversals suggests an inhibitory function of the NPR-4 receptor. Our previous results show that NPR-1 functions in ASE to control reversal length ( Fig. 1 A, C). NPR-1 has been shown to function through inhibitory signaling (Cheung et al., 2005;A. J. Chang et al., 2006;Rogers et al., 2006). Hence, npr-1 mutants might have increased activity in ASE sensory neurons. Increased activity of AVA during reversals in npr-1 mutants could be due to ASE-mediated change in the activity of AIB and AWC neurons, which are known to play important roles in modulating reversal behavior Chalasani et al., 2007;Piggott et al., 2011). Changes in the activity of AIB and AWC could in turn affect the AVA interneuron behavior.
Together, these results show that CREB1/CRH-1 inhibits FLP-18 expression in the C. elegans nervous system.

Starvation affects reverse body bends through increase in FLP-18 levels
Change in locomotion is required for effective exploration of the environment under different physiological conditions. Modulating reversal length is important to allow C. elegans to switch their navigational strategy during prolonged starvation. Our results thus far indicate that reversal length is affected by levels of FLP-18. This prompted us to investigate the expression of FLP-18 during starvation. To study the effect of starvation on FLP-18 levels, we starved the C. elegans for 24 h and extracted RNA from fed and starved animals. We estimated the levels of flp-18 transcripts through qPCR under both conditions and found that there was a 20-fold increase in FLP-18 levels upon starvation (data not shown). We also validated these data by quantitating the levels of Pflp-18::FLP-18::sl2::GFP in well-fed or 24 h starved C. elegans. Again, we found an increase in Pflp-18::FLP-18::sl2::GFP levels upon starvation ( p ϭ 0.0006, t ϭ 3.691, df ϭ 44.91; Fig. 4A). These two pieces of data indicate that starvation causes an increase in FLP-18 levels.
We next went on to ask whether the increase in FLP-18 levels could cause changes in reversal body bends. Upon analyzing the reversal lengths in fed C. elegans and animals starved for 24 h, we saw a significant decrease in the number of body bends in starved animals during each reversal ( p Ͻ 0.0001, t ϭ 4.619, df ϭ 30.71; Fig. 4B; https://www.amazon.com/clouddrive/share/jErueH1 BjmPVhpLuixNKGpG0XTYMQifULeVn0x6hY0r and https:// www.amazon.com/clouddrive/share/4nKmmo0ps2EK7JIMYCa H63unAACD5tmNlasb9rJDf3t). If the decreased reversal body bends is indeed due to increased FLP-18 levels, then loss of FLP-18 should abolish the decrease in body bends that we see in starved WT animals. We went on to analyze the reverse body bends in flp-18 mutants that were starved for 24 h along with control animals that were not starved. We found that the decrease in body bends per reversal that is seen in WT animals upon starvation is completely abolished in flp-18 mutant animals ( p ϭ 0.888, t ϭ 0.142, df ϭ 30.22; Fig. 4B; https://www.amazon.com/ clouddrive/share/RKqSKIVBWrVFLsqiAmL2cgwxcI92mV2Jfr GRsQfp6Sd and https://www.amazon.com/clouddrive/share/ry 5teKQOs5vNsYdp6WTnATS5S9c7kqhzg4Wkdl5kB7X). These data indicate that FLP-18 levels are regulated in C. elegans during starvation. Starved animals show increased FLP-18 expression; and consistent with the fact that increased FLP-18 levels could cause shorter reversals, analysis of these C. elegans showed that they indeed made shorter reversals. We next asked whether FLP-18 is functioning through NPR-1 and NPR-4 during starvation. To do this, we went on to look at the reversal length in the npr-1 npr-4 double mutants in animals that were either fed or starved for 24 h. Again, we found that the npr-1 npr-4 mutants behaved like the flp-18 mutants and had lost the ability to reduce reversal length as was seen in WT animals ( p ϭ 0.4001, t ϭ 0.856, df ϭ 24.13; Fig. 4B). Together, these experiments show that the FLP-18 pathway functioning through NPR-1 and NPR-4 is required for reducing the reversal length during starvation in C. elegans.
One discrepancy that we found in these experiments was that there was a significant reduction in reversal length in flp-18 mutants compared with our previous reversal data. Further, the npr-1 npr-4 double mutants appear to have a lower reversal length than flp-18 mutants at 24 h fed and starved conditions (Fig. 1A vs Fig. 4B for flp-18 and npr-1 npr-4 ). One possible reason could be the age of the animals that differs in both sets of experiments. At 78 h of development, which is when the later experiments were performed, all the animals have multiple eggs. This is not the case for the initial data, which were done at 54 h after egg-laying (including hatching) where the animals had no/few eggs (Chiba and Rankin, 1990). Although the starvation data implicate flp-18 and npr-1 npr-4 to be involved in modulating reversal length during starvation, more experiments would be required to pinpoint the exact mechanism of how these molecules function during satiety and starvation.

Discussion
Neuromodulators modify behaviors by shaping the properties of neural circuits. Here, we reveal the role of the neuropeptide FLP-18 in regulating reversal length in C. elegans. Our initial observations showed that flp-18-null mutant animals make longer reversals compared with WT control animals. Reversal length has been shown to be positively correlated with the probability of change in direction of C. elegans movement after reversal . Hence, reversal length is especially important for behaviors, such as chemotaxis and pathogen avoidance, which require the animals to change locomotory strategies to move toward the source of food or away from a pathogen (H. Luo et al., 2013). Similarly, the switch that C. elegans make from local search to global search after prolonged starvation also depends upon the regulation of reversal length. Our study identifies FLP-18 as a key molecule in the regulation of reversal length under different genetic and environmental conditions. Null mutants of creb1/crh-1 show shorter reversal length (Fig. 3A) and low chemotaxis indices (Y.D., unpublished data) consistent with defects in the ability of the mutants to change direction in response to chemical gradients. We show that FLP-18 expression is negatively regulated by the transcription factor CREB1/CRH-1. Further, flp-18; crh-1 double mutants show behaviors similar to flp-18-null mutants, indicating that FLP-18 is likely to be functioning downstream of CREB1/CRH-1. We further showed that, during starvation, WT C. elegans tend to make shorter reversals, which in turn may result in a change in the animal's exploratory strategy from local search to global search behaviors ; current study). Our experiments suggest that this decrease in the reversal length could be due to increased levels of FLP-18. This is further corroborated by the fact that flp-18-null mutants are unable to regulate reversal length under our assay conditions (Fig. 4).
Reversal circuit analysis revealed that NPR-1, NPR-4, and NPR-5 could be the potential candidates affecting reversal behavior through FLP-18 (Li and Kim, 2014). Mutants in npr-4 show significant increase in reversal length compared with WT animals but significantly less than that seen in flp-18 mutants, which suggests that NPR-4 is not the only receptor through which FLP-18 is modulating the reversal circuit. Reversal behavior analysis of npr-1 mutants also showed a similar behavior to npr-4 mutants, whereas npr-1 npr-4 double mutants showed a significant increase in reversal length comparable with that seen in flp-18 mutants. Behavioral analysis showed that NPR-5, which is expressed in sensory neurons and the first layer of interneurons, is not involved in this process. To further elaborate the site of action of FLP-18, we performed neuron-specific rescue experiments. Our rescue experiments suggest that NPR-4 functions in AVA whereas NPR-1 could function in the ASE sensory neurons. A possible circuit could involve NPR-1 functioning in the ASE sensory neuron to mediate changes in the AVA command interneuron through the AIB interneurons and/or the AWC sensory neurons, at the same time NPR-4 could affect the AVA neuron directly, which is also one of the neurons that expresses FLP-18, thereby allowing for a possible mechanism for the AVA neuron to modulate its own activity during reversals (illustrated in Fig. 5). However, it is plausible that NPR-1 could function in other sensory neurons or interneurons to allow for maintaining reversal length. Our data suggest that FLP-18 modulates the activity of reversal circuits at command interneuron and sensory neuron levels simultaneously. Finally, our data also suggest that shorter reversals during starvation could be mediated by the FLP-18/NPR-1/4 signaling pathway ( Fig. 4B; illustrated in Fig. 5).
We also studied the reversal circuit through calcium imaging during exploratory behavior. Calcium imaging of AVA in flp-18 mutants showed similar patterns as observed in npr-1 npr-4 double mutants. Surprisingly, calcium levels of AVA during reversals in npr-1 and npr-4 mutants were comparable and significantly lower than flp-18 and npr-1 npr-4 mutants (Fig. 2). These results suggest two important points. First, FLP-18 functions through NPR-1 and NPR-4 supporting our behavioral experiments. Second, the FLP-18/NPR-1/4 system appears to modulate the reversal circuit through inhibitory signaling. Together, these observations point to a possible role of FLP-18 in modulating reversal length by regulating the duration of activity and calcium levels of AVA through NPR-4 and NPR-1. Our work provides molecular insights into the modulation of the locomotory circuit by showing that the FLP-18/NPR-1/NPR-4 signaling pathway plays a critical role in modulating locomotion-based behaviors under various genetic and environmental conditions.