Mechanosensory Inputs Influence Caenorhabditis elegans Pharyngeal Activity via Ivermectin Sensitivity Genes

Corresponding author: John Keane, School of Biological Science, University of Sussex, Falmer, Brighton BN1 9QG, United Kingdom., j.keane{at}sussex.ac.uk (E-mail)

Communicating editor: P. ANDERSON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Mechanical stimulation induces opposite behavioral responses in the adult and dauer pharynx. Tail tap of adults inhibits pharyngeal pumping via a pathway involving the innexin gene unc-7 and components of the glutamatergic pathway encoded by the genes avr-14 and avr-15. Tail tap of dauers stimulates pumping through a mechanism involving Go and Gq. The nematocidal drug ivermectin is believed to kill worms by opening a glutamate-gated chloride channel (AVR-15) on pharyngeal muscle, causing complete pumping inhibition. However, ivermectin can also inhibit pumping in the absence of this channel. We propose that one of the ways ivermectin could prevent pumping, in the absence of the AVR-15 ivermectin-binding channel on pharynx muscle, is to target AVR-14 and AVR-15, which are expressed in the inhibitory pathway linking mechanosensation and pumping activity.

NEMATODE feeding is essential for survival. Knowledge of mechanisms that inhibit feeding is important for the development of drugs that can destroy populations of these medically and agriculturally important organisms. The relative simplicity of its body and ease of genetic manipulation in Caenorhabditis elegans have provided insights into the genetics of development and nervous system structure and function. C. elegans is related to pest nematodes and discovery of mechanisms used to inhibit feeding in this species should generalize to other species.

C. elegans is a small bacteria-eating soil nematode. Feeding is achieved by a pumping motion of the pharynx. Muscle contraction sucks fluid and bacteria into the lumen of the pharynx and relaxation expels the water and captures the bacteria. Captured bacteria are passed from the corpus in the anterior of the pharynx through the isthmus in a swallowing motion and into the terminal bulb where they are crushed by a grinder and passed into the intestine. The pharyngeal nervous system consists of 20 neurons. It is connected to the extrapharyngeal nervous system via gap junctions between the bilaterally symmetric RIP and I1 neurons. Although pumping can occur after complete ablation of the pharyngeal nervous system (AVERY and HORVITZ 1989 ), ablation studies have identified 4 neurons that are important for the regulation of the pump motion.

Genetic (RAIZEN et al. 1995 ) and laser ablation studies (AVERY and HORVITZ 1989 ) show that the pair of MC motor neurons controls the rate of excitation of pharyngeal muscle and positively influences adult pumping rates. Acetylcholine is implicated as the excitatory neurotransmitter released from MC onto pharyngeal muscle and nicotinic receptors on the muscle generate the excitatory postsynaptic potential. Immunological examination failed to show choline acetyltransferase in the MCs (Rand AND NONET 1997 ) but four observations suggest that they are nevertheless cholinergic. First, the behavioral effects of MC loss by laser ablation are mimicked by mutants of acetylcholine synthesis (cha-1, choline acetyltransferase) and packaging (unc-17, vesicle-associated acetylcholine transporter; AVERY and HORVITZ 1990 ) and by lesions in a gene encoding a pharyngeal muscle nicotinic acetylcholine receptor (nAchR) subunit (eat-2; J. MCKAY, personal communication). Second, contraction of pharyngeal muscle can be induced by application of nicotine, a nAchR agonist (RAIZEN et al. 1995 ). Third, an unc-17::GFP reporter transgene is expressed in the MCs (data not shown). Fourth, pumping rates can be reduced by application of D-tubocurarine, a nAchR antagonist (RAIZEN et al. 1995 ). The pair of inhibitory M3 motor neurons controls the timing of repolarization of pharyngeal muscle, which signals the end of a pump. M3 is believed to be glutamatergic on the basis of observations that glutamate pulses onto pharyngeal muscle imitate the effect of M3s and mutants of a pharyngeal muscle glutamate receptor subunit gene (avr-15) resemble M3-ablated worms (DENT et al. 1997 ). The NSM neurons are serotonergic (HORVITZ et al. 1982 ). Although ablation of these neurons has little measurable effect on behavior, exogenous serotonin accelerates pumping and suppresses locomotion. Serotonin from the NSMs may signal to the body that food is abundant and alter behavior appropriately. Neuron M4 regulates the swallowing motion of the isthmus, and it is the only pharyngeal neuron essential for worm survival. Details of pharyngeal anatomy are described in ALBERTSON and THOMSON 1976 and its behavior is summarized in AVERY and THOMAS 1997 .

Pumping is suppressed in wild-type worms during the dauer larval stage and in response to a touch stimulus in adults. Despite being a robust behavior, little is known about the pharynx/touch response circuit other than the fact that the RIP/I1 connection between the pharyngeal and extrapharyngeal nervous systems is required (CHALFIE et al. 1985 ). Inhibition of pumping during the dauer stage is biologically significant. Dauer larvae are a special developmental stage that forms in response to adverse environmental and nutritional conditions. Dauers exhibit numerous morphological and developmental modifications (CASSADA and RUSSELL 1975 ) that are designed to aid survival and possibly also dispersal to new food sources. Perhaps inhibition of pharyngeal activity could contribute to survival by reducing energy loss due to pumping when there is no food to eat or by reducing the risk of ingesting environmental toxins under conditions where nothing is to be gained by feeding anyway. A cuticle plug forms over the dauer mouth and may help to prevent desiccation and would also prevent feeding even if pumping was not suppressed (RIDDLE and ALBERT 1997 ).

To understand how pumping is inhibited we observed the effect of a touch stimulus on pharyngeal activity in two developmental stages: adults, which normally pump continuously in the presence of food, and dauers, which do not. We have looked at a number of mutants to identify genetic components required for this behavior. Our results indicate that (1) dauers and adults exhibit opposite behavioral output in the pharynx/touch response, (2) signaling via gap junctions and glutamate is important in adult and dauer pumping inhibition, and (3) a stimulatory pathway involving Gq/Go competes with the inhibitory pathway in both developmental stages.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Worm handling and nomenclature:
Unless otherwise stated, worms were grown at 20° using NGMSR plates seeded with Escherichia coli DA837 (DAVIS et al. 1995 ). Conventional C. elegans nomenclature is used (HORVITZ et al. 1979 ). Bristol N2 (BRENNER 1974 ) were used as wild type.

Mutations:
The following mutations were used.

LG I: unc-29(e1072am); egl-30(n686sd); avr-14(nr781, ad1305, nr391); daf-8(e1393ts); eat-18(ad1110); goa-1(n1134, n363).

LG II: tph-1(mg280); eat-2(ad465, ad570); cat-2(e1112); eat-2(ad570); adEx1818[rol-6(d) myo-2::eat-2⁺].

LG III: daf-7(e1372ts); eat-4(ad572, ky5); unc-25(e156); glr-1(n2461).

LG IV: osm-3(p802); mec-3(e1338); cha-1(p1186ts, md39ts).

LG V: avr-15(ad1051, nr785, nr395); glc-1(pk54::Tc1).

LG X: unc-7(e5, ad1565); unc-1(e719); mec-7(n434, e1343).

Adult pump rate assays:
One day prior to testing, L4 hermaphrodites were picked from an uncrowded nonstarved plate to a fresh seeded plate where they developed into adults at 20°. On the day of testing, adults were singly picked to an unseeded NGMSR plate briefly to remove excess E. coli from their cuticle. A number of clean adults were then transferred to another unseeded NGMSR plate and allowed to recover for 5 min at room temperature. The pump rate of a single adult was counted for 20 sec. Single pumps were scored as a complete backward movement of the terminal bulb grinder. The worm was then tapped once on the posterior one-third of its body (subsequently referred to as a tail tap) and the pump rate was recorded for another 20 sec. The worm was then discarded. A given individual therefore gave rise to two pump rate scores, one in which the worm is not tapped and the other where it is tapped on its posterior. Adults go through cycles of pumping and not pumping on plates in the absence of food. Pump rates were scored only from adults that were actively pumping at the start of the recorded period. Adults were scored within 40 min of being placed on an unseeded plate; after this time the plate was discarded. Pumps were counted at room temperature (22–25°).

Dauer pump rate assays:
Dauers for testing were identified according to these criteria: thin and dense body, presence of alae, and radial constriction of isthmus and terminal bulb (RIDDLE and ALBERT 1997 ). Except where specified, dauers were picked from starved plates grown at 20° and pump rates were scored within 2 days after the appearance of dauers on the plate. A platinum pick was used to transfer individual dauers from the starved plate to a fresh NGMSR plate whose surface was completely covered with E. coli (dauers leave tracks in the food, making them easier to find). The surface of the plate was immersed in light white oil (Sigma, St. Louis; M3516) to aid visual inspection of the pharynx. After 10–15 min recovery at room temperature, the pump rate of a single dauer was counted over 2 min. The same dauer was then heavily tapped once on the posterior one-third of its length with a pick and its pump rate was scored over 2 min. The dauer was then discarded. Plates were discarded when dauers had been on them for 40 min. Pumps were counted using a Wild M410 dissecting microscope (magnification, x64 and x20 eyepieces) at room temperature. Some special conditions were employed when testing the conditional cha-1(ts) mutants.

cha-1(ts) were grown at the permissive temperature of 15° until plates starved. Dauers were picked from plates for testing within 1 month of inoculating the plate. To study cha-1(ts) dauers under restrictive conditions, starved plates were transferred from 15° to 25° and left overnight. Dauers were then treated in the same way as any other strain. When cha-1(ts) dauers left at 25° overnight are returned to 15° on food plates they resume normal development.

Recording pumping activity per second:
Adults: unc-29(e1072am) was used instead of wild type, which move too fast to score using this approach. unc-29 encodes a levamisole-sensitive nicotinic acetylcholine receptor (FLEMING et al. 1993 ) expressed in body wall muscles and has no effect on adult pumping activity (RAIZEN et al. 1995 ). Adults were prepared and tested as described above with the following modification. Single worms were scored for 20 sec pretapped and posttapped. Worms were observed by use of a dissecting microscope placed next to a computer. Depressing a single key on the computer keyboard when the worm pumped recorded the time of pumping events over the course of the observation. After a population of worms was tested the average number of pumps was calculated for each consecutive second.

Dauers: Dauers were prepared as in Dauer pump rate assays and tested as in Recording pumping activity per second described above. These recordings could be carried out only on slow moving/uncoordinated dauers. In cases where a genotype to be tested was not slow it was combined with unc-29(e1072am) to produce a slow-moving double mutant. It is possible to measure pumping activity as a function of time using electropharyngeograms (EPGs; RAIZEN and AVERY 1994 ). Two factors prevented the use of EPGs for this purpose in this study. First, recorded EPG pump rates for any given strain were extremely variable, possibly influenced by random disruption of the dauer buccal plug due to suction of the dauer head. Second, we found that serotonin used to induce pumping also induced a rhythmic electrical activity in dauer EPG traces, which we call B-phase activity. This hindered identification of authentic pumps in a trace. On the basis of simultaneous EPG and video recording of dauers, B-phase activity appears to arise from pulsing of the excretory/secretory pore (NELSON and RIDDLE 1983 ).

Statistics:
The average percentage reductions in adult pump rates after tail tap shown in Fig 2 were normalized to take into account differences in initial pump rates between lines and were calculated as follows for each worm in each population:

View larger version (16K): In this window In a new window Download PPT slide   Figure 1. In the absence of food, tail tap reduces the relatively constant pump rate of adult worms. The effect of tail tap on adult unc-29 worms is shown. (A and C) Number of pumps per second over 20 sec before tail tap. (B) Pump rate of worms for subsequent 20 sec without tail tap. (D) Pump rate for subsequent 20 sec following tail tap. The average numbers of pumps during each 20-sec period are (A) 20.7 ± 2.08, (B) 20.8 ± 2.16, (C) 20.9 ± 1.91, and (D) 3.0 ± 0.81. Sixty individuals were tested in each case. D is significantly different from A–C (P < 0.001).

View larger version (33K): In this window In a new window Download PPT slide   Figure 2. Mutations in a number of neural signaling pathways confer resistance to the inhibitory effect of tail tap on adult pharyngeal activity. The average normalized percentage decrease in pump rate of mutant adults following harsh tail tap is shown. The pathways being examined are as follows: nos. 1 and 2, controls; 3–5, touch mutants; 6 and 7, Innexin unc-7; 8–16, glutamatergic pathway; 17–19, Go pathway; 20, GABA. *** indicates significant difference from control (P < 0.001). Wild type is the control in all cases except for nos. 9 and 18, which are compared to no. 2 (unc-29). No. 19 (unc-7; goa-1) is significantly different from no. 6 (unc-7; P < 0.001).

In certain mutant strains, some individuals pumped faster after tail tap. When this occurred, the normalized increase in pump rate was given a negative sign. The nonparametric Wilcoxon two-sample test was used to compare pairs of data sets (AVERY 1993 ; SOKAL and ROHLF 1995 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Pharyngeal activity responds to mechanical stimulus and is developmentally modified:
CHALFIE et al. 1985 found that light touch briefly inhibits pumping activity in adult worms. We have quantified the effect of harsh touch on pumping in an attempt to identify genetic components of the pathway in both adults and dauers. The pump rates of adult worms over a 20-sec period before and after a harsh tail tap are shown in Table 1. Wild-type (Table 1, no. 1) activity was 24.9 ± 1.41 pumps prior to tail tap and 4.1 ± 0.90 pumps after tail tap, which corresponds to a relative decrease in pump rate of 88.7 ± 1.98%. Under the conditions employed in this study, adult worms were found to have a relatively stable pump rate over the 20-sec period in the absence of mechanical stimuli. However, application of a tail tap almost completely inhibits pumping for a short time (3 sec) and then pumping recovers extremely slowly. These observations are presented in Fig 1 for unc-29 mutant adults, which behave similarly to wild-type N2 (Table 1, nos. 1 and 2). This behavior is extremely robust and we have used it to screen for mutants that resist pumping inhibition following tail tap (Fig 2 and Table 1).

The adult locomotory response to touch has been studied in detail in C. elegans (HUANG and CHALFIE 1994 ; WICKS and RANKIN 1995 ) and can be genetically and anatomically divided into separate neural circuits that mediate response to light and harsh touch. We tested two types of touch-response mutants to identify which circuit is involved in the pharynx/touch response. mec-7 mutants are defective in a ß-tubulin gene that is expressed in neurons responsible for gentle touch (CHALFIE and SULSTON 1981 ). mec-7 worms are insensitive to gentle touch. The two alleles of mec-7 that were tested (Table 1 and Fig 2, nos. 3 and 4) were not tap resistant. The gentle-touch neurons are not essential for this behavior but it is possible that they contribute to it redundantly with other circuits. WAY and CHALFIE 1989 discovered that response to harsh touch could be eliminated when the PVD neurons were killed in addition to the gentle-touch neurons. mec-3 encodes a homeodomain protein that is expressed in all of the touch neurons and in PVD. mec-3 mutants fail to express touch cell-specific structural genes, resulting in loss of light and harsh touch response. mec-3 is only slightly resistant to tail tap (Table 1 and Fig 2, no. 5), suggesting that the neural circuits underlying the locomotion/touch and pharynx/touch behaviors may not overlap. PVDs detect harsh touch to the central region of the body but adults lacking the PVDs can respond to harsh head and tail tap (WAY and CHALFIE 1989 ). The mechanosensory neurons involved are unknown.

In contrast to adult behavior, dauer larvae show an increase in pump rate after tail tap, as illustrated in Fig 3. Although the increase in tapped dauer pump rate is generally small it has been observed in every strain tested with a few notable exceptions. Mutation of eat-2 and eat-18 results in slow-pumping adults because of defective cholinergic signaling between motor neuron MC and the pharynx (RAIZEN et al. 1995 ). The eat-2-encoded nAchR subunit is expressed in pharyngeal muscle (J. MCKAY, personal communication) but the pharynx of eat-2 mutants remains sensitive to exogenous nicotine. eat-18 genetically interacts with eat-2 but eat-18 mutants are insensitive to exogenous nicotine. eat-2(ad570) and eat-18(ad1110) dauers (Table 2, nos. 24 and 25) exhibit slower untapped pumping rates than those of wt, suggesting that acetylcholine signaling persists at the pharynx even during the dauer stage. We do not know why the two eat-2 alleles exhibit different untapped pump rates in dauers (Table 2, nos. 24 and 26) despite having almost identical effects on adult pump rates. Expression of EAT-2 in pharyngeal muscle under the transcriptional control of the myogenic promoter myo-2 rescued eat-2(ad570) dauer pumping rates to wild-type levels (eat-2(ad570); adEx1818[rol-6(d) myo-2::eat-2⁺], untapped pump rate = 0.433 ± 0.128 pumps/min). Comparison of untapped and tapped dauer pumping rates (Table 2, nos. 24–26) reveals that the ability to pump faster after tail tap is lost in eat mutants. There may even be a tendency toward a reduced pump rate. We interpret these results to mean that cholinergic signaling from MC to the pharynx via nAchRs is necessary for the basal pumping rate in dauers and for the increase in pump rate following tail tap.

View larger version (12K): In this window In a new window Download PPT slide   Figure 3. Tail tap stimulates pharyngeal activity in dauers. Increased pump rate following harsh tail tap in dauers of four different genotypes is shown. Open bars, pump rate before tail tap; solid bars, pump rate after tail tap. Pretapped and posttapped pump rates are significantly different for nos. 23, 39, 45 (P < 0.05), and 44 (P < 0.1).

unc-7 is required for normal stimulus/response in adults and dauers:
Information about tail tap detected by the extrapharyngeal nervous system must pass to the pharyngeal nervous system either via connections made between the RIPs and I1s, which is the only direct neural link, or via a long-distance humoral pathway. The RIPs are coupled to the I1s by gap junctions. unc-7 encodes a gap junction channel subunit (STARICH et al. 1993 , STARICH et al. 1996 ) and unc-7 mutants are defective in both adult and dauer tail tap response. Tail tap does not inhibit adult pumping in unc-7 mutants to the same extent as wt (Fig 2 and Table 1, nos. 6 and 7). Some individual unc-7 adults actually pumped faster after tail tap, a response never seen in wt. unc-7 mutants also tend to have a faster untapped pump rate than that of controls (Table 1, no. 1 vs. nos. 6 and 7) but it is not commensurate with their resistance to tail tap pumping inhibition.

The unc-7 inhibitory pathway is also functional in dauers. The response of unc-7 dauers to tail tap is approximately six times greater than that of wt (Table 2, nos. 27 and 28). unc-7 allele ad1565 was originally identified in a screen for dauers that can pump fast. The unc-7 increase in pump rate is dependent, at least in part, on normal cholinergic communication between MC and pharyngeal muscle. The double mutants unc-7; eat-2 and unc-7; eat-18 (Table 2, nos. 29–31) do not show much of an increase in pump rate following tail tap. In fact, tail tap can actually reduce pump rates in these mutants. This indicates that the pathway in which unc-7 functions is upstream of, and inhibitory to, MC. unc-7; eat-2 and unc-7; eat-18 pump faster than eat-2 and eat-18 single mutants, suggesting that unc-7 also functions in an MC independent pathway. DENT et al. 2000 previously presented a model and some supporting data in which unc-7 is required for the flow of ivermectin-induced hyperpolarization from extrapharyngeal cells into the pharynx. Our results indicate that gap junctions are present in the pathway that transduces the mechanical stimulus from the tail to the pharynx. unc-7 may be required at a number of points along the pathway.

A glutamatergic pathway is involved in normal stimulus/response in adults and dauers:
Glutamate inhibits pharyngeal activity (LI et al. 1997 ) and glutamatergic neurons are present in the pharyngeal nervous system (DENT et al. 1997 ). We examined whether this transmitter is responsible for pump inhibition in tapped adults and untapped dauers. The eat-4 gene encodes a vesicular glutamate transporter (LEE et al. 1999 ) and eat-4 mutants exhibit no detectable glutamate signaling between M3 and pharynx muscle. Tail tap-induced pumping inhibition in eat-4 adults is not as pronounced as in controls (Fig 2, nos. 8 and 9 vs. nos. 1 and 2). The small effect of eat-4 on this behavior could indicate that glutamatergic signaling has a minor or mostly redundant role in this pathway. Another possibility is that glutamate signaling is relatively unperturbed at the required site. For instance, the observed loss of communication between M3 and pharyngeal muscle in eat-4 may arise because glutamate cannot be packaged into vesicles fast enough for it to keep up with the rapid rate of pharyngeal activity (RANKIN and WICKS 2000 ). However, neurons that are active only rarely, such as mechanosensory or interneurons, may be able to replenish their vesicular glutamate adequately to permit signaling with their postsynaptic partner. If glutamate is released from a synapse that is not frequently active in response to the single tail tap this could explain the small effect of eat-4 in this behavior.

Adults with lesions in any of three glutamate-gated chloride channel (GluCl) subunits, avr-14, avr-15, and glc-1 (DENT et al. 2000 ), are not very resistant to tail tap (Fig 2, Table 1, nos. 10–12), and neither are mutants of the AMPA-type glutamate receptor GLR-1 (HART et al. 1995 ; Fig 2, Table 1, no. 13). However, avr-14; avr-15 double mutants are quite resistant (Fig 2, Table 1, nos. 14–16). avr-14 is not expressed in the pharyngeal nervous system so it cannot be required for direct inhibition of pharynx muscle via M3. It is expressed in mechanosensory neurons, ring ganglia, and the ventral chord. eat-4 is also expressed in mechanosensory neurons, suggesting that mutations in the glutamatergic pathway may effect detection of harsh touch. eat-4 dauers, like wild type, exhibit a chronically suppressed pump rate, indicating that glutamate is not required for this long-term effect.

Go is required for normal response in dauers and adults:
The role of Go in pump/touch response in dauers is unusual. Go is encoded by the goa-1 gene (MENDEL et al. 1995 ) and null mutants do not form dauers. A reduction-of-function mutant, goa-1(n1134) (SEGALAT et al. 1995 ), produces some dauers on starved plates but they pump at wt rates (Table 2, no. 33). However, when goa-1(n1134) is combined as a double mutant with an uncoordinated (unc) mutation it pumps faster than controls both before and after tail tap (Table 2, nos. 34 and 36). Temporal analysis of the goa-1 response to tail tap shows an immediate rapid response in pump rate followed by a slow decay. The response is qualitatively and quantitatively different from unc-7, suggesting that separate pathways are involved (Fig 4). However, the goa-1 pathway is similar to unc-7 in its requirement for functional nAchRs (Table 2, no. 37 vs. no. 34). goa-1 adults are not resistant to tail tap-evoked pumping inhibition (Table 1, Fig 2, nos. 17 and 18) but an unc-7; goa-1 double mutant is more resistant than unc-7 alone (Table 1, Fig 2, no. 19 vs. no. 6), indicating that the goa-1-mediated stimulation of pumping is active in adults as well as dauers.

View larger version (27K): In this window In a new window Download PPT slide   Figure 4. Qualitative and quantitative differences in dauer tail tap response between unc-7 (A and B) and goa-1 (C and D) suggest that these genes function in different pathways. Graphs show pump rate per second over a 2-min period before (A and C) and after (B and D) tail tap. Thirty unc-7 dauers were tested for A and B. Sixty goa-1 dauers were tested for C and D. Bars show standard error of the mean. The strains used and average pump rates are unc-7(e5); daf-8(e1393ts), (A) 1.37 ± 0.389 pumps/min and (B) 7.02 ± 1.365 pumps/min; goa-1(n1134) unc-29(e1072am); daf-7(e1372ts), (C) 3.29 ± 0.704 pumps/min and (D) 16.76 ± 1.455 pumps/min.

Go and Gq often act antagonistically (HAJDU-CRONIN et al. 1999 ) and we found that a Gq mutant, egl-30(n686sd) (BRUNDAGE et al. 1996 ), eliminates the increase in pumping rate caused by tail tap in dauers (Table 2, no. 38). On the basis of this result we propose the following model. Harsh touch in adults and dauers causes release of an unknown neuromodulator that stimulates pumping and movement away from the stimulus via a G-protein-coupled receptor linked to Gq. Activation of Gq renders target tissues more responsive to acetylcholine signaling via nAchRs. In the absence of Gq there is no response, but in the absence of Go the response is large because nothing opposes the stimulation mediated by Gq. The response eventually decays due to receptor adaptation. Why do untapped unc-7; goa-1 and unc-29 goa-1 double mutants pump faster than the goa-1 single mutant? A possible explanation is that the presence of an unc causes the dauer body to bend and move and the worm perceives these as a light touch resulting in a small release of neuromodulator. We observe that unc; goa-1 double mutants move more often than controls and there is a positive correlation between movement and pumping (data not shown). All unc dauers tested (unc-1, unc-7, and unc-29) appear to pump faster than wt (Table 2, nos. 39, 28, and 35 vs. no. 23). Only unc-7 pumps significantly faster. The molecular products of these three loci are different and it is unlikely that all of them affect the dauer touch/pharynx response pathway.

Analysis of neurotransmitters potentially involved in the tail tap response:
Having ascertained that glutamate is unlikely to act directly on the pharynx to inhibit pumping in adults and dauers we wished to identify the neurotransmitter that did. We also wanted to identify the ligand responsible for signaling via the proposed Gq/Go pathway. A prime candidate for stimulating pumping in tapped dauers is serotonin because it stimulates pumping in adults (HORVITZ et al. 1982 ). tph-1(mg280) is a null mutation of the single tryptophan hydroxylase homolog of C. elegans, a key enzyme involved in serotonin biosynthesis, and tph-1 mutants exhibit defects in serotonin-dependent behaviors (SZE et al. 2000 ). tph-1 dauers pump slower than wt but show an increase in pump rate after tail tap (Table 2, no. 40 vs. no. 23). The untapped pump rate of unc-7; tph-1 dauers is the same as unc-7 alone, but the tapped pump rate is lower than that of unc-7 (Table 2, no. 41 vs. no. 28). The presence of tph-1 has a negligible effect on goa-1 (Table 2, no. 42 vs. no. 34). This could be interpreted in two ways: (1) that serotonin does increase pump rates in response to tap but it is not the agonist (or at least is not the only agonist) for the Gq/Go pathway or (2) that some serotonin remains in tph-1 mutants. If the former interpretation is true then an unknown neuromodulator must increase pump rates in tph-1 and unc-7; tph-1 mutants. Another good candidate is acetylcholine. A significant amount of data implicate it as the neurotransmitter at the MC/pharynx neuromuscular junction but it could also act elsewhere in the touch/pharynx pathway. Also, many of the existing egl-30 (Gq) alleles were identified in a screen for mutants that suppress the effects of muscarinic acetylcholine signaling in the pharynx (BRUNDAGE et al. 1996 ). Acetylcholine is synthesized by choline acetyltransferase, an enzyme encoded by the cha-1 gene. We tested two conditional alleles of cha-1 and found that pumping prior to tail tap is completely absent after dauers are left at the restrictive temperature overnight. However, cha-1 mutants can pump after tail tap at rates similar to eat-2 and eat-18 (cha-1(md39ts), 0.13 + 0.063 pumps/min; cha-1(p1186ts), 0.07 ± 0.046 pumps/min), indicating that acetylcholine is probably not responsible for increasing pumping rate via Gq.

Exogenous application of octopamine (HORVITZ et al. 1982 ), dopamine (our personal observation), and -aminobutyric acid (GABA; AVERY and HORVITZ 1990 ) can suppress pharyngeal activity. Mutants reportedly deficient in the neuromodulators octopamine (osm-3; HORVITZ et al. 1982 ; Table 2, no. 43 vs. no. 44) and dopamine (cat-2; SULSTON et al. 1975 ; Table 2, no. 45 vs. no. 23) exhibit dauer behavior similar to control levels. Therefore, they are not individually responsible for pumping suppression throughout the dauer stage and are not agonists of the Gq/Go pathway. GABA does not mediate pumping inhibition in adults as tail tap of the unc-25 GABA-deficient mutant (MCINTIRE et al. 1993 ) effectively inhibits pumping (Table 1, Fig 2, no. 20). We cannot exclude the possibility that respective neurotransmitter levels are insufficiently low in some of these mutants to cause an observable defect. It is also possible that these neurotransmitters play a redundant role in pumping inhibition and individual roles cannot be detected using a single-mutant approach.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Inhibitory and stimulatory pathways control pharyngeal output after tail tap:
The behavioral output of harsh tail tap is opposite in adults and dauers. However, the inhibitory pathway identified in adults is active in dauers, as unc-7 and glutamatergic pathway mutant dauers can pump faster than controls. The stimulatory pathway identified in dauers involving Gq/Go is similarly active in adults, as unc-7; goa-1 adults are more resistant to tail tap than is unc-7 alone. The behavioral output observed in these developmental stages appears to reflect a shift in dominance between the two competing pathways (Table 3). Despite testing many individually and multiply mutant worms we failed to completely eliminate the effects of tail tap in adults and dauers, suggesting that many redundant pathways underlie these behaviors.

unc-7 and glutamate may function in the same pathway:
We have identified genetic components of an inhibitory circuit connecting harsh touch receptors in the posterior body to the pharynx. Mutants of the innexin gene unc-7 and various combined mutants involved in glutamate signaling are resistant to pumping inhibition to a similar extent (Fig 2, nos. 6 and 7 and 14–16). We propose that these mutants work in the same pathway but communicate the inhibitory signal at different points within the circuit. Glutamate and gap junctions communicate the same signal at the same mechanosensory synapse in other touch-response circuits (LEE et al. 1999 ). This is unlikely to be the case here because unc-7 and avr genes do not both have to be mutant to observe a large behavioral effect. Glutamate signaling may be required in the sensory neurons and unc-7 may form the gap junction linking the pharyngeal to the extrapharyngeal nervous system. If unc-7 is essential for coupling RIP to I1 we would expect complete loss of adult tail tap inhibition, but this is not observed. Either the unc-7 lesion does not completely uncouple RIP and I1 or other inhibitory systems exist.

What is the ligand for the Gq/Go stimulatory pathway?
Receptors for the neuromodulators serotonin and dopamine can signal via Gq in vertebrates (WATSON and ARKINSTALL 1994 ); however, neither of these chemicals appears to be an agonist of the Gq/Go system in the dauer pharynx. Muscarinic acetylcholine receptors can also signal through Gq. Our analysis of cha-1(ts) mutants suggests that this may not be the ligand for the Gq-coupled receptor. However, this result is not conclusive as there is a chance that some acetylcholine remains within a cholinergic neuron belonging to the touch/pharynx-response pathway even after worms have remained at the restrictive temperature overnight. When cha-1(ts) adults are placed under conditions conducive to rapid pumping, at the restrictive temperature, conditions that should rapidly deplete any available acetylcholine, some individuals can still pump very slowly after 5 hr. In addition, dauers kept at the restrictive temperature overnight can still move away from a touch stimulus, indicating that acetylcholine remains in locomotory motor neurons. This is not unexpected as dauers tend to remain motionless when left undisturbed, conserving available neurotransmitter. As acetylcholine is released at body wall neuromuscular synapses, permitting dauers to move away from the tail tap stimulus, it is possible that this distant release is detected by the pharynx and contributes to the increase in pump rates.

Suppressed pumping during the dauer stage is not due (solely) to constitutive activation of the inhibitory unc-7/glutamate pathway:
unc-7 and glutamate pathway mutant dauers exhibit the same long-term suppression of pharyngeal activity observed in wt dauers. In the presence of exogenous serotonin (50 mM) some wt dauers are capable of pumping at rates of 75 pumps/minute when measured by EPG recording. This is significantly faster than rates observed in the absence of drug when dauers are on NGMSR plates (Table 2, no. 23). The dauer pharynx is therefore capable of pumping quite fast but is prevented from doing so by a mechanism that does not involve the unc-7 or the glutamatergic pathways discussed here.

Why does touch affect pharyngeal activity?
We know of no reason, at least under laboratory conditions, why it would be beneficial for adult worms to cease pumping, or for dauers to start pumping, following harsh touch. One could propose that it may be in the adult worms' interest to briefly shut down any nonessential energy-consuming distractions following a shock such as harsh tail tap to focus on escape from the stimulus. However, in many cases using this stimulus the adult worm completely freezes both locomotion and pumping. So perhaps both of these phenotypes are the result of a general shock response capable of overriding all other activities. The Gq/Go-mediated pathway is required for efficient pharyngeal activity during nondauer stages of development and for normal locomotion; egl-30 mutants pump and move slowly and goa-1 mutants move rapidly. We speculate that the small increase in pump rate observed in dauers is due to reception of a signal primarily intended to activate the locomotory circuit.

Ivermectin may kill worms by hijacking the unc-7/glutamatergic inhibitory pathway:
The inhibitory pharynx/touch pathway discussed in this article has been studied using mutants previously reported to confer resistance to the nematocidal drug ivermectin (combinations of avr-14, avr-15, and unc-7). The most direct mechanism by which ivermectin kills worms is to open AVR-15-containing glutamate-gated channels in pharyngeal muscle, hyperpolarizing the organ, and thus preventing pumping and feeding. DENT et al. 2000 have proposed another model in which ivermectin also kills worms by inhibiting pumping. In this model ivermectin irreversibly opens a glutamate-gated chloride channel composed of avr-14 and/or avr-15 in an extrapharyngeal neuron. This results in chloride influx and neuron hyperpolarization. Hyperpolarization descends to the pharynx via gap junctions encoded by unc-7 and inhibits pumping. This proposal has been tested by laser ablation of neuron I1 thereby uncoupling the pharynx from neurons containing ivermectin-sensitive GluCls in the extrapharyngeal nervous system. Laser-operated worms were resistant to ivermectin (DENT et al. 2000 ). This model is intriguing because it suggests that I1 is coupled directly to pharyngeal muscle via gap junctions. unc-7; eat-2 dauers pump faster than eat-2 single-mutant dauers, suggesting that unc-7-mediated pumping inhibition is partly MC independent. Perhaps this inhibition is due to unc-7-dependent coupling of I1, or a connected neuron, to pharyngeal muscle.

	ACKNOWLEDGMENTS

eat-2(ad570) adEx1818 [rol-6(d) myo-2::eat-2⁺] was provided courtesy of Dr. Jim McKay. The Caenorhabditis Genetics Center (CGC) provided some of the strains described in this article. This work was supported by research grant HL46154 from the U.S. Public Health Service to Leon Avery.

Manuscript received March 3, 2002; Accepted for publication January 28, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALBERTSON, D. and J. THOMSON, 1976  The pharynx of Caenorhabditis elegans.. Philos. Trans. R. Soc. Lond. B Biol. Sci. 275:299-325.[Medline]

AVERY, L., 1993  Motor neuron M3 controls pharyngeal muscle relaxation timing in Caenorhabditis elegans.. J. Exp. Biol. 175:283-297.[Abstract]

AVERY, L. and R. HORVITZ, 1989  Pharyngeal pumping continues after laser killing of the pharyngeal nervous system of C. elegans.. Neuron 3:473-485.[Medline]

AVERY, L. and H. R. HORVITZ, 1990  Effects of starvation and neuroactive drugs on feeding in Caenorhabditis elegans.. J. Exp. Biol. 253:263-270.

AVERY, L., and J. H. THOMAS, 1997 Feeding and defecation, pp. 679–716 in C. elegans II, edited by D. L. RIDDLE. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

BRENNER, S., 1974  The genetics of Caenorhabditis elegans.. Genetics 77:71-94.[Abstract/Free Full Text]

BRUNDAGE, L., L. AVERY, A. KATZ, U.-J. KIM, and J. E. MENDEL et al., 1996  Mutations in a C. elegans Gq gene disrupt movement, egg laying and viability. Neuron 16:999-1009.[Medline]

CASSADA, R. C. and R. L. RUSSELL, 1975  The dauer larva, a postembryonic developmental variant of the nematode Caenorhabditis elegans.. Dev. Biol. 46:326-342.[Medline]

CHALFIE, M. and J. SULSTON, 1981  Developmental genetics of the mechanosensory neurons of C. elegans.. J. Neurosci. 15:2434-2444.

CHALFIE, M., J. E. SULSTON, J. G. WHITE, E. SOUTHGATE, and J. N. THOMSON et al., 1985  The neural circuit for touch sensitivity in Caenorhabditis elegans.. J. Neurosci. 5:956-964.[Abstract]

DAVIS, M. W., D. SOMERVILLE, R. Y. N. LEE, S. LOCKERY, and L. AVERY et al., 1995  Mutations in the Caenorhabditis elegans Na,K-ATPase alpha-subunit gene, eat-6, disrupt excitable cell function. J. Neurosci. 15:8408-8418.[Abstract]

DENT, J. A., M. W. DAVIS, and L. AVERY, 1997  avr-15 encodes a chloride channel subunit that mediates inhibitory glutamatergic neurotransmission and ivermectin sensitivity in Caenorhabditis elegans. EMBO J. 16:5867-5879.[Medline]

DENT, J. A., M. M. SMITH, D. K. VASSILATIS, and L. AVERY, 2000  The genetics of ivermectin resistance in Caenorhabditis elegans.. Proc. Natl. Acad. Sci. USA 97:2674-2679.[Abstract/Free Full Text]

FLEMING, T., C. TORNOE, H. RIINA, J. COADWELL, J. LEWIS et al., 1993 Acetylcholine receptor molecules of the nematode Caenorhabditis elegans, pp. 65–80 in Comparative Molecular Neurobiology, edited by Y. PICHON. Birkhauser Verlag, Basel, Switzerland.

HAJDU-CRONIN, Y. M., W. J. CHEN, G. PATIKOGLOU, M. R. KOELLE, and P. W. STERNBERG, 1999  Antagonism between Go and Gq in Caenorhabditis elegans: the RGS protein EAT-16 is necessary for Go signaling and regulates Gq activity. Genes Dev. 13:1780-1793.[Abstract/Free Full Text]

HART, A. C., S. SIMS, and J. M. KAPLAN, 1995  Synaptic code for sensory modalities revealed by C. elegans GLR-1 glutamate receptor. Nature 378:82-85.[Medline]

HORVITZ, H. R., M. CHALFIE, J. E. SULSTON, and P. D. EVANS, 1982  Serotonin and octopamine in the nematode Caenorhabditis elegans.. Science 216:1012-1014.[Abstract/Free Full Text]

HORVITZ, R., S. BRENNER, J. HODGKIN, and R. HERMAN, 1979  A uniform genetic nomenclature for the nematode Caenorhabditis elegans.. Mol. Gen. Genet. 175:129-133.[Medline]

HUANG, M. and M. CHALFIE, 1994  Gene interactions affecting mechanosensory transduction in Caenorhabditis elegans.. Nature 367:467-470.[Medline]

LEE, R. Y. N., E. R. SAWIN, M. CHALFIE, H. R. HORVITZ, and L. AVERY, 1999  EAT-4, a homolog of a mammalian sodium-dependent inorganic phosphate cotransporter, is necessary for glutamatergic neurotransmission in Caenorhabditis elegans.. J. Neurosci. 19:159-167.[Abstract/Free Full Text]

LI, H., L. AVERY, W. DENK, and G. P. HESS, 1997  Identification of chemical synapses in the pharynx of Caenorhabditis elegans.. Proc. Natl. Acad. Sci. USA 94:5912-5916.[Abstract/Free Full Text]

MCINTIRE, S. L., E. JORGENSEN, and H. R. HORVITZ, 1993  Genes required for GABA function in Caenorhabditis elegans.. Nature 364:334-337.[Medline]

MENDEL, J. E., H. C. KORSWAGEN, K. S. LIU, Y. M. HAJDU-CRONIN, and W. A. SIMON et al., 1995  Participation of the protein Go in multiple aspects of behaviour in C. elegans.. Science 267:1652-1655.[Abstract/Free Full Text]

NELSON, F. K. and D. L. RIDDLE, 1983  Functional study of the C. elegans secretory-excretory system using laser microsurgery. J. Exp. Zool. 231:45-56.

RAIZEN, D. and L. AVERY, 1994  Electrical activity and behaviour in the pharynx of Caenorhabditis elegans.. Neuron 12:483-495.[Medline]

RAIZEN, D., R. LEE, and L. AVERY, 1995  Interacting genes required for pharyngeal excitation by motor neuron MC in Caenorhabditis elegans.. Genetics 141:1365-1382.[Abstract]

RAND, J. B., and M. L. NONET, 1997 Synaptic transmission, pp. 611–643 in C. elegans II, edited by D. L. RIDDLE. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

RANKIN, C. H. and S. R. WICKS, 2000  Mutations of the Caenorhabditis elegans brain-specific inorganic phosphate transporter eat-4 affect habituation of the tap-withdrawal response without affecting the response itself. J. Neurosci. 20:4337-4344.[Abstract/Free Full Text]

RIDDLE, D., and P. ALBERT, 1997 Genetic and environmental regulation of dauer larva development, pp. 739–768 in C. elegans II, edited by D. L. RIDDLE. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SEGALAT, L., D. A. ELKES, and J. KAPLAN, 1995  Modulation of serotonin-controlled behaviors by G_o in Caenorhabditis elegans.. Science 267:1648-1651.[Abstract/Free Full Text]

SOKAL, R., and F. ROHLF, 1995 Biometry: The Principles and Practice of Statistics in Biological Research. W. H. Freeman, New York.

STARICH, T., R. LEE, C. PANZARELLA, L. AVERY, and J. SHAW, 1996  eat-5 and unc-7 represent a multigene family in Caenorhabditis elegans involved in cell-cell coupling. J. Cell Biol. 134:537-548.[Abstract/Free Full Text]

STARICH, T. A., R. K. HERMAN, and J. E. SHAW, 1993  Molecular and genetic analysis of unc-7, a Caenorhabditis elegans gene required for coordinated locomotion. Genetics 133:527-541.[Abstract]

SULSTON, J., M. DEW, and S. BRENNER, 1975  Dopaminergic neurons in the nematode Caenorhabditis elegans.. J. Comp. Neurol. 163:215-226.[Medline]

SZE, J. Y., M. VICTOR, C. LOER, Y. SHI, and G. RUVKUN, 2000  Food and metabolic signalling defects in a Caenorhabditis elegans serotonin-synthesis mutant. Nature 403:560-564.[Medline]

WATSON, S., and S. ARKINSTALL, 1994 The G-Protein Linked Receptor Factsbook. Academic Press, New York.

WAY, J. C. and M. CHALFIE, 1989  The mec-3 gene of Caenorhabditis elegans requires its own product for maintained expression and is expressed in three neuronal cell types. Genes Dev. 3:1823-1833.[Abstract/Free Full Text]

WICKS, S. R. and C. H. RANKIN, 1995  Integration of mechanosensory stimuli in Caenorhabditis elegans.. J. Neurosci. 15:2434-2444.[Abstract]

This article has been cited by other articles:

				T. Janssen, S. J. Husson, M. Lindemans, I. Mertens, S. Rademakers, K. V. Donck, J. Geysen, G. Jansen, and L. Schoofs Functional Characterization of Three G Protein-coupled Receptors for Pigment Dispersing Factors in Caenorhabditis elegans J. Biol. Chem., May 30, 2008; 283(22): 15241 - 15249. [Abstract] [Full Text] [PDF]

				B. Shtonda and L. Avery CCA-1, EGL-19 and EXP-2 currents shape action potentials in the Caenorhabditis elegans pharynx J. Exp. Biol., June 1, 2005; 208(11): 2177 - 2190. [Abstract] [Full Text] [PDF]