Serotonin (5-HT) stimulates both pharyngeal pumping and egg laying in Caenorhabditis elegans. Four distinct 5-HT receptors have been partially characterized, but little is known about their function in vivo. SER-7 exhibits most sequence identity to the mammalian 5-HT7 receptors and couples to a stimulation of adenyl cyclase when expressed in COS-7 cells. However, many 5-HT7-specific agonists have low affinity for SER-7. 5-HT fails to stimulate pharyngeal pumping and the firing of the MC motorneurons in animals containing the putative ser-7(tm1325) and ser-7(tm1728) null alleles. In addition, although pumping on bacteria is upregulated in ser-7(tm1325) animals, pumping is more irregular. A similar failure to maintain “fast pumping” on bacteria also was observed in ser-1(ok345) and tph-1(mg280) animals that contain putative null alleles of a 5-HT2-like receptor and tryptophan hydroxylase, respectively, suggesting that serotonergic signaling, although not essential for the upregulation of pumping on bacteria, “fine tunes” the process. 5-HT also fails to stimulate egg laying in ser-7(tm1325), ser-1(ok345), and ser-7(tm1325) ser-1(ok345) animals, but only the ser-7 ser-1 double mutants exhibit an Egl phenotype. All of the SER-7 mutant phenotypes are rescued by the expression of full-length ser-7∷gfp translational fusions. ser-7∷gfp is expressed in several pharyngeal neurons, including the MC, M2, M3, M4, and M5, and in vulval muscle. Interestingly, 5-HT inhibits egg laying and pharyngeal pumping in ser-7 null mutants and the 5-HT inhibition of egg laying, but not pumping, is abolished in ser-7(tm1325);ser-4(ok512) double mutants. Taken together, these results suggest that SER-7 is essential for the 5-HT stimulation of both egg laying and pharyngeal pumping, but that other signaling pathways can probably fulfill similar roles in vivo.
SEROTONIN (5-HT) regulates key processes in nematodes, including pharyngeal pumping and egg laying. In Caenorhabditis elegans, four 5-HT receptors have been characterized: 5-HT1- (SER-4), 5-HT2- (SER-1), and 5-HT7-like G-protein-coupled receptors (GPCRs; SER-7) that appear to couple to Gαi/o, Gαq, and Gαs, respectively, as well as a novel 5-HT-gated Cl− channel (MOD-1) (Olde and Mccombie 1997; Hamdan et al. 1999; Ranganathan et al. 2000; Hobson et al. 2003). However, the physiological role of these receptors is unclear.
Feeding in C. elegans consists of two processes: the synchronous contraction of the corpus, anterior isthmus, and terminal bulb, and peristalsis of the posterior isthmus (Avery and Horvitz 1987). Serotonin stimulates pumping from ∼60 to 250 pumps/min and increases the frequency and decreases the duration of action potentials in pharyngeal muscles (Horvitz et al. 1982; Raizen et al. 1995; Niacaris and Avery 2003). These effects are dependent on both the MC and M3 pharyngeal motorneurons and both appear to be regulated directly by 5-HT (Raizen et al. 1995; Niacaris and Avery 2003). However, the 5-HT receptors involved have not been identified. The M4 motorneuron synapses posteriorly on the pm5 muscles of the pharynx and is necessary for isthmus peristalsis (Albertson and Thompson 1976; Avery and Horvitz 1987). Laser ablation of the M4 motorneuron results in failure of the pm5 muscles to contract and larval arrest due to starvation (Avery and Horvitz 1989). On the basis of these observations, it appears that the M4 motorneuron plays a more important role in the regulation of isthmus peristalsis than in regulating the overall rate of pumping.
Serotonin also stimulates egg laying in C. elegans. The egg-laying machinery includes 16 muscles (8 vulval and 8 uterine) and 2 types of neurons, the hermaphrodite-specific neurons (HSNs) and ventral type C neurons (VCs), both of which appear to be serotonergic (White et al. 1986; Rand and Nonet 1997). Multiple 5-HT receptors are involved in the regulation of egg laying. Vulval muscle appears to express a Gαq-coupled 5-HT receptor (SER-1) that is coupled to the entry of external Ca2+ through EGL-19, an L-type Ca2+ channel (Shyn et al. 2003; Dempsey et al. 2005). 5-HT switches the muscle from an inactive to an active state capable of contraction (Waggoner et al. 1998). In contrast, the HSNs, which synapse onto the vulval muscle, appear to express Gαi/o-coupled receptors that inhibit neurotransmitter release and egg laying (Bany et al. 2003; Shyn et al. 2003).
In this study, we have identified an essential role for a C. elegans 5-HT7-like receptor (SER-7) in the regulation of both pharyngeal pumping and egg laying, suggesting that Gαs-coupled signaling pathways may play a more important regulatory role than previously appreciated. Activation of SER-7-dependent, Gαs-coupled, signaling pathways could stimulate these processes directly, for example, by increasing the PKA-dependent phosphorylation of key channel proteins and/or indirectly by opposing the action of Gαi/o-mediated inhibition of Gαq-stimulated pathways, as has been described for the Gαq/Gαi/o-signaling network in the regulation of neurotransmitter release (Nurrish et al. 1999).
MATERIALS AND METHODS
Culture and maintenance of strains and reagents:
General techniques for culture and handling of worms have been described (Brenner 1974). The putative ser-7 knockout strains, tm1325 and tm1728, were obtained from the National Bioresources Project (Tokyo Women's Medical University, Tokyo), the ser-7(tm1325) ser-1(ok345) (DA2109) double mutant was obtained from Leon Avery, and all other strains were obtained from the Caenorhabditis Genetics Center (Oklahoma Medical Research Foundation, Oklahoma City, OK). The N2 Bristol strain was used as a control and is referred to as “wild type” in the text. All experiments used well-fed adults grown at 20° on standard nematode growth medium (NGM) seeded with the Escherichia coli strain OP50. ser-7(tm1325) and ser-7(tm1728) were outcrossed six times and three times, respectively. The ser-7(tm1325);ser-4(ok512) double mutant was constructed using standard techniques and confirmed by PCR.
Green fluorescent protein (GFP) reporter constructs were obtained from Andrew Fire (Carnegie Institute of Washington, Washington, DC). cAMP kits were purchased from Assay Designs (Ann Arbor, MI). [3H]-LSD (specific activity 86 Ci/mmol) was obtained from Du-Pont– New England Nuclear (Boston). All other chemicals were purchased from Sigma-Aldrich (St. Louis).
Creation of native and mutant ser-7b-receptor constructs for expression in mammalian cells:
The cloning of native ser-7b for mammalian expression was described previously (Hobson et al. 2003). To create a chimeric SER-7B receptor with GFP fused in the predicted third-intracellular loop of the receptor (see Figure 1B), primers were designed on either side of the predicted third-intracellular loop [ser-75′F-(CCCGGGTATGGCCCGTGCAGTCAACATATC), ser-75′R (GGTACCAACCGTTCGGAGGCATCCAC), ser-73′F (GAATTCCCACGAAATGGATCAGCTG) and ser-73′R (GCCGGCCTAGACGTCACTTGCTTCGTG)] and used to clone the ser-7b cDNA. Restriction sites for XmaI, KpnI, EcoRI, and NgoMIV, respectively, used to subclone the PCR products into the GFP expresson vector pPD117.01 and create the chimeric construct pOT7I3, are indicated in italic. A C. elegans cDNA pool prepared from all life stages was used as a template. To prepare ser-7b with E314A/K316A mutations in the third-intracellular loop (pOT7I3M), the GeneTailor site-directed mutagenesis system was used as per manufacturer's instructions (Invitrogen, Carlsbad, CA). The following primers were used for mutagenesis [SER7Mut-F (CAGACGGAGAAAAGTGcATGCgcAGCCCGGAAAAC) and SER7Mut-R (ACTTTTCTCCGTCTGCCGAATCAATGGTCCTTTGCA)]. Mutagenized nucleotides are indicated in lowercase. The ser-7b cDNA constructs were cloned into pFLAG-CMV2 (Sigma-Aldrich) by amplification with the following primers [SER-7CDsF (GCGGCCGCGATGGCCCGTGCAGTCAACATATC) and SER-7CDsR (TCTAGACTAGACGTCACTTGCTTCGTG)]. NotI and XbaI restriction sites used to subclone the ser-7b cDNA into pFLAG-CMV2 are noted in italic. All constructs were confirmed by sequencing (MWG Biotech, High Point, NC).
Ligand binding and coupling:
All constructs were transiently transfected into COS-7 cells using lipofectamine 2000 (Invitrogen), essentially as described by Hobson et al. (2003). Radioligand binding was assayed as described previously (Huang et al. 1999, Hobson et al. 2003). Briefly, crude membrane preparations (15 μg protein) were incubated with [3H]-LSD (2–60 nm) and Bmax and Kd values were calculated as described below. For inhibition binding, membranes (30 μg) were incubated with various concentrations of ligand and 10 nm [3H]-LSD. Nonspecific binding was determined in the presence of a 1000-fold excess of cold LSD. Saturation and inhibition binding were determined at room temperature after 1 hr in the dark. Reactions were terminated by filtration onto 96-well microplates with GF/B filters (Packard) previously soaked with 0.3% polyethyleneimine. Filters were then washed three times with ice-cold TEM (50 mm Tris.HCl, 0.5 mm EDTA, 50 mm MgCl2; pH 7.4) buffer, dried overnight, and the radioactivity quantified by liquid scintillation counting. All radioligand binding data were analyzed by nonlinear regression analysis using Deltagraph (Deltagraph, version 4.05e; Deltapoint). An F-test was used to determine if the data best fit a one- or two-site-binding model. To assay coupling, COS-7 cells, transfected as described above, were incubated with ligand and cAMP levels were determined as previously described (Hobson et al. 2003).
Creation of rescue constructs:
The rescue constructs are illustrated in Figure 1 and the corresponding transgenic lines are designated as follows: ser-7∷gfp (erEx1517), ser-1∷gfp (erEx1617). The ser-1∷gfp and ser-7∷gfp constructs are full-length translational fusions with sequence coding for GFP inserted in the predicted C terminus and third-intracellular loops of the receptors, respectively.
All rescue constructs were created by overlap fusion PCR (Hobert 2002). To create the full length ser-7∷gfp translational fusion with sequence coding for GFP inserted into the predicted third-intracellular loop of the receptor (exon III, in the same position as the chimeric cDNA construct), three different PCR fragments were fused by two rounds of PCR. Primers were used to amplify (1) the 5′-end of the gene and half of exon III [A3ILPF and BFL5R (CCGTTCGGAGGCATCCACAG)], (2) the 3′-end of the gene including the other half of exon III and 100 nt downstream of the 3′-UTR from a C. elegans genomic DNA preparation [CFL3F (CCACGAAATGGATCAG CTGAG) and DFL3R (CACCTTCTGAAATGCACATAATTGC)], and (3) a portion of pOT7I3 spanning a portion of the third-intracellular loop including gfp. The fragment of the third-intracellular loop containing gfp contained a 150-nt overlap on either side of the gfp with both the 5′ and 3′ PCR fragment, so when combined and amplified using A3ILPF and DFL3R, a full-length gene product with GFP inserted into exon III and the ser-7 3′-UTR was generated. The construct was injected into ser-7(tm1325) animals with pRF4 to create the transgenic strains, erEx1517.
To create the full-length ser-1∷gfp rescue construct with sequence coding for GFP inserted into the predicted C terminus of the receptor, three PCR fragments were fused by two rounds of PCR. Primers were designed to amplify (1) the 5′-end of the gene including part of exon IV [(SER-1promexonIVF: 5′-CGTTTTCACGGACCTCCTCCCACAC) and (SER-1promexonIVR: 5′- CGACGCAACCCAATCAGGAATCTGC)], (2) the 3′-end of the gene and 3′-UTR with part of exon IV [(SER-1exonIVF: 5′-GACTCTCATCAAATGTATTCGCGA) and (SER-13'UTRR: 5′-GTGTGAGGTGTGCGGCAAATCTAC)], and (3) a portion of pOT17S1, a plasmid containing the full-length ser-1 cDNA with sequence for gfp inserted into exon IV, spanning the C-terminal tail of the receptor and GFP [(SER-1exonIVGFPF: 5′- CCGCAACTCAAGATCTTGCAA and SER-1exonIVGFPR: 5′- GTGTACGTATCCGGAACAATTG)]. The fragment of the C-terminal tail containing GFP contained a 100-nt overlap either side of gfp with the 5′ and 3′ PCR fragments, so when combined and amplified using SER-1promexonIVF and SER-13'UTRR a full- length ser-1 gene with the ser-1 3′-UTR and GFP in the tail of the receptor was created. The construct was injected in ser-1(ok345) animals to create erEx1617.
Overlap fusion PCR constructs (at 0.05–5 ng/μl) were co-injected with the rol-6 plasmid (pRF4) into C. elegans gonads by standard techniques (Kramer et al. 1990; Mello and Fire 1995). Lines expressing the construct were screened by observing the roller phenotype and by GFP expression using a confocal-inverted laser-scanning microscope (Olympus-Fluoview, Olympus America, Melville, NY). At least three different overlap fusion PCR reactions were injected and five or more lines carrying the arrays were analyzed for each construct.
All behavioral assays were performed on young adult N2's synchronized by growing fourth-stage larvae at 20° for 24 hr prior to assay. For 5-HT-stimulated pumping, worms were placed on plates containing 13 mm 5-HT (creatinine sulfate salt) in the presence or absence of Escherichia coli OP50, left for 15 min, and the number of pumps (defined as backward grinder movements in the terminal bulb) was counted for 30 sec and then multiplied by two to obtain pumps per minute. Due to the relative impermeability of the nematode cuticle, high concentrations of 5-HT have been used routinely by C. elegans researchers to examine 5-HT-dependent phenotypes. To assay pumping on bacteria, either pumps were counted for 30 sec or single animals were followed for 10 min and pumps/10 sec were counted every 30 sec. For the latter assay to measure “fast pumping” on bacteria, at least 7 animals per genotype were assayed and the results pooled. Basal pumping rates were measured as described by Keane and Avery (2003). Egg laying was assayed in synchronized animals that had already begun to lay eggs on bacteria. Worms were placed on plates containing OP50 alone, 13 mm 5-HT alone, or both 5-HT and OP50 and eggs laid were counted after 1 hr. To determine the number of eggs in the uterus, i.e., to determine if the mutant animals retained eggs and exhibited an egg-laying defective (Egl)-phenotype, fourth-stage larvae were grown at 20° for 24 hr, when they had already begun to lay eggs on bacteria, and the number of eggs in the uterus were counted after mounting the worms on 2% agarose pads on microscope slides. All assays were performed at 20°–22°.
Pharyngeal pumping and egg laying were analyzed using Student's t-test (Student, 1908). To analyze variation in the pumping assay, where pumps/10 sec were recorded every 30 sec for 10 min from individual animals, a Kolmogorov-Smirnov test, which makes no assumptions about the distribution of the data, was used.
Electropharyngeograms (EPGs) were recorded essentially as described by Raizen and Avery (1994). Briefly, worms were placed on slides containing 10 mm 5-HT in Ascaris saline (4 mm NaCl, 125 mm sodium acetate, 24.5 mm KCl, 5.9 mm CaCl2, 4.9 mm MgCl2, 5 mm hepes, pH 7.4; del Castillo and Morales 1967) and recordings were begun by sucking the head of the worm into the tip of a low-impedance glass micropipette (2–6 MΩ). Pipettes were pulled from Corning 8161 patch clamp glass with ID 1.65 mm, OD 1.20 mm (Warner Instrument, Hamden, CT) by a P-87 Flaming/Brown micropipette puller (Sutter Instruments, Novato, CA) to have tip openings between 20 and 40 μm. Signals were amplified ×100 and low-pass filtered at 1 kHz with an Axoprobe-1A microelectrode amplifier (Molecular Devices, Sunnyvale, CA) and an FLA-01 amplifier (Cygnus Technologies, Southport, NC). All data were recorded and stored on a two-channel scientific digital tape recorder (DT-200, Microdata Instruments, S. Plainfield, NJ) and analyzed using Matlab (The Mathworks, Natick, MA).
ser-7 encodes a 5-HT7-like receptor with a unique pharmacology:
We previously reported the partial characterization of a C. elegans 5-HT receptor (SER-7B; Hobson et al. 2003). In this study, we have expanded the pharmacological characterization of SER-7B, and more importantly, evaluated its potential function in animals carrying the putative ser-7 null alleles, tm1325 and tm1728. Multiple SER-7 isoforms have been identified in the C. elegans EST database that differ in their C termini (Figure 1A). When transiently expressed in COS-7 cells, SER-7B exhibits a pharmacological profile that differs from its closest mammalian homolog, the 5-HT7 receptor, even though both receptors couple to a 5-HT-dependent increase in cAMP levels (Figure 2). For example, SER-7B has a low affinity for many specific mammalian 5-HT7 receptor ligands, including methysergide and SB-269970 (Table 1) and the binding potencies for SER-7B expressed in COS-7 cells are different to mammalian 5-HT7 receptors (Table 2). Interestingly, gramine, a serotonin receptor antagonist that can block firing of the inhibitory pharyngeal motorneuron M3 in response to 5-HT has a pKi identical to that of 5-HT (Niacaris and Avery 2003). However, it is worth noting that nematode-specific cell lines are not available and, although nematode receptors expressed in mammalian cells appear to couple to the G-protein(s) predicted on the basis of sequence alignments, no studies have yet confirmed that the pharmacology of a cloned nematode GPCR heterologously expressed in mammalian cells mimics the pharmacology exhibited in vivo.
5-HT does not stimulate pharyngeal pumping in putative ser-7 null mutants:
To clarify the physiological role of SER-7, the effects of exogenous 5-HT were examined in the putative ser-7 null mutants, ser-7(tm1325) and ser-7(tm1728). ser-7(tm1325) has a 742-bp deletion in exon V that results in the deletion of a portion of transmembrane VII and most of the C terminus of the receptor (Figure 1A). ser-7(tm1728) has a 666-bp deletion also resulting in the loss of exon V (Figure 1A). Therefore, ser-7(tm1325) and ser-7(tm1728) are predicted null alleles, as transmembrane VII is required for ligand binding and the C terminus is required for effective G-protein coupling and receptor trafficking (Cheung et al. 1989; Duvernay et al. 2004).
In wild-type animals, exogenous 5-HT dramatically stimulated pharyngeal pumping (Figure 3). In contrast, 5-HT had little effect on pharyngeal pumping in either ser-7(tm1325) or ser-7(tm1725) mutants (Figure 3). Since pumping in both mutants appeared to be dramatically upregulated when the animals were maintained on bacteria (Figure 3; see below), but not 5-HT, serotonergic signaling may be redundant in the regulation of pumping, and other signaling pathways, possibly mediated by neuropeptides, also might be involved. Indeed, a role for neuropeptides in the regulation of nematode pharyngeal pumping has been suggested previously (Brownlee et al. 1995; Rogers et al. 2001). In fact, 5-HT actually inhibited pumping on bacteria in the ser-7 null mutants (Figure 3), perhaps through presynaptially localized 5-HT autoreceptors, as discussed more fully below.
Pharyngeal pumping on bacteria is upregulated, but more variable, in ser-7 null mutants:
To examine the upregulation of pharyngeal pumping on bacteria more carefully, an assay was developed in which the pumping rates of individual animals (pumps/10 sec) were measured on bacteria every 30 sec for 10 min (Figure 4). In wild-type animals, bacterially stimulated pumping was consistent and rates/10 sec exhibited a modest variation around the mean (from 32 to 48 pumps/10 sec, mean 43 ± 0.3 pumps/10 sec; Figure 4A). In contrast, while pumping on bacteria also was dramatically increased in ser-7(tm1325) animals, rates were more variable than those observed in wild-type worms, resulting in slightly lower overall mean pumping rates (5–48 pumps/10 sec, mean 38.3 ± 0.6 pumps/10 sec; Figure 4B). A similar, but more pronounced response was observed in tph-1(mg280) animals that lacked tryptophan hydroxylase, an enzyme required for the biosynthesis of 5-HT (Figure 4G). Previous reports suggested that tph-1(mg280) animals did not increase their rate of pharyngeal pumping on bacteria (Sze et al. 2000). However, in our assay, tph-1(mg280) worms were capable of elevating their pumping rates to near wild-type levels, but only for brief periods. Most of the time, they pumped at significantly lower rates (2–48 pumps/10 sec, mean 30.4 ± 0.6 pumps/10 sec; Figure 4G).
The inability to maintain a “fast-pumping” phenotype was more pronounced in tph-1(mg280) than in the ser-7(tm1325) animals, suggesting that other 5-HT receptors might be involved. At least two distinct 5-HT receptors are expressed in the pharynx, a Gαq-coupled 5-HT receptor in pharyngeal muscle (SER-1; Tsalik et al. 2003) and SER-7 in the pharyngeal nervous system (Hobson et al. 2003; see below). Therefore, a putative ser-1 null mutant ok345 also was examined for 5-HT and bacterially stimulated pumping (Figure 4D). ser-1(ok345) has an 859-nt deletion and 25-nt insertion removing exon IV that in turn deletes a portion of the third-intracellular loop (10 aa), the sixth and seventh transmembrane domains, and a portion of the C terminus (41 aa) from the predicted receptor, suggesting that the receptor was nonfunctional (Figure 1A). ser-1(ok345) mutants, in contrast to the ser-7(tm1325) animals, dramatically increased pumping in response to 5-HT, indicating that SER-1 is not the major 5-HT receptor involved in the 5-HT-dependent upregulation of pumping (Figure 3). However, the ser-1(ok345) animals also exhibited a more variable pumping rate on bacteria (mean 39.2 ± 0.5 pumps/10 sec; Figure 4D) than wild-type animals. As predicted, the altered pumping in ser-7(tm1325) ser-1(ok345) double mutant was nearly identical to that observed in the tph-1(mg280) animals (Figure 4, F and G). Taken together, these data suggest that serotonergic signaling is not essential for the upregulation of pharyngeal pumping on bacteria, but may play a role in “fine tuning” the process. Most importantly, the pumping phenotypes observed in the ser-1(ok345) and ser-7(tm1325) animals could be rescued using the corresponding full-length ser-7∷gfp or ser-1∷gfp translational fusions, respectively (see Figure 1A for rescue constructs and Figures 3 and 4 for the rescue data).
Localization of ser-7 expression and rescue of 5-HT-stimulated pharyngeal pumping in ser-7(tm1325) animals:
Initial attempts to rescue 5-HT-stimulated pharyngeal pumping in ser-7(tm1325) animals using the minimal ser-7 promoter 2 kb upstream of the ATG and a ser-7b cDNA with sequence coding for GFP inserted into the third-intracellular loop were unsuccessful (0.5–0.005 ng/μl), as transgenic worms could not be isolated. Hobson et al. (2003) have noted previously that this minimal ser-7 promoter drove expression exclusively in the M4. The insertion of GFP into the third-intracellular loop had no apparent effect on coupling, as SER-7B and SER-7BI3GFP had nearly identical EC50s for the 5-HT- dependent stimulation of cAMP levels, when the receptors were transiently expressed in COS-7 cells (30 ± 3 nm vs. 47 ± 5 nm, respectively, Figure 2). In addition, the expression of a modified SER-7B (SER-7B E314A/K316AI3GFP) designed to couple less effectively to G-proteins by mutation of key amino acids in the third-intracellular loop of the receptor (Figure 1B; Obosi et al. 1997), also failed to generate robust transgenic lines under the control of the same promoter . As predicted, when SER-7B E314A/K316AI3GFP was expressed in COS-7 cells, the 5-HT stimulation of cAMP levels was dramatically reduced compared to SER-7B, with an EC50 of 1.8 ± 0.5 μM and a maximal activation of less than half that of SER-7B, suggesting that SER-7B E314A/K316I3GFP was functional, but attenuated (Figure 2).
In contrast to most C. elegans introns, the first three introns in the ser-7 gene are large, 0.7, 3.3 and 1.5 kb, respectively, suggesting that they might contain elements important for the regulation of expression (Figure 1A). Therefore, the rescue of 5-HT- dependent pumping in ser-7(tm1325) animals was attempted with a PCR fusion encoding the full-length ser-7 gene with sequence coding for GFP inserted into the predicted third-intracellular loop of the receptor (ser-7∷gfp). As noted above, the insertion of GFP into this position of the receptor has little apparent effect on G-protein coupling (Figure 2). ser-7∷gfp DNA injected into C. elegans at 0.5–0.05 ng/μl produced lines that exhibited marked GFP fluorescence in a number of pharyngeal neurons, including the MCs, M4, I2s, I3, M5, and occasionally in the M3s, I4, I6, and M2s, supporting a general neuromodulatory role for 5-HT in the regulation of pumping and feeding (Figure 5A). Strong GFP fluorescence also was observed in the vulval muscles (Figure 5B). However, these lines exhibited a stuffed pharynx phenotype and did not carry effectively. When ser-7∷gfp DNA was injected from 0.05 to 0.005 ng/μl, GFP fluorescence decreased dramatically and was barely detectable at the lower expression levels. However, under these conditions, 5-HT dramatically stimulated pharyngeal pumping in the rescued ser-7(tm1325)(+) animals (Figure 3). For example, basal rates of pumping were similar in wild-type (WT), ser-7(tm1325) and ser-7(1325)(+) animals, and were stimulated over fivefold by 5-HT in both the wild-type and rescued ser-7(tm1325)(+) worms. Similarly, these rescued ser-7(1325)(+) animals exhibited a wild-type pumping phenotype on bacteria (Figure 4C).
SER-7 regulation of pharyngeal pumping appears to involve the MC pharyngeal motorneuron:
To investigate the possible mechanism of the SER-7-mediated stimulation of pharyngeal pumping EPGs, which measure changes in pharyngeal muscle membrane potential, were examined from wild-type, ser-7(tm1325) and ser-7(tm1325)(+) animals (Figure 6A). In response to exogenous 5-HT, EPGs from WT worms exhibit two initial positive transients, called E-phase transients (Raizen et al. 1995; Figure 6A). E1 appears to result from input from the MC motorneurons and E2 from the depolarization of the pharyngeal muscle (Raizen et al. 1995). For example, laser ablation of the MC results in the inability of the worm to increase pharyngeal pumping and a loss of E1 from the EPG (Raizen et al. 1995). Similarly, EPGs from ser-7(tm1325) worms incubated with 5-HT also lack an E1 transient, suggesting that cholinergic input from the MC to the pharyngeal muscle is absent in these worms (Figure 6A). For example, EPGs from nearly all (>90%, 200 pumps examined from 10 worms; Figure 6B) of the wild-type and ser-7(tm1325)(+) animals contained an E1 spike, whereas EPGs from <5% of the ser-7(tm1325) (200 pumps examined from 10 worms; Figure 6B) animals had an E1 spike. The EPGs from the ser-7(tm1325) animals also exhibited small positive transients similar to those observed in MC− worms or eat-2 mutants that express altered nicotinic acetylcholine receptors on their pharyngeal muscles (Raizen et al. 1995; Figure 6A inset). The EPG data along with the observation that SER-7 is expressed in the MCs suggests that SER-7 mediates the increase in the overall rate of pharyngeal pumping by increasing the firing of the MC motorneurons.
SER-7 is also essential for 5-HT-stimulated egg laying:
Since SER-7 was robustly expressed in vulval muscle, ser-7(tm1325) and ser-7(tm1728) animals were examined for 5-HT-stimulated egg laying (Figure 7). In contrast to wild-type animals, 5-HT had no effect on egg laying in either ser-7(tm1325) or ser-7(tm1728) mutants and the 5-HT stimulation of egg laying was rescued in ser-7(tm1325) animals carrying a full-length ser-7∷gfp translational fusion (Figure 7). Previous studies have suggested a role for a Gαq-coupled 5-HT receptor in vulval muscle (Shyn et al. 2003). Indeed, the expression of SER-1 in vulval muscle also appears to be essential for 5-HT-stimulated egg laying and 5-HT-stimulated egg laying in ser-1(ok345) mutants lacking a functional SER-1 receptor could be rescued by the expression of SER-1 in the vulval muscle (Dempsey et al. 2005; Xiao and Komuniecki, personal communication).
The ser-7(tm1325), ser-7(tm1728) and ser-1(ok345) animals failed to exhibit any obvious Egl phenotype (Table 3), suggesting that serotonergic signaling may be at least partially redundant in the regulation of egg laying, as was observed above for pharyngeal pumping, and that other pathways, perhaps mediated by one of the many neuropeptide receptors in the C. elegans genome, can also stimulate egg laying (Schinkmann and Li, 1992; Waggoner et al. 2000). In contrast, ser-7(tm1325) ser-1(ok345) double mutants exhibited a modest, but statistically significant, Egl phenotype, suggesting that both 5-HT receptors may couple to at least some common downstream target(s) (Table 3). Interestingly, as observed above for pumping, 5-HT also appeared to dramatically inhibit normal egg laying on bacteria in ser-7 null mutants (Figure 7). In contrast, 5-HT did not inhibit egg laying on bacteria in ser-7(tm1325);ser-4(ok512) double mutants and actually stimulated egg laying in these animals when applied alone (Figure 7). As discussed more fully below, we hypothesize that SER-4 may function presynaptically to inhibit the release of neurotransmitters involved in the stimulation of egg laying. Finally, ser-4(ok512) mutants appeared to retain fewer eggs than wild-type animals, and also appeared to be slightly hyperactive for egg laying in the presence of 5-HT (Table 3; Figure 7).
Serotonin regulates a number of key processes in nematodes, including pharyngeal pumping, locomotion, and egg laying, as well complex behaviors such as the enhanced slowing response on bacteria (Horvitz et al. 1982; Sawin et al. 2000). Indeed, 5-HT appears to prime starved C. elegans for the exploitation of abundant food sources. In this study, we have demonstrated that a predicted Gαs-coupled 5-HT receptor (SER-7) is essential for the 5-HT stimulation of both pharyngeal pumping and egg laying. SER-7 is most closely related to mammalian 5-HT7 receptors, on the basis of sequence alignments and the observation that both receptors couple to Gαs when expressed in mammalian cells. However, as predicted, the pharmacological profiles of the two receptors differ significantly. For example, SER-7 has very low affinity for 5-CT, a potent agonist of mammalian 5-HT7 receptors.
Mammalian 5-HT7 receptors couple to Ca2+/calmodulin-sensitive adenyl cyclase isoforms and are expressed in both the CNS and peripheral tissues. 5-HT7 receptors have been implicated in the modulation of a variety of processes, including the relaxation of vascular smooth muscle, circadian rhythms, REM sleep, hypothermia, and migraine (Thomas and Hagan 2004). In rat hippocampal neurons, 5-HT7 receptors reduce the period of hyperpolarization between periods of neuronal excitability by enhancing the hyperpolarization-activated current Ih and inhibiting Ca2+-activated K+ channels (Gill et al. 2002). These observations fit well with proposed roles for SER-7 in the neuromodulation of pharyngeal pumping and egg laying discussed below.
Surprisingly, SER-7 does not appear to be necessary for the stimulation of either pharyngeal pumping or egg laying under physiological conditions, when the worms are on bacteria, suggesting that serotonergic signaling may be redundant and/or overlapping with other signaling pathways. However, both SER-7 and SER-1 appear to “fine tune” the elevated rate of bacterially stimulated pumping. Interestingly, previous reports suggested that 5-HT was essential for the upregulation of pharyngeal pumping on bacteria, on the basis of studies with tph-1 mutants that lack tryptophan hydroxylase and have dramatically reduced 5-HT levels (Sze et al. 2000). In contrast, the results of our study suggest that this tph-1 mutant can indeed upregulate pumping on bacteria, albeit less efficiently than wild-type worms, but does not maintain this “fast-pumping” phenotype.
5-HT facilitates rapid pharyngeal muscle contraction–relaxation cycles through its direct action on the MC and M3 motorneurons (Raizen et al. 1995; Niacaris and Avery 2003). The MCs are excitatory cholinergic neurons that depolarize pharyngeal muscle and regulate the rate of pumping through nicotinic cholinergic receptors expressed on pharyngeal muscle (Raizen et al. 1995). Ablation of the MCs causes a drastic reduction in both 5-HT and bacterially stimulated pumping. More importantly, EPGs from 5-HT- stimulated MC− worms reflect the lack of MC input. Typically, the excitation phase (E phase) of the EPG consists of two transients associated with the depolarization of the pharyngeal muscle. In 5-HT-stimulated MC− and ser-7(tm1325) worms, only a single E-phase transient is observed, suggesting that input from the MCs is compromised (Raizen et al. 1995; Figure 6). Taken together, these data suggest that SER-7 plays a direct role in the 5-HT stimulation of acetylcholine release from the MCs and that Gαs-coupled pathways may play a more general role in regulating the neurotransmitter release from presynaptic neurons than previously described.
The regulation of egg laying in C. elegans has been studied extensively, through laser ablation of VC4/5, inactivation of the HSNs in egl-1 mutants, measurements of Ca2+ transients, the rescue of individual receptor mutants, and a wealth of pharmacological studies. However, no role for SER-7 or for any Gαs-coupled receptor in the regulation of egg laying has been described. Both the VCs and HSNs secrete multiple neurotransmitters, including acetylcholine, 5-HT, and at least one, but probably several, related peptides (Schinkmann and Li 1992; Rand and Nonet 1997). Acetylcholine can either stimulate egg laying, as demonstrated by the addition of the nicotinic agonist levamisole or inhibit egg laying, by the activation of Gαi/o-coupled muscarinic receptors (GAR-2?) on the HSNs that inhibit neurotransmitter release (Weinshenker et al. 1995; Bany et al. 2003). Physiologically, the overall effects of acetylcholine appear to be inhibitory (Bany et al. 2003). 5-HT acts directly on vulval muscle to increase the rate and alter the temporal pattern of egg laying, switching the muscle from a pattern of sporadic Ca2+ activity to a pattern of continuous activity (Waggoner et al. 1998; Shyn et al. 2003). 5-HT increases the frequency of Ca2+ transients in vulval muscle through a pathway that appears to involve a Gαq-coupled 5-HT receptor (SER-1), Gαq (EGL-30), PKC (TPA-1), and a voltage-gated L-type Ca2+ channel (EGL-19).
The expression of both SER-1 and SER-7 in vulval muscle is essential for 5-HT stimulated egg laying, but the precise roles of the two receptors remain to be determined. It is clear that the inactivation of either SER-1 or SER-7 alone does not have any gross effect on the egg-laying system, as the effects are 5-HT-specific and the mutation of either receptor alone does not yield any obvious Egl phenotype. Presumably, 5-HT signaling in vulval muscle is at least partially redundant, as hypothesized above for pharyngeal pumping. For example, the egg-laying neurons secrete a number of different FARPs and their role in the regulation of egg laying has been predicted (Schinkmann and Li 1992). However, ser-7 ser-1 double mutants do exhibit an Egl phenotype, suggesting that at least some of the downstream effectors of the SER-1 and SER-7 mediated pathways may overlap. Either or both SER-1 and SER-7 may be involved in the phosphorylation of the vulval muscle L-type Ca2+ channel, EGL-19. EGL-19 is essential for 5-HT-dependent vulval muscle Ca2+ transients and egg laying, as partial loss-of- function egl-19 mutations strongly interfere with the ability of 5-HT to stimulate both processes (Shyn et al. 2003). Most importantly, neither the VC nor HSN egg-laying neurons are required for these effects, confirming a direct role for EGL-19 in vulval muscle (Shyn et al. 2003).
In mammals, multiple GPCRs act through cAMP/PKA pathways to regulate cardiac L-type Ca2+ channels. For example, the robust upregulation of ICa by the β-adrenergic receptor/cAMP/PKA pathway is well documented and appears to require both the specific subcellular localization of PKA through AKAP and the phosphorylation of the EGL-19-like, α1c-channel subunit (Gao et al. 1997). Indeed, EGL-19 contains a number of consensus PKA phosphorylation sites and the 5-HT-dependent stimulation of SER-7 may directly regulate the phosphorylation state of EGL-19 and the activity of Ca2+ channel. A similar but less well-documented role for PKC-mediated phosphorylation of cardiac L-type Ca2+ channels also has been described and might explain the dependence of 5-HT-dependent egg laying on SER-1 and EGL-30 (Gαq) (Trautwein and Hescheler 1990). However, while the direct PKA/PKC-dependent phosphorylation of EGL-19 is the most straightforward model, a number of other scenarios are possible, including the phosphorylation of vulval muscle K+ channels.
Surprisingly, 5-HT dramatically inhibited both pharyngeal pumping and egg laying when ser-7 null animals were placed on bacteria. As noted above, 5-HT failed to stimulate either pumping or egg laying in ser-7(tm1325) or ser-7(tm1728) animals, but both processes were dramatically stimulated when these mutant animals were placed on bacteria. The mechanism of this 5-HT inhibition is unclear, but may involve the feedback inhibition of neurotransmitter release mediated by presynaptic 5-HT autoreceptor(s) (SER-4/MOD-1?) on key neurons regulating these processes. Since ser-7(tm1325) or ser-1(ok345) animals are not Egl and lay eggs and elevate pumping on bacteria, other redundant receptors, perhaps mediated by neuropeptides, must be operating to elevate both egg laying and pumping in these mutants. The exogenous application of 5-HT in these egg-laying experiments may be inhibiting the release of these other stimulatory neurotransmitters. Although 5-HT dramatically stimulates egg laying, both acetylcholine and 5-HT also appear to inhibit neurotransmitter release from the HSNs by the presynaptic activation of Gαo-coupled pathways. For example, Bany et al. (2003) have demonstrated that acetylcholine released from the VCs inhibits egg laying by inhibiting the presynaptic release of 5-HT and presumably other neuromodulators from the HSNs by activating a Gαo (GOA-1)-coupled muscarinic receptor (GAR-2?) on the HSNs. Similarly, Shyn et al. (2003) demonstrated that 5-HT released from the HSNs can feedback inhibit through GOA-1 and silence the HSNs, although the 5-HT autoreceptor involved was not identified. The role of presynaptic GOA-1 (C. elegans Gαo) signaling in inhibiting neurotransmitter release in ventral cord neurons is well documented. The presynaptic activation of GOA-1 appears to move UNC-13, a protein required for synaptic vesicle priming, away from neurotransmitter release sites (Nurrish et al. 1999). More importantly, the expression of a constitutively active GOA-1 in the HSNs was sufficient to induce an Egl-phenotype (Moresco and Koelle 2004). Similarly, the expression of a constitutively active Gαo-coupled GPCR, EGL-47, in the HSNs also dramatically inhibited egg laying, suggesting that GOA-1 integrates signaling from multiple receptors in the HSNs (Moresco and Koelle 2004).
As predicted from these observations, 5-HT did not inhibit egg laying on bacteria in ser-7;ser-4 double mutants. In contrast, 5-HT still dramatically inhibited pumping in the ser-7;ser-4 double mutants, suggesting that additional 5-HT receptors are probably involved. These studies are continuing with the intention of identifying the downstream effectors linked to SER-7 and the signaling pathways involved in the 5-HT inhibition of pumping and egg laying.
We thank Dominique Durand for the use of his electrophysiology equipment, Leon Avery for the ser-7(tm1325) ser-1(ok345) double mutant, Shohei Mitani at the National Bioresources Project for the ser-7(tm1325) and ser-7(tm728) mutants, and the Caenorhabditis elegans Genetics Center, which is supported by the National Institutes of Health National Center for Research Resources, for additional strains. We also thank John Plenefisch and Sushant Khandekar for critical discussion of the work. This work was supported by National Institutes of Health grant AI-145147 awarded to R.W.K.
Communicating editor: K. Kemphues
- Received April 20, 2005.
- Accepted September 20, 2005.
- Copyright © 2006 by the Genetics Society of America