Embryo series courtesy of Einhard Schierenberg
| Home | About WormBook | Author Instructions | Sponsors | e-Alerts | Feedback | WormAtlas | WormBase |
Table of Contents
Abstract
The role of neuropeptides in modulating behavior is slowly being elucidated. With the sequencing of the C. elegans genome, the extent of the neuropeptide genes in C. elegans can be determined. To date, 113 neuropeptide genes encoding over 250 distinct neuropeptides have been identified. Of these, 40 genes encode insulin-like peptides, 31 genes encode FMRFamide-related peptides, and 42 genes encode non-insulin, non-FMRFamide-related neuropeptides. As in other systems, C. elegans neuropeptides are derived from precursor molecules that must be post-translationally processed to yield the active peptides. These precursor molecules contain a single peptide, multiple copies of a single peptide, multiple distinct peptides, or any combination thereof. The neuropeptide genes are expressed extensively throughout the nervous system, including in sensory, motor, and interneurons. In addition, some of the genes are also expressed in non-neuronal tissues, such as the somatic gonad, intestine, and vulval hypodermis. To address the effects of neuropeptides on C. elegans behavior, animals in which the different neuropeptide genes are inactivated or overexpressed are being isolated. In a complementary approach the receptors to which the neuropeptides bind are also being identified and examined. Among the knockout animals analyzed thus far, defects in locomotion, dauer formation, egg laying, ethanol response, and social behavior have been reported. These data suggest that neuropeptides have a modulatory role in many, if not all, behaviors in C. elegans.
Neuropeptides are short sequences of amino acids that function either directly or indirectly to modulate synaptic activity. In addition, neuropeptides may also function as primary neurotransmitters. As in mammalian systems, the number of predicted neuropeptides in C. elegans is well over one hundred (Li et al., 1999; Pierce et al., 2001); however, most of the neuropeptides fall into two large families: the insulin-like peptides (Pierce et al., 2001; Li et al., 2003) and the FMRFamide (Phe-Met-Arg-Phe-NH2)-related peptides or FaRPs, which are referred to as FLPs in C. elegans (Li et al., 1998; Li, 2005). The remaining non-insulin, non-FLP peptides are classified as the neuropeptide-like proteins or NLPs. The NLPs are a diverse group of neuropeptides that have little similarity among each other (Nathoo et al., 2001). With a few striking exceptions, the functions of the different neuropeptides remain largely unknown in C. elegans. This overview serves to summarize the current state of the neuropeptide field.
Neuropeptides are typically derived from larger precursor molecules, which undergo posttranslational processing and sometimes modifications to yield mature peptides (see Figure 1). A single neuropeptide precursor molecule can give rise to a single neuropeptide, multiple distinct neuropeptides, multiple copies of a single neuropeptide, or any combination thereof. As an additional mechanism to increase neuropeptide complexity in mammals, a single precursor molecule can be differentially cleaved to yield different sets of peptides in different cell types (Strand, 1999; Salio et al., 2006). Furthermore, in the mollusk Aplysia peptides from a single precursor molecule can be sorted into different nerve terminals (Sossin et al., 1990). Whether such differential processing and trafficking occurs in C. elegans is unclear.
Like mammalian precursor molecules (Steiner, 1998), the initial cleavages in C. elegans occur C-terminal to dibasic residues flanking the peptide sequence (Rosoff et al., 1993; Marks et al., 1995, 1997, 1998, 1999a, 2001; Husson et al., 2005, 2006); however, cleavages C-terminal to mono- and tribasic residues have also been reported (Rosoff et al., 1993; Marks et al., 1997, 2001). The enzymes responsible for the initial endoproteolytic cleavage are kex2/subtilisin-like proprotein convertases, which must themselves be cleaved to become active. Cleavage of proprotein convertase is dependent on a chaperonin protein SBT-1 7B2 (Lindberg et al., 1998; Sieburth et al., 2005). Four C. elegans proprotein convertases, kpc-1, egl-3/kpc-2, aex-5/kpc-3, and bli-4/kpc-4, are present in C. elegans (Thacker and Rose, 2000). egl-3/kpc-2 is expressed in many, but not all neurons in the nervous system (Kass et al., 2001). Loss of egl-3/kpc-2 results in defects in egg-laying, mechanosensation, and locomotion (Kass et al., 2001; Jacob and Kaplan, 2003), indicating that EGL-3/KPC-2 cleaves precursors whose peptides have diverse functions. Moreover, using a polyclonal anti-FMRFamide antibody that recognizes the Arg-Phe-amide but not the non-amidated Arg-Phe-OH moiety (Marder et al., 1987), Kass and co-workers (2001) found decreased FMRFamide-like immunoreactivity in egl-3/kpc-2 mutants, suggesting that egl-3/kpc-2 cleaves some, but not all FLP precursors and that other proprotein convertases are active in the same cells. Among the other proprotein convertases, a kpc-1 deletion mutant shows mild locomotory defects and slow growth, suggesting that KPC-1 cleaves precursors of peptides involved in movement and growth (Thacker and Rose, 2000). Mutations in aex-5/kpc-3 cause defecation defects (Thomas, 1990). aex-5/kpc-3 is expressed in muscle (Thacker and Rose, 2000), and has been proposed to cleave a precursor molecule in muscle to produce a peptide that serves as a retrograde signal to regulate exocytosis (Doi and Iwasaki, 2002). A major function of BLI-4/KPC-4 is to cleave procollagen into collagen so that it can be deposited in the cuticle to give the cuticle structural integrity. Hence, null alleles of bli-4/kpc-4 cause lethality (Thacker et al., 1995). However, transcripts of bli-4/kpc-4 are also expressed in the nervous system (Thacker et al., 1995; Thacker and Rose, 2000), although phenotypes associated with loss of bli-4/kpc-4 neural transcripts are unknown. Collectively, these data indicate that multiple proprotein convertases are active in neurons.
|
Figure 1. Processing of a neuropeptide gene product: flp-1 as an example. After translation of the flp-1A transcript, pre-pro-FLP-1A is cleaved by signal peptidase in the endoplasmic reticulum to release the signal peptide. The propeptide pro-FLP-1A is further cleaved C-terminal to mono-, di-, or tribasic residues (indicated by K and R) by proprotein convertases, such as EGL-3/KPC-2. The basic amino acids are removed by carboxypeptidases E, such as EGL-21, to yield the basic neuropeptides. The FLP-1A peptides are further modified by the addition of an amide group donated from the C-terminal glycine, a reaction which may be catalyzed by T19B4.1, to yield the active peptides. After release, peptides are removed from the synaptic cleft by proteolytic degradation, which may be mediated by NEP-1. With the exception of PNFMRFYamide, all flp-1A encoded peptides have been biochemically isolated (Li, 2005).
A peptidomic approach was also taken to determine the relative contribution of each proprotein convertase in neuropeptide processing (Husson et al., 2006). FLP and NLP peptides were isolated from wild type and different proprotein convertase mutants. kpc-1(gk8) and bli-4(e937)/kpc-4 mutants showed a similar peptide profile as wild type, suggesting that their contribution to peptide processing is minor or that the alleles examined, such as bli-4/kpc-4(e937), do not completely remove gene function. For two alleles of egl-3(n729 and gk238)/kpc-2 no peptides were isolated (Husson et al., 2006), confirming previous work that EGL-3/KPC-2 is the major active proprotein convertase in neurons (Kass et al., 2001). However, as previously reported (Kass et al., 2001), a residual level of FMRFamide-like immunoreactivity remained. Surprisingly, a much decreased peptide profile was seen in aex-5(sa23)/kpc-3 mutants (Husson et al., 2006), providing strong evidence that aex-5/kpc-3 is also responsible for neuropeptide precursor processing in neurons.
Once the precursor molecules are cleaved by the proprotein convertases, the basic residues themselves are removed from the peptide sequences by the activity of carboxypeptidase E. egl-21 encodes a neural-specific carboxypeptidase E that is expressed in about 60% of the neurons (Jacob and Kaplan, 2003). Loss of egl-21 carboxypeptidase E causes more severe phenotypes than those seen in the egl-3/kpc-2 proprotein convertase mutants. egl-21 null alleles show defects in egg laying, locomotion, mechanosensation, and defecation as well as extremely low levels of FMRFamide-like immunoreactivity (Jacob and Kaplan, 2003). These data suggest that cleavage by EGL-21, like EGL-3/KPC-2, yields peptides that function in multiple behaviors and that EGL-21 is one of the carboxypeptidases E that cleaves FLP precursor molecules. Two other carboxypeptidases are present in the C. elegans genome, but their roles in neuropeptide processing have not been investigated (Jacob and Kaplan, 2003).
To protect themselves from degradation, neuropeptides are commonly modified at the N- or C-terminus. In many circumstances, the modification also confers biological activity to the neuropeptide, including in C. elegans (Schinkmann and Li, 1992). The most common known modification in C. elegans is amidation. Based on the presence of a C-terminal glycine, which donates an amino group in the amidation process, all of the FLPs and many of the NLPs are likely to be amidated. In mammals two enzymes, peptidylglycine-alpha-hydroxylating monooxygenase (PHM) and peptidyl-alpha-hydroxyglycine alpha-amidating lyase (PAL) act sequentially to catalyze amidation; the enzymes are synthesized on the same molecule as adjacent domains on a bifunctional protein, peptidyl-α-hydroxyglycine α-amidating lyase (PAM; Eipper et al., 1993). C. elegans contains at least one PAM-like and one PHM molecule (Han et al., 2004). Decreased activity of T19B4.1, which encodes a monooxygenase, leads to resistance to the acetylcholinesterase inhibitor aldicarb, suggesting that T19B4.1 may be a PAM-like molecule that processes neuropeptides (Sieburth et al., 2005). Whether T19B4.1 and/or other enzymes are involved in neuropeptide amidation has not been determined.
Bioactive neuropeptides are located in dense core vesicles derived from the trans-Golgi network. By contrast, many of the classical small molecule transmitters are located in small, clear vesicles that are clustered at the synaptic zones. The processing of the neuropeptide precursor molecules starts in the endoplasmic reticulum with the removal of the signal peptide and continues in the Golgi complex and the dense core vesicles themselves as the vesicles are transported to the nerve terminal (Strand, 1999). UNC-104 kinesin is necessary for the transport of small, clear vesicles in C. elegans (Hall and Hedgecock, 1991) and can also function as the motor for dense core vesicles. Mutations in unc-104, for instance, cause an increase of FMRFamide-like immunoreactivity in neuronal cell bodies (Schinkmann, 1994; Jacob and Kaplan, 2003). Furthermore, fast, but not slow, anterograde transport of IDA-1, a transmembrane protein localized to dense core vesicles, is lost in unc-104 mutants, suggesting that at least two distinct motors can transport dense core vesicles (Zahn et al., 2004). Disruption of unc-116, which encodes a kinesin molecule distinct from UNC-104, decreases overall FMRFamide-like immunoreactivity (Schinkmann, 1994), suggesting that UNC-116 kinesin also plays a role in dense core vesicle trafficking.
In contrast to small clear vesicles, dense core vesicles are not localized at synaptic zones, but are more diffusely scattered around the nerve terminal (Strand, 1999; Salio et al., 2006). Whereas contents in small clear vesicles can be released by focal increases of calcium at the synaptic zone, release of neuropeptides from dense core vesicles appears to be dependent on a general increase of calcium throughout the nerve terminal, which can occur after high levels of stimulation (Strand, 1999; Salio et al., 2006). The exact mechanism for dense core vesicle movement to the cell membrane is unknown. Some mechanisms may be conserved for small clear and dense core vesicles. For instance, in C. elegans UNC-13 is necessary to prime both types of vesicles for release (Richmond et al., 1999; Sieburth et al., 2007). A cytoplasmic protein that promotes vesicle release by bridging between dense core vesicles and the plasma membrane is calcium-dependent activator protein (CAPS; Renden et al., 2001; Grishanin et al., 2002). Similarly, C. elegans UNC-31 CAPS also promotes dense core vesicle release (Sieburth et al., 2007); its activity appears to be modulated by IDA-1 (Cai et al., 2004). Mutations in pkc-1 protein kinase I cause increased punta fluorescence of neuropeptide precursor molecules, but not of GFP-SNB-1 puncta associated with synaptic vesicles, suggesting that PKC-1 is specifically necessary for dense core vesicle release (Sieburth et al., 2007). PKC-1 is expressed in the cholinergic ventral cord motor neurons and not the GABAergic motor neurons (Sieburth et al., 2007), indicating that other protein kinases function to promote dense core vesicle release in the GABAergic motor neurons. After release of the vesicle's contents, neuropeptides are cleared from the cleft by the action of proteolytic enzymes, one of which may be NEP-1 neprilysin (Sieburth et al., 2005). Hence, unlike small molecule transmitters, which can be recycled and re-loaded into synaptic vesicles, neuropeptides must be synthesized de novo in the cell body and transported down to the axon terminal.
In addition to their release at synapses, neuropeptides also act as hormones, i.e., as long range signaling molecules. Some of the first isolated mammalian neuropeptides, for instance, were hormones released from the pituitary, adrenal glands, and the gut (Strand, 1999). Similarly, neuropeptides in C. elegans released from neurons or non-neuronal cells (see below) may act as hormones (see below). To monitor expression of neuropeptides into the pseudocoelom, Sieburth et al. (2007) took advantage of the scavenger activity of the coelomocytes, which continuously endocytose fluid from the pseudocoelom (Fares and Grant, 2002), and monitored release of GFP-tagged neuropeptide precursors into the pseudocoelom by the appearance of GFP in the coelomocytes.
Before the sequencing of the C. elegans genome, one common method to identify neuropeptide candidates was to use different antibodies from the mammalian field to stain the C. elegans nervous system. Initial work reported immunoreactivity for cholecystokinin, Substance P, melanocyte-stimulating hormone, met-enkephalin, beta-endorphin, and possibly adrenocorticotopic hormone (S. McIntire, pers. comm.). Immunoreactivity was also detected for FMRFamide (Schinkmann and Li, 1992), which was initially isolated from invertebrates (Price and Greenberg, 1977) but for which related peptides were later found in mammals (Dockray, 2004). Subsequent sequencing of the C. elegans genome (C. elegans Sequencing Consortium, 1998) indicated that with the exception of Substance P- and FMRFamide-related peptides, none of these mammalian peptide families have been identified in C. elegans (Nathoo et al., 2001). Because of the similarity to the FLP neuropeptides, anti-cholecystokinin antibodies are likely to have cross-reacted with FLPs. The antigens to which the other mammalian antisera cross-reacted are unknown.
With the completion of the C. elegans genome, researchers were able to scan the genome for candidate genes encoding neuropeptides. Certain groups focused on specific neuropeptide families. For instance, several groups collectively identified forty genes that encode insulin-like molecules (see Table 1; Duret et al., 1998; Gregoire et al., 1998; Kawano et al., 2000; Pierce et al., 2001; Li et al., 2003). Our lab group identified twenty-four flp genes (flp-1 to flp-23 and flp-28) encoding peptides with a C-terminal RFamide moiety by cDNA isolation and BLAST searches (see Table 2; Li et al., 1998; Kim and Li, 2004; unpubl. obs.), while McVeigh and co-workers (2005; A. Maule, pers. comm.) used EST data mining to identify five additional flp genes, flp-24 to flp-27 and flp-32 (see Table 2). Husson and Schoofs (2007) identified flp-33 by isolating a FLP not encoded by any previously identified flp gene (see Table 2). Some of the flp genes also encode non-FLP peptides (see Table 2). Hart and co-workers used similarity and pattern-based scans in more general BLAST screens to identify other neuropeptide genes (Nathoo et al., 2001). Using the characteristics of neuropeptide precursor processing, the pattern-based scans were designed to search for peptide sequences that were flanked by mono- or dibasic sequences. 34 non-insulin-like, non-FLP-like genes were identified and are referred to collectively as the neuropeptide-like protein or nlp genes (Li et al., 1999; Nathoo et al., 2001; A. Hart, pers. comm.); an additional eight nlp genes were recently identified in a different screen and using peptidomic analysis (see Table 3; Couillault et al., 2004; Husson et al., 2005). A total of 113 neuropeptide genes encoding over 250 putative neuropeptides have now been identified in C. elegans (see Tables 1, 2 and 3).
What is striking about the neuropeptide genes is how many of them are clustered on a chromosome, suggesting that they arise from tandem gene duplications. In some cases, the neuropeptide genes are only a few hundred or thousand base pairs apart. For instance, within about 25,000 bp on chromosome I there are seven ins genes, ins-24 through ins-30, which do not appear to be part of an operon. Similarly, many of the nlp genes are also physically close (Nathoo et al., 2001). While the flp genes are clustered on certain chromosomes, particularly the X chromosome, only a few (e.g., flp-2, 3, and 28) are as physically close as some of the ins genes.
Because many of the flp neuropeptide genes encode multiple, distinct peptides, a nomenclature was developed to designate the distinct peptides produced by one gene. Each peptide is now designated by the gene name and a number. For instance, the flp-1 gene encodes eight distinct peptides designated as FLP-1-1, FLP-1-2, etc. (see Table 4). As more of the NLP and INS peptides are isolated and their sequences confirmed, these peptides may also be given specific designations.
Table 1. Neuropeptide genes encoding insulin-like peptides in C. elegans
|
Putative peptides | Expression Pattern | Function or Phenotype | Receptor | References## | |||||
|---|---|---|---|---|---|---|---|---|---|---|
|
|
ASI, ASJ, PQR, other neurons, hindgut, pharyngeal muscle, hypodermis | promotes repro-ductive growth | DAF-2 | 1 | |||||
|
|
ASI, ASJ, ASH, ADF, AIA, AIM, ASE, ASG, AWA, BAG, NSM, intestine, vulval muscles | DAF-2 antagonist? | DAF-2? | 2, 3, 4 | |||||
|
|
amphidial, labial, ventral cord, & tail neurons, pharynx, vulva | 2 | |||||||
|
|
amphidial, labial, lateral, ventral cord, & dorsal projecting neurons | 2 | |||||||
|
|
amphidial, labial, ventral cord, dorsal projecting, & tail neurons, hypodermis | DAF-2 | 2, 5 | ||||||
|
|
amphidial, labial, ventral cord, lateral projecting, & tail neurons, vulva | 2 | |||||||
|
|
amphidial, labial, ventral cord, & tail neurons | DAF-2 | 2 | ||||||
|
|
amphidial, labial, ventral cord, & tail neurons | 2 | |||||||
|
|
amphidial, labial, ventral cord, & tail neurons, vulva | 2 | |||||||
|
|
ASI, ASJ | Overexpression causes embryonic and larval arrest | DAF-2? | 2 | |||||
|
|
2 | ||||||||
|
|
labial, ventral cord, & tail neurons | 2 | |||||||
|
|
2 | ||||||||
|
|
2 | ||||||||
|
|
2 | ||||||||
|
|
2 | ||||||||
|
|
2 | ||||||||
|
|
2 | ||||||||
|
|
amphidial, ventral cord, tail, & pharyngeal neurons | DAF-2 antagonist? | DAF-2? | 2 | |||||
|
|
Overexpression causes larval arrest | DAF-2? | 2 | ||||||
|
|
2 | ||||||||
|
|
amphidial, ventral cord, & tail neurons | 2 | |||||||
|
|
amphidial, labial, ventral cord, lateral process projecting, & tail neurons | Modulates acetyl-choline signaling | 2, 6 | ||||||
|
|
amphidial, labial, & ventral cord neurons | 2 | |||||||
|
|
2 | ||||||||
|
|
2 | ||||||||
|
|
2 | ||||||||
|
|
2 | ||||||||
|
|
2 | ||||||||
|
|
2 | ||||||||
|
|
2 | ||||||||
|
|
Overexpression causes larval arrest$; Modulates acetyl-choline signaling | DAF-2? | 2$; 6 | ||||||
|
|
2$ | ||||||||
|
|
2$ | ||||||||
|
|
2, OST | ||||||||
|
|
2, OST | ||||||||
|
|
2 | ||||||||
|
|
2, OST | ||||||||
|
|
2 | ||||||||
|
|
2, OST | ||||||||
|
|
|||||||||
|
|
EST | ||||||||
| #Genes for which ESTs, ORFeomes (OST), or cDNAs have been isolated are in bold; encoded peptides are based on sequence homologies to other insulin-like peptides. Cosmid and LG data only indicated for ins-31 a; ins-31 b and ins-31 c are the same. $ Unclear which ins-31 construct was used for overexpression and functional data. See References. ##References are as follows: 1, Li et al., 2003; 2, Pierce et al., 2001; 3, Kodama et al., 2006; 4, Tomioka et al., 2006; 5, Kao et al., 2007; 6, Sieburth et al., 2005. EST or OST in EST or ORFeome databases indicated only if not identified in canonical reference and sequence spans at least one intron. Modified from Li (2005). | ||||||||||
Table 2. Neuropeptide genes encoding FMRFamide-related peptides (FLPs) in C. elegans
|
Putative peptides§ | Expression pattern† | Function or Phenotype | Receptor‡ | References## | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
|
AIA, AIY, AVA, AVE, AVK, RIG, RMG, M5 | involved in locomotion, egg laying, and fat deposition; SADPNFLRF-NH2 inhibits frequency of pharyngeal action potentials; modulates acetylcholine signaling | (C25G6.5, Y58G8A.4, C16D6.2, Y59H11AL.1) | 1-11, KA | ||||||||||||
|
|
AIA, RID, PVW, I5, MC (ASI, M4, head muscles, an extra pair of cells in the head) | T19F4.1a/b | 2, 9, 10, 12 | |||||||||||||
|
|
IL1, PQR; SP, CP9 | SAEPFGTMRF-NH2 inhibits frequency of pharyngeal action potentials | C53C7.1a (Y58G8A.4, C16D6.2) | 2,6-9, 12 | ||||||||||||
