Olfactory Plasticity Is Regulated by Pheromonal Signaling in Caenorhabditis elegans

See allHide authors and affiliations

Science  24 Sep 2010:
Vol. 329, Issue 5999, pp. 1647-1650
DOI: 10.1126/science.1192020


Population density–dependent dispersal is a well-characterized strategy of animal behavior in which dispersal rate increases when population density is higher. Caenorhabditis elegans shows positive chemotaxis to a set of odorants, but the chemotaxis switches from attraction to dispersal after prolonged exposure to the odorants. We show here that this plasticity of olfactory behavior is dependent on population density and that this regulation is mediated by pheromonal signaling. We show that a peptide, suppressor of NEP-2 (SNET-1), negatively regulates olfactory plasticity and that its expression is down-regulated by the pheromone. NEP-2, a homolog of the extracellular peptidase neprilysin, antagonizes SNET-1, and this function is essential for olfactory plasticity. These results suggest that population density information is transmitted through the external pheromone and endogenous peptide signaling to modulate chemotactic behavior.

Caenorhabditis elegans is attracted to a series of odorants, such as benzaldehyde (bz) (1). However, after extended exposure to the odor in the absence of food, worms stop approaching the odorant and disperse from it—a behavioral plasticity called olfactory adaptation or food-odor associative learning (hereafter called olfactory plasticity) (24). The odorant probably serves as a signal of remote food, and the naïve chemotaxis to the odorant is probably beneficial for C. elegans in nature, but if worms fail to reach food after a certain period of odor sensation they no longer approach the odorant. For the best chance of survival, animals may have to optimize the extent of this behavioral switching by assessing environmental conditions, including the population density of the species, which can critically affect the survival strategy (57). Over the course of our olfactory plasticity assays (fig. S1), we noticed that the extent of olfactory plasticity shows positive correlation with the density of animals on the cultivation plates (Fig. 1A).

Fig. 1

Olfactory plasticity is regulated by population density and dauer pheromone. (A) Two to 12 parental adults per plate were put on the culture plates, and the progeny was tested for olfactory plasticity (n ≥ 6 assays). (B) Olfactory plasticity of daf-22 mutants was tested with or without adding 0.03 unit/ml of crude dauer pheromone to the culture plates (n ≥ 6 assays). Open bars and filled bars represent mock preexposed condition and bz preexposed condition, respectively. *P < 0.01, **P < 0.001 for all figures; (B) Student’s t test and (A) Dunnett’s test were used for multiple comparisons. Error bars represent SEM in all figures.

Population density is known to regulate the developmental decision of dauer larva formation, and for this regulation the population density information is transmitted by the dauer pheromone, which is continuously secreted by the animals (8). Mutants of daf-22, which encodes an enzyme in the pheromone biosynthesis pathway (9), showed a defect in olfactory plasticity (Fig. 1B and figs. S2 and S3). The chemical nature of dauer pheromones was recently described as a set of sugar derivatives called ascarosides (1013). Addition of either crude pheromone extract or individual synthetic pheromones restored olfactory plasticity to the daf-22 mutants (Fig. 1B and fig. S4). Long-term incubation with the pheromone (more than 24 hours) before the assay is needed to exert full effect of the pheromone (fig. S5). We therefore conclude that population density information is transmitted via the dauer pheromone ascarosides, which enhance olfactory plasticity.

To learn how olfactory plasticity is regulated, we screened for mutants defective in the plasticity of chemotaxis to bz (14, 15). Genetic fine-mapping and rescue experiments revealed that one of the mutants, pe356, had a mutation in a splicing acceptor of the nep-2 gene, which encodes a C. elegans homolog of the mammalian peptidase family neprilysin (Fig. 2, A to C, and figs. S2, S3, S6A, and S8). Neprilysin is a type II transmembrane protein with a zinc metalloprotease activity in the C-terminal ectodomain and is known to degrade extracellular peptides (16, 17). A deletion mutant of nep-2, pe379, also showed a severe defect in olfactory plasticity (Fig. 2, A and B). Metalloprotease activity is probably essential for regulation of olfactory plasticity because a mutant form of nep-2, nep-2(FExxF), in which zinc-binding amino acids at the active center were mutated (18), failed to rescue the mutant phenotype (fig. S7).

Fig. 2

The nep-2 neprilysin gene is essential for olfactory plasticity. (A) Olfactory plasticity of nep-2 mutants pe356 and pe379. (n ≥ 3 assays) (B) Gene model of nep-2 and lesions in nep-2 mutants. (C) Domain structure and motifs in the predicted NEP-2 protein. (D and E) Tissue- and cell-specific rescue experiments. nep-2 cDNA was expressed by tissue- or cell-specific promoters in nep-2(pe356) mutants, and the transgenic animals were subjected to olfactory plasticity assay [(D) n ≥ 4 assays and (E) n ≥ 5 assays]. Open bars and filled bars represent mock preexposed conditions and bz preexposed conditions, respectively. Dunnett’s test was used for multiple comparisons.

Observation of the fluorescent reporter Venus fused to the nep-2 promoter suggested that nep-2 is mainly expressed in muscles, glia-like cells (GLRs), and several classes of neurons (Fig. 3, B and C). Expression in each of these tissues is driven by separate cis-acting elements in the promoter region (Fig. 3, A and D to F). The translational fusion protein NEP-2::Venus was localized to the cell surface in both muscles and neurons (fig. S7, D to H). Tissue- and cell-specific rescue experiments show that expression of nep-2 in the nervous system is important for olfactory plasticity (Fig. 2D) and that nep-2 can be functional when expressed in a variety of neurons (Fig. 2E). These observations are consistent with NEP-2 acting as an ecto-peptidase that degrades paracrine or endocrine peptides so that accumulation of the peptide in extracellular space in the nep-2 mutant causes inhibition of olfactory plasticity.

Fig. 3

Expression patterns of nep-2 and snet-1. (A) Structure of the truncated nep-2 promoters, nep-2p, nep-2p(H), nep-2p(S), and nep-2p(HS), used for the expression of Venus reporter. (B to F) Expression pattern of Venus reporter driven by nep-2p, nep-2p(H), nep-2p(S), or nep-2p(HS). Bright-field image of (B) is shown as an inset. (C) is a magnified image of the depicted region of (B). (D) to (F) show the same head region as shown in (C). (G to I) Expression pattern of the Venus reporter driven by snet-1p. Bright-field image of (G) is shown as an inset. (H) shows a bright-field image of the head region observed in (I). (J and K) Expression pattern of Venus reporter driven by snet-1p in the daf-22 background with or without addition of 0.1 unit per milliliter of dauer pheromone to the culture plate. Question marks indicate an unidentified pair of lateral neurons.

To identify neuropeptides that are negatively regulated by NEP-2, we further screened for mutations that suppress the olfactory plasticity defect of nep-2(pe379). One of these mutations, pe1063, strongly suppressed the defect of nep-2 (Fig. 4A). Fine-mapping and rescue experiments showed that the causative gene was C02F12.3, which we named snet-1 (suppressor of nep-2) (Fig. 4B and figs. S2, S3, S6B, and S8). snet-1 is predicted to encode a small protein of 101 amino acids, with a signal sequence and dibasic cleavage sites typical of neuropeptide precursors (Fig. 4C). The predicted peptide region had partial similarity with Aplysia L11 peptide, which is a peptide secreted from the L11 neuron (fig. S9A) (19). Rescue experiments with truncated constructs suggested that the peptide region is important for the regulation of olfactory plasticity (fig. S9, B and C). When Venus was fused to the coding sequence, the fluorescence was observed in coelomocytes—the scavenger cells located in the body cavity—indicating that the protein is secreted by the action of the signal sequence (fig. S10, A, B, and G). Intracellular localization of SNET-1::Venus fusion protein and its dependence on UNC-104 kinesin-3 suggest that SNET-1 is transported to the neurites by kinesin-dependent anterograde transport and probably secreted there (fig. S10, C to F) (20). Olfactory plasticity defect of nep-2 reappeared when snet-1 was expressed by various cell-specific promoters in the nep-2;snet-1 double mutant (Fig. 4D). These observations are consistent with SNET-1 being a secreted peptide and a substrate of the NEP-2 peptidase. Because null mutation of snet-1 may not be fully epistatic to nep-2 (Fig. 4A), there may be also other targets of NEP-2.

Fig. 4

snet-1 encodes a peptide and regulates olfactory plasticity. (A) Olfactory plasticity of nep-2, snet-1, and nep-2;snet-1 mutants (n ≥ 5 assays). (B) Gene model of snet-1 and the pe1063 mutation. (C) Domain structure in the predicted SNET-1 protein. (D) snet-1 cDNA was expressed in the nep-2(pe379);snet-1(pe1063) double-mutant background by cell-specific promoters, and the transgenic animals were subjected to olfactory plasticity assay (n ≥ 5 assays). (E) Relative levels of snet-1p::venus fluorescence in adult daf-22 mutant animals with or without 0.1 unit per milliliter of crude pheromone in the culture plates (n = 80 animals for ASK and n = 40 animals for other neurons). (F) Olfactory plasticity of the daf-22(m130);snet-1(pe1063) double mutant (n ≥ 5 assays). (G) A possible role for NEP-2 and SNET-1 in pheromone-dependent regulation of olfactory plasticity. In (A), (D), and (F), open bars and filled bars represent mock preexposed conditions and bz preexposed conditions, respectively. [(A) and (F)] Tukey’s test, (D) Dunnet’s test, and (E) Student’s t test with Bonferroni correction were used for multiple comparisons.

The snet-1p::Venus promoter fusion reporter was expressed in several neurons, including the sensory neurons ASK and the interneurons AIB, AIM, and PVQ (Fig. 3, G to I), suggesting that SNET-1 may be regulated by sensory inputs, such as sensation of the pheromone. Because overexpression of snet-1 by increased gene dosage leads to defect in olfactory plasticity, quantitative regulation of the snet-1 gene expression may be important (fig. S11). Using quantitative reverse transcription polymerase chain reaction (RT-PCR), we indeed found that the transcript of snet-1 was increased in the adult animals of daf-22 pheromone–less mutants as compared with the wild type (fig. S12). In the daf-22 mutant background, snet-1p::Venus was expressed in ASI sensory neurons in addition to the cells observed in the wild-type background, and the supernumerary expression was suppressed by the addition of the pheromone to the culture plates (Figs. 3, J and K, and 4E). ASI neurons are known to regulate dauer larva formation, and expression of the genes daf-7 transforming growth factor–β (TGFβ), daf-28 insulin, and several chemoreceptors such as srd-1, str-2, and str-3 in ASI is regulated by the dauer pheromone (2123). Of particular interest is that expression of str-3 and daf-7 is regulated through the pheromone receptor SRBC-64/SRBC-66 on ASK sensory neurons in larvae (24), but mutants of these receptors did not show any defect in olfactory plasticity (fig. S13). These observations suggest that down-regulation of snet-1 expression by the pheromone, probably in an SRBC-64/66–independent manner, results in promotion of olfactory plasticity by reducing the SNET-1 peptide signaling. In support of the notion that SNET-1 conveys the pheromonal signal, null mutation of snet-1 suppressed the olfactory plasticity–defective phenotype of daf-22 to the wild-type level (Fig. 4F).

Throughout the animal kingdom, greater population density leads to increased rates of dispersal behavior because high population density usually leads to competition for limited resources within a habitat (57). Population density–dependent olfactory plasticity may serve as one of such adaptive mechanisms. Our results indicate that nep-2 regulates olfactory plasticity by acting in a cell-nonautonomous manner. Mammals possess seven neprilysin family proteins: neprilysin, endothelin-converting enzyme 1 (ECE1), ECE2, phosphate-regulating neutral endopeptidase (PHEX), neprilysin-2, damage-induced neuronal endopeptidase (DINE), and Kell (16). Although little is known about the loss-of-function phenotypes of the neprilysin family proteins, recent biochemical studies revealed that neprilysin scavenges amyloid β peptide efficiently and with this mechanism can prevent Alzheimer’s disease (17). On the basis of our observations, we propose a model in which SNET-1 peptide is secreted as an environmental signal that prevents olfactory plasticity and NEP-2 turns off the SNET-1 signals by degrading the excess peptides, creating a balance sensitive to environmental signals. By acting on an unknown receptor in the olfactory sensory circuit, SNET-1 negatively regulates olfactory plasticity (Fig. 4G and fig. S14). Given that daf-22, nep-2, and snet-1 regulate the plasticity of the response to several types of odorants (fig. S3), the pheromonal information may also regulate other behaviors through the peptide signaling. Our identification of the SNET-1 pathway that regulates olfactory plasticity sheds light on the complexity of the regulatory network underlying simple sensory behavior in C. elegans and suggests similar regulation might be present in other animals.

Supporting Online Material

Materials and Methods

Figs. S1 to S14


References and Notes

  1. Materials and methods are available as supporting material on Science Online.
  2. The authors acknowledge J. Ragains for synthesis of the dauer pheromone ascarosides. Y.I. and T.I. were supported by a Grant-in-aid for Scientific Research, J.C. by NIH grant CA24487, and R.A.B. by an NIH K99 Pathway to Independence Award (GM087533). We thank S. Mitani and P. Sengupta for srbc-64 and srbc-66 strains.
View Abstract

Navigate This Article