Plasticity of pheromone-mediated avoidance behavior in C. elegans

Abstract

Caenorhabditis elegans secretes a complex cocktail of small chemicals collectively called ascaroside pheromones which serves as a chemical language for intra-species communication. Subsets of ascarosides have been shown to mediate a broad spectrum of C. elegans behavior and development, such as gender-specific attraction, repulsion, aggregation, olfactory plasticity, and dauer formation. Recent studies show that specific components of ascarosides elicit a rapid avoidance response that allows animals to avoid predators and escape from unfavorable conditions. Moreover, this avoidance behavior is modulated by external conditions, internal states, and previous experience, indicating that pheromone avoidance behavior is highly plastic. In this review, we describe molecular and circuit mechanisms underlying plasticity in pheromone avoidance behavior which pave a way to better understanding circuit mechanisms underlying behavioral plasticity in higher animals, including humans.

Keywords:

• Pheromone
• avoidance behavior
• plasticity
• C. elegans

Introduction

Animals consistently modulate their behavior in response to the ever-changing environmental conditions and internal states. Thus, a major goal in neuroscience is to understand how animals sense environmental signals and internal metabolic status and how neuronal circuits process and translate this information into the appropriate behavioral programs. The correct sensation, integration, and processing of the external and internal cues is critical for the maintenance of physiology and behavior, and dysregulation of these pathways leads to multiple psychiatric and/or other neurological disorders. However, it is tremendously challenging to identify the underlying neuronal circuits and molecular and cellular mechanisms in higher animals. The nervous system of these animals is complex, and it is difficult to correlate specific behavioral changes with disruption of function of particular neurons and molecules.

The nematode C. elegans provides an excellent model system in which to explore the neural and molecular basis of environmentally influenced behavior. C. elegans has a relatively simple nervous system with only 302 neurons and over 7,000 synapses in hermaphrodites and 385 neurons and over 8,000 synapses in males, and their wiring diagrams have been completely reconstructed (Bhattacharya, Aghayeva, Berghoff, & Hobert, 2019; Cook et al., 2019; Hall & Russell, 1991; Jarrell et al., 2012; White, Southgate, Thomson, & Brenner, 1986). However, C. elegans displays a broad spectrum of behaviors, such as locomotion, chemosensation, nociception, foraging, and feeding. In addition, C. elegans exhibits more complex behaviors, including social and sleep-like behaviors as well as learning and memory [See review (Ardiel & Rankin, 2010; Byrne et al., 2019; Rengarajan & Hallem, 2016)]. Moreover, most, if not all, of these behaviors are plastic and can be modulated by external and/or internal conditions [See review (Ardiel & Rankin, 2008; Garcia & Portman, 2016; Hobert, 2003; Sasakura & Mori, 2013; Sengupta, 2013)]. Here, we review recent findings regarding particular chemosensory behaviors, avoidance or attraction elicited by pheromone cues, as well as discuss their modulation by circuit state, sex, previous experience and the feeding state in C. elegans.

Ascaroside pheromones

Pheromones are blends of released chemicals that play major roles in intraspecies chemical communication (Karlson & Luscher, 1959). C. elegans secretes a complex cocktail of small chemicals that are collectively called ascaroside pheromones; these affect many aspects of C. elegans biology [See review (Butcher, 2019; Edison, 2009; Ludewig & Schroeder, 2013; McGrath & Ruvinsky, 2019; J. Park, Choi, Dar, Butcher, & Kim, 2019; Schroeder, 2015)]. Since C. elegans ascaroside pheromones were discovered as a dauer-inducing metabolite in 1982 (Golden & Riddle, 1982), the chemical components of ascaroside pheromones have been identified as hundreds of structurally related compounds (Artyukhin et al., 2013; Butcher, Fujita, Schroeder, & Clardy, 2007; Butcher, Ragains, Kim, & Clardy, 2008; Butcher et al., 2009; Jeong et al., 2005; Pungaliya et al., 2009; Srinivasan et al., 2008; Srinivasan et al., 2012).

As a dauer-inducing cue, ascaroside pheromones act as a population density indicator to regulate growth in the early larval developmental stage: high pheromone concentrations cause animals to enter into an alternative non-aging, stress-resistant, and developmentally arrested dauer larval stage (Butcher et al., 2007; Butcher et al., 2009; Golden & Riddle, 1982, 1984; Jeong et al., 2005; Pungaliya et al., 2009; Srinivasan et al., 2008) (Table 1). Subsequently, over a dozen research groups have shown that ascaroside pheromones play additional roles in several behaviors, including gender-specific attraction and repulsion, aggregation, and foraging (Artyukhin et al., 2013; Chute et al., 2019; Fagan et al., 2018; Greene, Dobosiewicz, Butcher, McGrath, & Bargmann, 2016; Jang et al., 2012; Macosko et al., 2009; Narayan et al., 2016; D. Park et al., 2017; Pungaliya et al., 2009; Srinivasan et al., 2008; Srinivasan et al., 2012; von Reuss et al., 2012; Zhang, Sanchez-Ayala, Sternberg, Srinivasan, & Schroeder, 2017) (Table 1). Moreover, complexity of ascaroside signaling further modulates these biological processes. Thus, C. elegans ascaroside pheromones elicit long-term changes in development and physiology as well as short-term behavioral changes, making it an outstanding model organism to explore the mechanisms by which biologically relevant cues are sensed and integrated to direct critically important developmental and behavioral programs.

Table 1 Structure and function of scaroside pheromones in C. elegans.

Ascr#3 avoidance behavior and its plasticity

A few pheromones that elicit chemosensory responses have been well-characterized in C. elegans. Srinivasan et al. showed that adult wild-type males were attractive to a blend of femto- or pico-molar concentrations of three ascarosides, namely ascr#2 (asc-C6-MK, C6), ascr#3(asc-ΔC9, C9), and ascr#4, whereas adult wild-type hermaphrodites avoided higher concentrations of ascr#2 or ascr#3 (Srinivasan et al., 2008). This group also showed that in males, ascr#3 appeared to be sensed by two chemosensory neuron-types, the ASK amphid neurons and the male-specific CEM neurons (Srinivasan et al., 2008). Macosko et al. further showed that hermaphrodites from social feeding strains with low npr-1 neuropeptide receptor gene acitivity (215-valine) or npr-1 loss-of-function mutant (ad609lf) strains exhibited attraction behavior to a 10 nM mixture of ascr#2, ascr#3, and ascr#5 (asc ωC3, C3) but not to any one of these pheromone components individually (Macosko et al., 2009). However, wild-type solitary feeding hermaphrodites with high npr-1 activity (215-phenylalanine) avoided the ascaroside mixture (Macosko et al., 2009). The ASK pheromone-sensing neurons elicited acute Ca²⁺ transient upon exposure to this pheromone mixture: pheromone-evoked Ca²⁺ response was much stronger in npr-1(lf) mutant animals than in wild-type animals (Macosko et al., 2009). This enhanced Ca²⁺ response in the ASK neurons upon pheromone exposure may contribute to attraction behavior toward the pheromone in npr-1(lf) mutant animals. These two studies indicate that pheromone responses in C. elegans vary depending on sex and a neuropeptide circuit that controls solitary or social feeding behavior.

Jang et al. further dissected molecular and circuit mechanisms by which worms modulate pheromone-mediated chemosensory behaviors depending on circuit state and sex (Jang et al., 2012). Using the pheromone drop test that detects acute avoidance responses upon pheromone exposure (Hilliard, Bargmann, & Bazzicalupo, 2002; Jang & Bargmann, 2013), the group showed that adult wild-type hermaphrodites specifically avoided nano-molar concentrations of ascr#3 but not of ascr#2 or ascr#5 (Jang et al., 2012). Calcium imaging, genetic analysis, behavioral analysis, and tools to manipulate neuronal activity supported that the ADL amphid chemosensory neurons sense ascr#3 and mediate ascr#3 avoidance in adult wild-type hermaphrodites (Jang et al., 2012). Subsequently, the ADL ascr#3 Ca²⁺ activities were significantly decreased in npr-1(lf) hermaphrodites and wild-type males, indicating that reduced ADL ascr#3 responses are required for ascr#3 attraction behavior in npr-1(lf) hermaphrodites and wild-type males (Jang et al., 2012). Strikingly, the ADL neurons drove ascr#3 avoidance in wild-type hermaphrodites via ADL chemical synapses likely onto the AVA and AVD command interneurons for backward locomotion and drove attraction through gap junctions via the RMG neurons in npr-1(lf) hermaphrodites (Jang et al., 2012). In males, the ADL neurons exhibited decreased ascr#3 responses which were independent of npr-1 activities and were possibly due to differences in their own sexual state and/or inputs from other sexually dimorphic neurons (Cook et al., 2019; Jang et al., 2012). These results suggest that the ADL neurons are capable of contributing to either attraction to or avoidance from ascr#3 and that their sensitivity to ascr#3 is modified in worms based on the function of the npr-1 gene and animal’s sex (Figure 1(A,B)). Furthermore, since the npr-1 genotype seems to reflect a stress-related behavioral state (de Bono, Tobin, Davis, Avery, & Bargmann, 2002; Rogers, Persson, Cheung, & de Bono, 2006), ascr#3 avoidance behavior could be modulated by stressful conditions. More recently, Fagan et al showed that the ADF neurons of males sensed and promoted attraction to ascr#3, while ADF-ablated hermaphrodites were repelled to ascr#3. In this sexual dimorphic process, the sexual regulator tra-1 and its direct target mab-3 cell-autonomously regulated the male state of ADF to promote ascr#3 attraction and the hermaphrodite state of ADF to permit ascr#3 repulsion (Fagan et al., 2018).

Figure 1. Models for behavioral plasticity of ascr#3 avoidance. (A) In npr-1(lf) hermaphrodite, ASK and RMG antagonize ADL chemical synapses and decrease ascr#3 avoidance. (B) In wild-type male, ADL ascr#3 response is decreased due to sexual dimorphism. ASK and RMG circuit antagonizes ADL output to further reduce ascr#3 avoidance. (C) In starved conditions, secretion of INS-18 from the intestine is decreased, which activates the DAF-2 signaling of ADL, resulting in increase in synaptic release from ADL to downstream neurons, and promotes enhanced ascr#3 avoidance. (D) In pre-exposed condition, ADL-SMB synaptic activities that are inactive in naïve animals is altered to promote enhanced ascr#3 avoidance.

The ascr#3 avoidance behavior is further modulated by the feeding state. Ryu et al. found that starved animals showed increased ascr#3 avoidance behavior compared to well-fed animals and that this effect was dependent on the daf-2 insulin-like receptor gene (Ryu et al., 2018). Although daf-2 mutant animals exhibited normal ADL Ca²⁺ activities upon ascr#3 exposure, they exhibited reduced synaptic transmission from the ADL ascr#3 sensing neurons to first-order AVA command interneurons (Ryu et al., 2018). The group further showed that daf-2 acted through the canonical IGF pathway in the ADL neurons to downregulate the snb-1 presynaptic protein (Ryu et al., 2018). Remarkably, an intestinal insulin-like peptide, INS-18, acted upstream of DAF-2 in an inhibitory manner and modulated ascr#3 avoidance, and the INS-18 release was reduced in starved worms (Ryu et al., 2018). This study significantly advances our understanding of how the feeding state can fundamentally change the behavioral output via hormonal signaling (Figure 1(C)).

Hong et al. investigated whether ascr#3 avoidance behavior is modulated by previous experience. The group demonstrated that in hermaphrodites, exposure to the ascr#3 pheromone during the critical period of the L1 larval stage led to enhanced ascr#3-specific avoidance behavior as an adult (Hong et al., 2017). Interestingly, this increased ascr#3 avoidance was not attributed to increase in the synaptic transmission via ADL–AVA synapses but to the recruitment of the SMB motor neurons into the circuit activated by the ascr#3-sensing ADL neurons (Hong et al., 2017). The SMB motor neurons, which are one of over twenty post-synaptic neurons (Cook et al., 2019; White et al., 1986), exhibited Ca²⁺ activities upon ascr#3 exposure only in ascr#3 pre-exposed animals (Hong et al., 2017). These results indicated that ascr#3-experienced animals form a long-lasting memory or imprint for ascr#3 experience (Figure 1(D)). Thus, this work provides fundamental circuit mechanisms underlying how individual synapses are functionally or anatomically altered upon sensory imprinting.

Avoidance behavior to other ascarosides

In addition to ascr#3, osas#9 ascaroside was characterized as a single pheromone component which causes avoidance behavior in wild-type hermaphrodites (Artyukhin et al., 2013; Chute et al., 2019). osas#9 is an interesting ascaroside since its side chain is derived from the neurotransmitter octopamine and it is produced mainly by starved young L1 larvae (Artyukhin et al., 2013). Chute et al. found that starved but not well-fed, young adult hermaphrodites avoided micro-molar concentrations of osas#9 (Chute et al., 2019). This osas#9 avoidance was mediated by not the ADL neurons but the nociceptive ASH neurons via the osas#9-sensing tyra-2 tyramine/octopamine G protein-coupled receptor (GPCR) (Chute et al., 2019). Interestingly, the group found that trya-2 expression was increased in starved worms, suggesting that changes in the expression level of the osas#9-chemoreceptor gene depending on the feeding state are responsible for starvation-dependent osas#9 avoidance. The ways in which the feeding state regulates the expression of the trya-2 gene remain to be explored. Moreover, since the osas#9-sensing ASH neurons also participate in the RMG gap junction circuit (Cook et al., 2019; White et al., 1986), synergistic avoidance with ascr#3 and osas#9 needs to be examined in starved animals, which will provide further molecular and circuit mechanisms underlying starvation-dependent ascaroside avoidance behavior.

Park et al. found that micromolar concentrations of ascr#1 (daumone, C7) elicited avoidance behavior in young adult hermaphrodites (D. Park et al., 2017). The group also suggested that the ascr#1 avoidance behavior is mediated by the daf-16/FoxO gene and glutamatergic transmission via the AWB chemosensory neurons but not either by ADL or ASH neurons (D. Park et al., 2017). However, roles of the AWB neurons in ascr#1 avoidance need to be further characterized for better understanding the ascr#1 avoidance behavior.

Avoidance behavior to other pheromones

Worms appear to avoid other putative pheromone components. Zhou et al. found that the internal fluid from injured worms elicited repulsive behavior in adult hermaphrodites via the ASI and ASK chemosensory neurons (Zhou et al., 2017). The group also found that this internal fluid did not affect the lifespan of worms, suggesting that this fluid acts as not a harmful repellent chemical but as an alarm pheromone (Zhou et al., 2017). This potential pheromone was none of the known ascarosides but comprised at least three nonvolatile components (Zhou et al., 2017). Notably, this avoidance was decreased in starved animals, indicating that this behavior is also dependent on the feeding state (Zhou et al., 2017). The chemical identity of these pheromones needs to be examined to further unravel the molecular and circuit mechanisms underlying this distinct pheromone avoidance behavior.

Pristionchus pacificus is a necromantic insect-dwelling nematode and a facultative predator of C. elegans (Serobyan, Ragsdale, & Sommer, 2014). It was reported that P. pacificus secreted a few ascarosides, indicating that C. elegans avoids this predator nematode (Choe et al., 2012). Interestingly, Liu et al., found that starved, but not well-fed, P. pacificus secreted additional non-volatile chemicals which elicited a strong avoidance behavior in C. elegans (Liu et al., 2018). They identified these chemical signals as a mixture of sulfolipids and found them to be similar to a known C. elegans repellent, sodium dodecyl sulfate (SDS) (Liu et al., 2018). These sulfolipids appeared to be detected redundantly by multiple chemosensory neurons, including the ADL, ASH, ASI, and ASJ neurons (Liu et al., 2018). Taken together, these results suggest that C. elegans ensures rapid and life-saving avoidance behavior against the predator by detecting predator-secreted ascaroside as a pheromone as well as sulfolipid as a kairomone (Brown, Eisner, & Whittaker, 1970; Stowe, Turlings, Loughrin, Lewis, & Tumlinson, 1995).

Conclusions and future outlook

Connectomics, the description of comprehensive maps of anatomical connections between every neuron in an organism, is a rapidly evolving field in neuroscience. These large-scale wiring diagrams are expected to offer a unique opportunity to understand not only the structural details of neural connectivity but also neural functions. However, previous studies demonstrated that the wiring diagrams are insufficient to decipher circuit and neural mechanisms of behavior. Studies on the plasticity of C. elegans pheromone avoidance further support that behavior cannot be predicted by anatomy alone and eventually requires delicate physiological and behavioral analysis at high resolutions. Studies on C. elegans’ ascr#3 pheromone avoidance behavior and its functional circuits have provided plentiful information on anatomical and functional logics of the circuit mechanisms including the ASK–RMG–ADL “hub and spoke” and “push and pull” circuit models which are conserved in higher organisms (Bargmann & Marder, 2013). It is intriguing to examine whether ascr#3 avoidance behavior is further modulated by temperature, mating experience, maternal age, and other internal or external conditions (Parida, Neogi, & Padmanabhan, 2014; Perez, Francesconi, Hidalgo-Carcedo, & Lehner, 2017; Shi & Murphy, 2014), that could lead to findings of unexpected but crucial circuit mechanisms. Moreover, it will help us more deeply understand how the nervous system is functionally organized to generate behavioral outcomes.

Understanding how individual synapses are functionally or anatomically altered in response to changes in environmental conditions and internal states is the essential first step toward being able to dissect out the molecular and neuronal mechanisms underlying many forms of behavioral plasticity. Given that the structures and functions of neural circuits are evolutionarily conserved, these works will lead to the development of a general framework for understanding how circuits are modulated in higher animals, including humans, and in disease states (Aitman et al., 2011; Aryal & Lee, 2019; Wangler et al., 2017). Moreover, these works will not only pave the way to identify molecular mechanisms underlying behavioral plasticity but also provide a useful platform to develop tools to modulate behaviors.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This study was supported by the National Research Foundation of Korea [NRF-2020R1A4A1019436, NRF-2018R1A2A3074987].