Description

Plants synthesize many structurally diverse organic molecules to interact with the organisms around them. These compounds can act at the receptor level in target organisms to aid the plant in defense or to facilitate a symbiotic relationship (Bahar et al., 2007; Wink, 2018). We have not exhausted these plant-derived secondary metabolites as a source of novel drug therapeutics for pressing human health challenges because screening them for potential function as neuronal actuators is difficult. Here, we explore (1) using C. elegans as an efficient and practical organism for screening plant compounds for potential bioactivity in the nervous system, and (2) using these methods in an undergraduate classroom to bring real scientific discoveries to teaching laboratories. C. elegans are well studied, have essential genetic tools, short generation times, and easy culture techniques (Corsi et al., 2015; O’Reilly et al., 2014). Moreover, C. elegans navigates its environment by relying on chemosensation, triggering chemotaxis in response to odorants (Bargmann, 2006; Margie et al., 2013). These characteristics make it a potentially effective organism for screening phytochemicals and teaching in an undergraduate laboratory setting.

We first screened five commercially available plant compounds for potential actuators of neuronal function using C. elegans chemotaxis behavior (Figure 1A), including (S)-(+)-carvone, eugenol, cis-jasmone, linalool, and (–)-ɑ-thujone (Figure 1B). The chemotaxis response in the N2 wild type strain differed across plant compounds and solvent control (ANOVA, F(5)=10.52, p=7.59×10^-5). In particular, (S)-(+)-carvone elicited a strong repulsion effect compared to solvent (p=0.0004) and all other compounds ((S)-(+)-carvone versus: eugenol p=0.005, cis-jasmone p=0.003, linalool p=0.008, (–)-ɑ-thujone p=0.00009). No other compounds significantly differed between each other or solvent control at the concentration that was tested (20 mM), which is surprising as these compounds have been previously shown to produce interesting neuromodulatory and behavioral effects in other animals: cis-jasmone repels aphids (Birkett et al., 2000), (–)-ɑ-thujone is a GABA_A receptor antagonist (Höld et al., 2000), and both linalool (Batista et al., 2008) and eugenol (Chung et al., 2014) alleviate glutamate-induced nociceptive pain. It may be that dose-response curves or different genetic strains would be required to fully explore the effects of these compounds in nematodes. Regardless, to our knowledge, this is the first demonstration of carvone as a repellent in C. elegans and we conducted follow-up experiments to gain insight into potential mechanisms.

We next tested whether there is genetic variation in chemotaxis response elicited by plant compounds by comparing chemotaxis behavior across divergent C. elegans strains (Figure 1C). Such variation between strains could allow for genome wide association studies to identify potential neuronal substrates (Cook et al., 2017). We tested (S)-(+)-carvone as well as (–)-ɑ-thujone due to its known interactions with GABA_A receptors (Olsen, 2000). Across four wild isolate strains and the N2 wild type strain, there was a significant effect of compound, but not strain, on chemotaxis behavior (2-way ANOVA, Compound: F(2)=4.679, p=0.0136; Strain: F(4)=1.821, p=0.1392). (S)-(+)-carvone elicited a negative chemotaxis response compared to both solvent (p=0.0409) and (–)-ɑ-thujone (p=0.0201). The chemotaxis behavior elicited by (–)-ɑ-thujone did not significantly differ from solvent (p=0.9444), mirroring results from the first screening experiment. These findings indicate that screening the number of strains and replicates required for genome wide association mapping is outside the scope of a single small undergraduate laboratory course, but leaves open the possibility that screening additional wild isolates could be valuable.

To explore the potential neuronal mechanisms for the repellant effects of carvone, we screened worm strains with mutations in chemosensory signaling pathways (Figure 1D). We used tax-4 mutants, which lack the alpha subunit of a cyclic nucleotide-gated ion channel required for chemosensation (Komatsu et al., 1996), and osm-9 mutants, that lack a transient receptor potential (TRP) channel subunit that is also needed for chemosensation (Bargmann, 2006; Colbert et al., 1997). The tax-4 and osm-9 channels are expressed in non-overlapping subsets of sensory neurons. In addition to using mutant strains, we also examined both stereoisomers of carvone, (R)-(–)-carvone and (S)-(+)-carvone, which are recognized as different odors by humans and thus may bind different olfactory receptors (Leitereg et al., 1971). Both worm strain, compound, and their interaction had a significant effect on chemotaxis behavior (2-way ANOVA, Strain: F(2)=11.085, p=0.0002; Compound: F(2)=64.929, p=3.88×10^-13; Strain * Compound: F(4)=7.476, p=0.0001). The two stereoisomers of carvone were more repellant than solvent control (solvent vs. (R)-(–)-carvone, p<0.0001; solvent vs. (S)-(+)-carvone, p=4.0×10^-6) and (R)-(–)-carvone seemed to elicit stronger responses than (S)-(+)-carvone (p=9.2×10^-6). Chemotaxis behavior in the tax-4 strain was significantly different than both osm-9 (p=0.0006) and N2 wild type strains (p=0.0006). We did not detect any differences in responses to carvone between osm-9 and N2 (p=0.9998) in this assay. Given the variance in these assays, however, more replicates will be needed to fully elucidate the role of osm-9 in carvone chemosensation and the apparent difference in sensitivity to carvone stereoisomers.

In summary, we have shown here that the terpenoid carvone, derived from caraway, dill and spearmint, has a repellent chemotaxis effect in C. elegans. Our results support other studies showing carvone is an effective insect repellent (Franzios et al., 1997; Tripathi et al., 2003). The repulsion effect of the carvone enantiomers was reduced in tax-4 mutants, suggesting that both enantiomers are sensed by tax-4-expressing sensory neurons. Overall, this work shows that carvone is a C. elegans repellant and also establishes the feasibility of screening for bioactive plant compounds in a classroom undergraduate research experience (CURE) course. This approach allows students to pursue genuine scientific discoveries within a few classroom sessions.