Introduction
To successfully infect a host, a parasite or pathogen must first encounter it (Combes Reference Combes, Toft, Aeschlimann and Bolis1991, Reference Combes2001). This is a major limiting step to successful transmission; thus, many parasites display strategies to situate themselves in the same spatiotemporal window as potentially compatible hosts (e.g., Reece et al. Reference Reece, Prior and Mideo2017). Parasites with a motile infectious stage found in the environment are particularly known to exhibit various behaviours to facilitate host encounter (Combes et al. Reference Combes, Bartoli, Théron, Lewis, Campbell and Sukhdeo2002). For instance, many free-living infectious stages show directional movement towards chemical cues associated with host presence (i.e., positive chemotaxis) or increased general activity levels (chemokinesis). This has been reported in a range of different parasites, from protozoa to digenean flatworms (e.g., Haberl et al. Reference Haberl, Kalbe, Fuchs, Ströbel, Schmalfuss and Haas1995; Haas et al. Reference Haas, Haberl, Hofmann, Kerschensteiner and Ketzer1999).
That said, chemically mediated host-finding behaviours have been most widely reported for parasitic species within the Phylum Nematoda. Parasites within this incredibly diverse phylum can infect plants, vertebrates, and invertebrates, with host-seeking behaviours reported for nematode infectious stages using any of these taxa (see reviews by Gang and Hallem Reference Gang and Hallem2016; Čepulytė and Bu̅da Reference Čepulytė and Bu̅da2022). Infectious juveniles/larvae can use multiple sensory inputs to locate and contact suitable hosts, including olfaction and gustation, as well as sensing heat, humidity, vibrations, and electric fields (e.g., Byers and Poinar Reference Byers and Poinar1982; Dillman et al. Reference Dillman, Guillermin, Lee, Kim, Sternberg and Hallem2012a; reviewed by Gang and Hallem Reference Gang and Hallem2016). Host-seeking behaviours are particularly well-studied for a particular subset of insect-infecting nematodes, entomopathogenic nematodes (EPNs), given their importance in the biocontrol of pest insects and use as a model system for studies with parasitic nematodes (Lacey et al. Reference Lacey, Grzywacz, Shapiro-Ilan, Frutos, Brownbridge and Goettel2015; Lewis et al. Reference Lewis, Campbell, Griffin, Kaya and Peters2006; Rumbos and Athanassiou Reference Rumbos and Athanassiou2017; Shapiro-Ilan et al. Reference Shapiro-Ilan, Han and Dolinksi2012).
EPNs are represented by dozens of species in two genera, Steinernema and Heterorhabditis (reviewed by Kaya and Gaugler Reference Kaya and Gaugler1993; Dillman and Sternberg Reference Dillman and Sternberg2012; Stock et al. Reference Stock, Campos-Herrera and Shapiro-Ilan2025). Unlike other insect-infecting nematodes, EPNs are obligate killers because they can only feed on dead hosts and carry a symbiotic bacterium that causes insect death while also preventing colonization by opportunistic bacteria and fungi (Dillman et al. Reference Dillman, Chaston, Adams, Ciche, Goodrich-Blair, Stock and Sternberg2012b). When IJs in the environment encounter a suitable insect host, they enter it through different means depending on the EPN species, such as through the spiracles, anus, or direct penetration of the cuticle (Lewis et al. Reference Lewis, Campbell, Griffin, Kaya and Peters2006). Bacteria-induced host death often occurs within 24–48 h, after which the colonizing third-stage infective juveniles (IJs) mature into adults, mate, and lay eggs, generally completing 2–3 generations within the insect cadaver before IJs emerge to seek new hosts. Host specificity varies among EPNs – some species show considerable variation in their ability to establish an infection within a given type of insect and cause host death (Kaya and Gaugler Reference Kaya and Gaugler1993; Simoes and Rosa Reference Simoes and Rosa1996).
EPNs respond to many chemical stimuli (see review by Zhang et al. Reference Zhang, Li, Kesner and Robert2021). As expected, this includes insect-generated cues in the form of excretory products, cuticle, or gut contents, especially in combination with carbon dioxide (Grewal et al. Reference Grewal, Gaugler and Lewis1993). Various studies have shown one or more behavioural responses by IJs when exposed to airborne cues by greater wax moth larvae (a.k.a. waxworms) and crickets (e.g., Campbell and Kaya Reference Campbell and Kaya2000; Dillman et al. Reference Dillman, Guillermin, Lee, Kim, Sternberg and Hallem2012a), or to chemicals from supernatants of crushed insects (e.g., Čepulytė et al. Reference Čepulytė, Tiškevičiūtė, Osinska and Būda2024). However, not all EPN species respond similarly to the same host cue, with stronger responses often reflecting greater host suitability (e.g., Lewis et al. Reference Lewis, Ricchi and Gaugler1996; Dillman et al. Reference Dillman, Guillermin, Lee, Kim, Sternberg and Hallem2012a), and the same cue can invoke different behaviours in the same species of EPN (Hallem et al. Reference Hallem, Dillman, Hong, Zhang, Yano, DeMarco and Sternberg2011). For instance, propanoic acid and hexanal were among five identified host odorants that stimulated jumping but not chemotaxis in IJs of S. carpocapsae, while heptanol and nonanol did the opposite (Hallem et al. Reference Hallem, Dillman, Hong, Zhang, Yano, DeMarco and Sternberg2011). EPNs can also respond to insect-derived cues within a soil matrix, including frass chemicals such as nitrogen metabolism and waste products like uric acid (e.g., Schmidt and All Reference Schmidt and All1979; reviewed by Zhang et al. Reference Zhang, Li, Kesner and Robert2021).
Interestingly, EPNs do not only respond to chemicals generated by uninfected insects. As an example, EPN-infected host cadavers release certain compounds (e.g., prenol and 1-nonene) that can either repel or attract IJs in a dose-dependent manner (Baiocchi et al. Reference Baiocchi, Lee, Choe and Dillman2017; Čepulytė et al. Reference Čepulytė, Tiškevičiūtė, Osinska and Būda2024, Reference Čepulytė, Osinska, Apšegaitė, Tiškevičiūtė and Būda2025). A surprising number of plant-generated chemicals have also been demonstrated to attract IJs (reviewed by Zhang et al. Reference Zhang, Li, Kesner and Robert2021), especially herbivore-induced volatiles, as this serves to recruit EPN natural enemies of the attacking insects (e.g., Bai et al. Reference Bai, Yu, Hu, Fu, Wu, Li and Zhang2024). A recent study even reported that both fungivorous insects (fruit fly larvae) and the EPN S. diaprepesi were attracted to a fungus-produced chemical, octenol (Wu and Duncan Reference Wu and Duncan2020). In addition to responding to odorants signifying potential insect hosts, i.e., in a foraging context, EPNs can also react to chemical cues indicating the presence of predators such as mites (Kamali et al. Reference Kamali, Olabiyi, Diepenbrock, Stelinski and Duncan2025).
When it comes to host-finding, it has been suggested that different species of EPN typically fall into one of two foraging categories depending on the relative motility of their insect hosts, with implications for their responses to host-associated cues: “cruisers” that use fairly stationary insects and “ambushers” which infect motile insects, with some EPNs considered to have intermediate strategies (Campbell et al. Reference Campbell, Lewis, Stock, Nadler and Kaya2003; Lewis et al. Reference Lewis, Campbell, Griffin, Kaya and Peters2006; but see Wilson et al. Reference Wilson, Ehlers and Glazer2012). Correspondingly, EPNs may respond to host cues in a hierarchical sequence depending on their foraging category (Lewis et al. Reference Lewis, Campbell, Griffin, Kaya and Peters2006). Because cruisers disperse within their environment through crawling and seek hosts with low mobility, they are attracted to olfactory cues during this phase so as to narrow down their location whereas ambush foragers do not show this attraction when crawling, given that they primarily wait for passing hosts rather than engaging in chemotaxis (Lewis et al. Reference Lewis, Gaugler and Harrison1993). After contact with a host or host-associated cues, cruising EPNs often switch to a more restricted search area to narrow down host location; however, EPNs with an ambush strategy may then show an attraction to olfactory cues that were absent pre-contact (Lewis et al.Reference Lewis, Gaugler and Harrison1992, Reference Lewis, Grewal and Gaugler1995).
All EPNs can engage in crawling behaviour during which they may examine their environment for host-associated cues but standing and jumping are unique behaviours seen in various species of Steinernema considered to be ambushers (Campbell and Kaya Reference Campbell and Kaya2002; Lewis et al. Reference Lewis, Campbell, Griffin, Kaya and Peters2006). While many EPNs can raise the anterior portion of their body to some extent and wave it around (i.e., head-waving), standing (a.k.a. nictation) involves a nematode elevating a substantial proportion of its body from the substrate and then balancing on a bend in its tail while engaging in head-waving (Campbell and Gaugler Reference Campbell and Gaugler1993; Campbell and Kaya Reference Campbell and Kaya1999a, Reference Campbell and Kayab, Reference Campbell and Kaya2000; Ishibashi and Kondo Reference Ishibashi, Kondo, Gaugler and Kaya1990). Standing behaviour may thus facilitate environmental scanning for cues indicating host presence, such as volatile chemicals. While in the standing position, an IJ may form a loop with its body such that it is then propelled through the air, i.e., jumping, when released (Campbell and Kaya Reference Campbell and Kaya1999b).
The primary function of jumping is to rapidly contact and attach to a mobile potential host as part of an overall ambushing strategy (Campbell and Kaya Reference Campbell and Kaya1999a), but this behaviour is only triggered by stimuli presented in a particular manner. Specifically, there are many reports of ambushing Steinernema species primarily responding to host cues when standing, but not when in a crawling position (e.g., Campbell and Kaya Reference Campbell and Kaya1999a, Reference Campbell and Kaya2000). By waving in response to host-associated cues while standing, IJs can quickly react by jumping, thereby increasing their likelihood of successful host attachment (Campbell and Kaya Reference Campbell and Kaya1999a, Reference Campbell and Kaya2000). That the sensitivity of ambushing IJs to olfactory cues is heightened after physical contact with the cuticle of a suitable insect host (e.g., Lewis et al. Reference Lewis, Grewal and Gaugler1995; Baiocchi et al. Reference Baiocchi, Braun and Dillman2019) may further ensure the success of nematodes with this foraging strategy by priming them to seek host entry points from which stimulating odorants are emanating (Lewis et al. Reference Lewis, Ricchi and Gaugler1996).
Here, we investigate the behavioural responses of S. carpocapsae IJs, typically considered as ambush foragers (Campbell et al. Reference Campbell, Lewis, Stock, Nadler and Kaya2003; Lewis et al. Reference Lewis, Campbell, Griffin, Kaya and Peters2006; but see Wilson et al. Reference Wilson, Ehlers and Glazer2012), after physical contact with aqueous cues indicating the presence of suitable or non-suitable hosts, including the alkene 1-pentadecene (1-P). This compound is produced by certain bacteria (e.g., Kumari Reference Kumari2019) and higher plants (e.g., Ney and Boland Reference Ney and Boland1987) but is also found in the secretions of tenebrionid beetles such as Tribolium spp. (e.g., Tschinkel Reference Tschinkel1975; Villarverde et al. Reference Villaverde, Juarez and Mijailovsky2007) and various weevils (e.g., Cai et al. Reference Cai, Macfadyen, Hua, Xu and Ren2022), along with other insects such as black flies (McGaha et al. Reference McGaha, Young, Burkett-Cadena, Iburg, Beau and Hassan2015) and ants (Coll et al. Reference Coll, Hefetz and Lloyd1987). In Tribolium spp., it is an information-carrying chemical (semiochemical) as it is an attractant at low concentrations but a repellent at high concentrations, thus acting to regulate beetle density by modulating sociality (Đukić et al. Reference Đukić, Andrić, Glinwood, Ninkovic, Andjelković and Radonjić2021), and has also been characterized as a sex pheromone (Keville and Kannowski Reference Keville and Kannowski1975). As we previously found that the main components of Tribolium secretions, methyl-1,4-benzoquinone and ethyl-1,4-benzoquinone, affected S. carpocapsae survival and activity (Smith et al. Reference Smith, Tay and Koprivnikar2023), we wished to further investigate the potential behavioural effects of 1-P. Notably, Rivera et al. (Reference Rivera, Martini, Khrimian and Stelinski2017) reported that a pheromone produced by male weevils was highly attractive to two species of EPN (Steinernema diaprepesi and Heterorhabditis indica). This suggests the potential for EPNs to use this type of chemical cue to recognize and locate suitable hosts (i.e., as a kairomone), warranting study with a different insect semiochemical.
Material and methods
Nematode and host cue sources
Third-stage IJs of S. carpocapsae were obtained from a commercial source (Koppert Biological Systems) and used within 2 months after refrigerated storage at 4°C as per the supplier’s instructions. We tested IJ behavioural responses to 1 of 5 treatments: 1) cues from a model insect host, 2) the semiochemical 1-P, 3) solvent control for 1-P, 4) cues from an incompatible host, and 5) deionized water (DW). For treatment 1, we used larvae (waxworms) of the greater wax moth (Galleria mellonella), a model insect frequently used for laboratory tests with EPNs (see Baiocchi and Dillman Reference Baiocchi and Dillman2015). Waxworms were purchased from a local pet store (PetSmart). Macerated waxworms in DW are a commonly used stimulus in EPN behavioural studies (e.g., Čepulytė et al. Reference Čepulytė, Tiškevičiūtė, Osinska and Būda2024). To generate a solution containing the cues of a compatible host, we froze 3 waxworms at –20°C before crushing them in 10 mL of DW. 1-pentadecene (98% purity) was purchased from Sigma-Aldrich (13360-61-7). To make the 1-P solution, we dissolved 1-P in 95% ethanol and then diluted this with DW to a concentration of 0.16 mg/mL (as per Smith et al. Reference Smith, Tay and Koprivnikar2023). We chose this concentration of 1-P because it reflects the average ratio of its contribution to the quinone-containing secretions produced by T. castaneum (75% ethy-1,4-benzoquinone, 15% methy-1,4-benzoquinone, and 10% 1-P) when considering the values reported by other studies (Loconti and Roth Reference Loconti and Roth1953; Villaverde et al. Reference Villaverde, Juarez and Mijailovsky2007). The solvent control (treatment 3) thus reflected the concentration of ethanol (4%) in DW used in treatment 2.
For an incompatible host cue (treatment 4), we used eastern mud snails (Ilyanassa obsoleta) that were field-collected in the Bay of Fundy and then kept frozen at –20°C. While there have been reports of Steinernema species responding to cues from terrestrial slugs (Dillman et al. Reference Dillman, Guillermin, Lee, Kim, Sternberg and Hallem2012a; Nermut et al. Reference Nermut, Půža and Mráček2012), we believed that S. carpocapsae would be unlikely to respond to cues from snails found on intertidal mudflats, given that they would not encounter these in natural settings. Like treatment 1), we thawed 3 snails and crushed them in 10 mL of DW to create a solution (similar to Nermut et al. Reference Nermut, Půža and Mráček2012 for slug cues). Finally, we had a neutral control consisting of DW (treatment 5). All solutions were prepared 1 h before the experiment commenced and kept at room temperature (~23°C). To ensure that the IJs were exposed to the concentrations for treatments 2 and 3 as indicated above after IJ transfer in DW, we prepared working solutions of 1-P and ethanol at 0.1776 mg/mL and 4.44%, respectively. After transferring individual IJs in 20 uL DW, the solutions in each well were thus diluted to the final concentrations of 0.16 mg/mL for 1-P (treatment 2) or 4% ethanol (treatment 3).
Nematode behavioural assay
The various behaviours of EPN IJs are well-described elsewhere (e.g., Campbell and Gaugler Reference Campbell and Gaugler1993; Campbell and Kaya Reference Campbell and Kaya2002; Robinson Reference Robinson, Chen, Chen and Dickson2004). We focused on 3 specific behavioural measures here: sinusoidal (side-to-side) movements, head-waving, and overall activity (any visible movement). These 3 behaviours reflected our focus on chemokinesis (non-directed movement in response to chemical stimuli) rather than chemotaxis (directed movement). Side-to-side sinusoidal movement was specifically meant to measure klinokinesis (angular velocity/rate of turning), in keeping with Benhamou and Bovet (Reference Benhamou and Bovet1989), whereas overall activity reflects the speed or frequency of movement, i.e., orthokinesis (reviewed by Mueller and Fagan Reference Mueller and Fagan2008). We note that while we used sinusoidal movement as a marker of kinesis in our study, given the lack of a substrate for the IJs to propel themselves forward (see below), this behaviour also inherently involves some element of taxis which we could not consider given our use of a liquid medium. We also observed IJs for ‘looping behaviour’ (Campbell and Kaya Reference Campbell and Kaya2000) but later excluded it as a measure because looping was only observed in 2 individuals.
We used 96-well tissue culture plates to conduct our behavioural assays. The IJs were allowed to acclimate in a Petri dish of DW at room temperature for 30 min before the experiment. Before their use in the behavioural assays described below, a subset of these IJs was observed for movement under a dissecting microscope to ensure their viability for experimentation. After this period, we individually transferred IJs contained within 20 μL of DW by pipetting them into single wells of the tissue culture plates. Because we had 16 replicates (individual nematodes in wells) for each of the 5 treatments, there were 80 IJs in total. Given the time needed to observe each IJ, we used 8 plates that each had 10 wells (2 from each of the 5 treatments), randomly assigned a treatment and IJ to stagger the experiment start times. Once nematode transfer was complete, we then added 180 μL of solution representing one of the 5 treatments described above. We conducted observations of individual IJs immediately after adding the treatment solutions to wells, and then again after 15 min, 30 min, 1 h, 4 h, and 8 h. Each IJ was observed for 30 sec under a dissecting microscope at each of the 6 time points. During this observation period, we noted the presence or absence of sinusoidal movement or head-waving behaviour, as well as the total number of movements by each IJ.
We choose to expose S. carpocapsae IJs within small volumes of solution rather than to airborne volatiles or filter papers to which solutions had been added (protocols described in Baiocchi and Dillman Reference Baiocchi and Dillman2015) for various reasons. First, when EPNs (including S. carpocapsae) are applied for pest insect biocontrol purposes, this often occurs in a gel or liquid medium (e.g., Schroer and Ehlers Reference Schroer and Ehlers2005), thus it is useful to understand how dissolved host cues affect EPN behaviours under these conditions. It is also important to note that Kruitbos et al. (Reference Kruitbos, Heritage, Hapca and Wilson2009) demonstrated a significant effect of substrate type on body waving by S. carpocapsae as this was more frequently seen for IJs observed on peat and leaf litter than on either sandy loam or sand. Similarly, Kruitbos et al. (Reference Kruitbos, Heritage, Hapca and Wilson2010) reported greater chemotaxis by S. carpocapsae while in a peat medium compared to sand and highlighted the difficulties of drawing conclusions regarding the behaviours of soil-transmitted parasites in nature depending on the experimental media used. By instead using liquid media, we aimed to eliminate the influence of substrate on IJ behaviours, although their ability to stand, wave, and jump is affected by the presence of materials to which they can adhere, as well as moisture level (Ishibashi and Kondo Reference Ishibashi, Kondo, Gaugler and Kaya1990; Kruitbos et al. Reference Kruitbos, Heritage, Hapca and Wilson2009). This said, the focus of our study was not on directed movement (taxis) or jumping, but rather on kinesis (non-directed movement). Our focal behaviours (sinusoidal movement, head-waving, and overall level of movement) were thus best measured by observing worms in a transparent medium, especially to conduct repeated observations of the same individuals over an 8-h period.
Our approach also broadly simulated how S. carpocapsae IJs may come into physical contact with host chemical cues dissolved into a soil matrix while foraging in natural settings. While it has been difficult to definitively establish the natural distribution of S. carpsocapsae IJs within the soil matrix, they likely occur within the organic thatch layer or upper 1 cm of soil and are potentially more motile than lab-based studies suggest (reviewed by Wilson et al. Reference Wilson, Ehlers and Glazer2012). As such, S. carpocapsae IJs could find themselves physically surrounded by insect-produced chemical cues within the soil or organic thatch. Given the numerous reports that EPNs with ambusher strategies, like S. carpocapsae, only respond to airborne volatile cues after physical contact with host-associated cues (Lewis et al. Reference Lewis, Campbell, Griffin, Kaya and Peters2006), we wanted to focus on the latter aspect.
Statistical analysis
We removed 2 IJs from our dataset that did not display any behaviours for the entire observation period (i.e., had a zero for each measure at each of the time points) as these most likely reflected dead or non-viable IJs, meaning that we obtained data for 78 IJs in total across the 5 treatments. We then used separate generalized linear mixed models (GLMM) to conduct repeated-measures analyses and examine the fixed effects of treatment (categorical variable) and time (continuous variable), as well as their interaction, for each of our 3 behavioural measures. As the total number of movements for each IJ tended to have a relatively high number of zeros at the 4- and 8-h time points, we analyzed this measure using a Poisson distribution with a log link function. Because sinusoidal movement and head-waving were noted as yes/no for each IJ at each time point, i.e., these were binary categorical responses, we used a binary distribution and logit link function for the GLMMs analyzing these behaviours. All GLMMs included Tukey LSD posthoc tests for treatment. Analyses were conducted using IBM SPSS version 30.0.
Results
Treatment had a significant effect on the total number of movements (orthokinesis) exhibited by S. carpocapsae IJs (F4,287 = 6.559, P <0.001), as did time (F1,218 = 64.71, P <0.001; Figure 1a), and there was also a significant interaction between treatment and time (F4,193 = 5.371, P <0.001). IJs exposed to the waxworm solution showed significantly more movements (mean ± S.E. = 18.39 ± 1.32) compared to those in all of the other treatments (P <0.05 for all pairwise comparisons) while IJs exposed to 1-P moved significantly more (mean ± S.E. = 12.96 ± 1.39) than those in the DW and ethanol treatments (P <0.05 for all pairwise comparisons), with a trend for the 1-P-exposed IJs to move more than those exposed to the snail cue (P = 0.074). IJs in the DW treatment showed the fewest number of movements overall (mean ± S.E. = 4.76 ± 1.13) and were significantly less active than IJs in the other 4 treatments (P <0.05 for all pairwise contrasts). The total number of IJ movements decreased over the 8-h observation period; however, the positive effect of 1-P exposure on orthokinesis was more evident over the first hour, whereas that of waxworm solution treatment remained relatively consistent.
Behavioural responses by infective juveniles (IJs) of the nematode Steinernema carpocapsae during exposure to one of five experiment treatments represented by deionized water, the insect semiochemical 1-pentadecene (1-P), a solvent control (EtOH), or macerated tissue from either the eastern mud snail Ilyanassa obsoleta or caterpillars of the greater wax moth Galleria mellonella (i.e., waxworms). Nematode behaviours include: a) mean (±S.E.) total number of movements by individual IJs, b) occurrence of head-waving by IJs, and c) occurrence of sinusoidal (side-to-side) movement by IJs.
The head-waving behaviour of IJs was also significantly affected by treatment (F4,342 = 7.769, P <0.001) and time (F1,174 = 14.233, P <0.001), but not by their interaction (F4,173 = 1.297, P = 0.273; Figure 1b) as IJs exposed to 1-P consistently exhibited more head-waving than nematodes in the other 4 treatments (P <0.05 for all pairwise contrasts). IJ sinusoidal movement (klinokinesis) was significantly affected by treatment (F4,328 = 6.983, P <0.001), time (F1,277 = 22.236, P <0.001), and an interaction between treatment and time (F4,216 = 4.333, P = 0.002; Figure 1c). With the exception of comparing to the DW treatment, IJs exposed to 1-P exhibited fewer sinusoidal movements compared to IJs in the other treatments (P <0.05 for all pairwise contrasts), whereas IJs in the waxworm treatment exhibited side-to-side movements more frequently than those in any of the other 4 treatments (P <0.05 for all pairwise contrasts). IJ sinusoidal movement decreased over time in all treatments; however, this reduction in side-to-side movement was much sharper for nematodes exposed to DW, while less pronounced over time for IJs in the waxworm treatment.
Discussion
We found that IJs of the entomopathogenic nematode S. carpocapsae exhibited strong behavioural responses when exposed to aqueous cues associated with insect hosts through increases in non-directed movement (kinesis). IJs exposed to a solution containing macerated tissue from a known host, caterpillars of the greater wax moth G. mellonella (i.e., waxworms), showed the greatest total movement and sinusoidal (side-to-side) movements. IJs exposed to the insect semiochemical 1-pentadecene (1-P) also exhibited high levels of activity and were additionally the most likely to engage in head-waving behaviour. Our study thus demonstrates that S. carpocapsae IJs show an increase in two components of kinesis (overall activity and sinusoidal movement, i.e., orthokinesis and klinokinesis, respectively) and another behaviour (head-waving) that is important for making host contact. This is also only the second report of EPNs showing a behavioural response to an insect semiochemical or pheromone, suggesting that this is a promising area for future study when it comes to understanding the host-finding strategies used by EPNs.
While various studies have reported that S. carpocapsae exhibits behavioural responses to volatiles that signal host presence in the form of jumping or limited chemotaxis (e.g., Campbell and Kaya Reference Campbell and Kaya2000; Baiocchi et al. Reference Baiocchi, Lee, Choe and Dillman2017), kinesis is also an important part of foraging strategy and has been reported from both parasitic and free-living nematodes (Riga Reference Riga, Gaugler and Bilgrami2004; Robinson Reference Robinson, Chen, Chen and Dickson2004). Directed movement (taxis) towards a chemical stimulus in an intensity-dependent manner is clearly advantageous for locating resources with a heterogenous environmental distribution; however, non-directed movement (kinesis) can be beneficial when patchy resources have a fairly coarse spatial distribution, especially if these resources show little temporal predictability (Mueller and Fagan Reference Mueller and Fagan2008).
Both ortho- and klinokinesis can generally aid nematodes in finding ideal conditions (Lee Reference Lee and Lee2002). Specifically, high levels of movement and turning rate can increase the likelihood of chance host encounter by IJs who are then able to switch to directed movement (taxis) and localize their search (Riga Reference Riga, Gaugler and Bilgrami2004). As such, both ortho- and klinokinesis contribute to area-restricted foraging, which is an energy-efficient approach. That S. carpocapsae IJs were significantly more active when exposed to macerated waxworm tissue and 1-P may indicate that they respond more strongly to host chemical cues via kinesis rather than taxis in keeping with their typical categorization as ambush foragers but future studies considering both types of behaviour will be necessary. For instance, Manduca et al. (Reference Manduca, Zeni, Casadei, Tarasco, Lucchi and Benelli2025) reported that S. carpocapsae IJs tended to increase their speed and turning rate after contact with insect faeces through lab-on-a-chip observations, which was proposed as IJ exploitation of variations in klinotaxis to actively search for a nearby living host after perceiving chemical signals from it.
When we consider the two specific behaviours (head-waving and sinusoidal movement) exhibited by S. carpocapsae in our study, IJs showed remarkably different responses to the host-associated chemical cues. Notably, although exposure to either the waxworm or 1-P solution increased klinokinesis (overall activity), side-to-side movements were most often performed by IJs in the waxworm treatment, whereas head-waving was by far most frequently seen for IJs in the 1-P treatment. This divergent pattern of 1-P and macerated insect solution effects may reflect that these cues convey different information about host proximity and attachment opportunity, thus triggering S. carpocapsae behaviour(s) appropriate to each situation.
Previous studies have reported increased taxis and head-thrusting by S. carpocapsae IJs after physical contact with cues (cuticle or gut contents) indicating the presence of an appropriate insect host (e.g., Grewal et al. Reference Grewal, Gaugler and Lewis1993; Lewis et al. Reference Lewis, Grewal and Gaugler1995; Baiocchi et al. Reference Baiocchi, Braun and Dillman2019). These particular behavioural responses are thought to prepare IJs for an immediate opportunity to successfully attach to a nearby host (Lewis et al. Reference Lewis, Campbell, Griffin, Kaya and Peters2006). Once inside their host, i.e., post-contact with internal cues, IJs release various excretory/secretory proteins, with effects ranging from metabolic breakdown of host tissue to immunomodulation (reviewed by Okakpu and Dillman Reference Okakpu and Dillman2022; Li and Wu Reference Li and Wu2024). For instance, when S. carpocapsae IJs were activated by in vitro exposure to insect tissue homogenate, they released proteins that were lethal or reduced host immune responses (e.g., Lu et al. Reference Lu, Macchietto, Chang, Barros, Baldwin, Mortazavi and Dillman2017; Jones et al. Reference Jones, Tafesh-Edwards, Kenney, Toubarro, Simões and Eleftherianos2022). That exposure to macerated waxworm tissue in our study resulted in an overall increase in movement is thus consistent with S. carpsocapsae IJs responding to cuticle or gut material but also suggests that this may be complemented by the stimulation of sinusoidal movement (orthokinesis), especially as side-to-side actions also facilitate taxis.
In contrast, 1-P does not contain host materials as it is a semiochemical secreted by tenebrionid beetles (e.g., Keville and Kannowski Reference Keville and Kannowski1975; Đukić et al. Reference Đukić, Andrić, Glinwood, Ninkovic, Andjelković and Radonjić2021), weevils (e.g., Cai et al. Reference Cai, Macfadyen, Hua, Xu and Ren2022), and other insects such as black flies (McGaha et al. Reference McGaha, Young, Burkett-Cadena, Iburg, Beau and Hassan2015) and ants (Coll et al. Reference Coll, Hefetz and Lloyd1987). It is thus likely that 1-P signals the potential presence of insect hosts in a way which stimulates IJs to search their immediate environment through a combination of overall movement (klinokinesis) and head-waving to detect other host cues. As we did not consider chemotaxis here, directed movement by S. carpocapsae and other EPNs in response to 1-P should be examined, especially in cruising species that should more actively track hosts (Lewis et al. Reference Lewis, Gaugler and Harrison1993). This is particularly important given that 1-P is also produced by other organisms within natural environments, such as certain bacteria and higher plants (e.g., Ney and Boland Reference Ney and Boland1987; Kumari Reference Kumari2019).
That all three focal behaviours showed a reduction over the 8-h observation period is to be expected given the limited energy reserves of IJs but the drop off was particularly steep for total number of movements by IJs exposed to 1-P and for sinusoidal movements by DW-exposed IJs. In contrast, the movements of waxworm-exposed IJs remained more consistent over time. This suggests that IJs may become less sensitive to some host-associated chemical cues over time if host encounter does not occur, as well as possible attenuation of those cues. As 1-P is a volatile compound that dissipates into the air (Tian et al. Reference Tian, Hao, Wu, Hu, Zhou, Liu and Zhang2022), this is likely why it primarily induced klinokinesis during the first hour, although the positive effect of 1-P on head-waving was more consistent through the 8-h observation period. By the 8-h point, IJs exposed to waxworm cues exhibited the most overall activity and head-waving, and the second-most sinusoidal movements, reflecting the strength of this host-generated signal. Our results also have relevance for future behavioural assays with IJs, given the interaction between time and cue type.
Our study is only the second to report a behavioural response by EPNs to an insect semiochemical or pheromone (i.e., as a kairomone), and the first to demonstrate an effect on S. carpocapsae. Rivera et al. (Reference Rivera, Martini, Khrimian and Stelinski2017) found increased taxis by S. diaprepesi, S. riobrave, and Heterorhabditis indica in response to a sex pheromone produced by male weevils (Diaprepes abbreviatus). In addition, nematodes in the genus Pristionchus, which are necromenic parasites that enter an insect host but wait for it to die in order to feed on the microbiota rather than causing death like EPNs, are attracted to certain insect sex pheromones, including those produced by various beetles (e.g., Hong et al. Reference Hong and Sommer2006, Reference Hong, Svatoš, Herrmann and Sommer2008; Herrmann et al. Reference Herrmann, Mayer, Hong, Kienle, Minasaki and Sommer2007). Considering the remarkable range of chemical cues to which EPNs show behavioural responses (see review by Zhang et al. Reference Zhang, Li, Kesner and Robert2021), future studies with other insect semiochemicals and pheromones may be highly informative. Insects produce a wide range of such compounds for different purposes, including mate attraction, aggregation or separation, navigation, and to warn conspecifics of danger (reviewed by Yew and Chung Reference Yew and Chung2015), many of which can be artificially synthesized (Souza et al. Reference Souza, Bandeira, Bergmann and Zarbin2023). This presents many opportunities for study.
Our observations of greater S. carpocapsae IJ behavioural responses if exposed to insect host cues (tissue and semiochemical) are generally consistent with the findings of previous studies for these compounds, but also for exposure to ethanol and another alkene, 1-nonene. Čepulytė et al. (Reference Čepulytė, Tiškevičiūtė, Osinska and Būda2024) found that S. carpocapsae was neither attracted nor repelled by ethanol in terms of taxis, and ethanol exposure did not affect IJ head-waving or kinesis in our study. However, high 1-nonene concentrations were attractive (Čepulytė et al. Reference Čepulytė, Tiškevičiūtė, Osinska and Būda2024). We also did not see IJ behavioural responses resulting from exposure to the macerated tissue of a gastropod, the eastern mud snail Ilyanassa obsoleta. Other studies have found that cues from terrestrial slugs induced behavioural responses by EPNs (e.g., Dillman et al. Reference Dillman, Guillermin, Lee, Kim, Sternberg and Hallem2012a; Nermut et al. Reference Nermut, Půža and Mráček2012) despite the inability of the tested EPNs to establish infections in the slugs, leading the authors to suggest that this might instead indicate a mechanism of dispersal through phoresy or EPN ability to scavenge on remains of non-insect cadavers. However, as the eastern mud snail is found on intertidal mudflats, it is improbable that EPNs would respond to its tissues for any reason, which likely explains the lack of behavioural responses by S. carpocapsae IJs seen here.
Understanding how EPN IJs respond to host-associated cues is important as they are a model system for studying how parasitic nematodes find suitable hosts and successfully establish infections (Riga Reference Riga, Gaugler and Bilgrami2004; Lewis et al. Reference Lewis, Campbell, Griffin, Kaya and Peters2006; Gang and Hallem Reference Gang and Hallem2016), and also because EPNs are a valuable tool in the biocontrol of pest insects (Shapiro-Ilan et al. Reference Shapiro-Ilan, Han and Dolinksi2012; Lacey et al. Reference Lacey, Grzywacz, Shapiro-Ilan, Frutos, Brownbridge and Goettel2015; Rumbos & Athanassiou Reference Rumbos and Athanassiou2017). Greater knowledge of how EPNs respond to host cues in different media is consequently valuable as this can help to optimize application strategies, as well as possibly prime IJs to quickly engage in host-finding behaviours and increase their likelihood of successful establishment. Future studies investigating how insect secretions such as semiochemicals and pheromones affect various EPN behaviours will thus be valuable.
Financial support
This work was supported by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada to JK (RGPIN-2020-04622) and NSERC USRA to AT.
Competing interests
The authors declare none.
Ethical standard
This study did not involve experimentation with humans or other animals regulated by the Canadian Council of Animal Care or institutional ethics committees.
