Influence of symbiotic and non-symbiotic bacteria on pheromone production in Steinernema nematodes (Nematoda, Steinernematidae) Free

Ascarosides are small signaling molecules secreted by nematodes that function as pheromones. Several studies have shown that these molecules influence various aspects of nematode behavior and development (Jeong et al., 2005; Butcher et al., 2007; Srinivasan et al., 2008, 2012). A significant amount of research has been done in the model organism Caenorhabditis elegans, which emits ascarosides to induce the formation of the dauer (non-feeding, environmentally resistant) stage (Golden and Riddle, 1982; Jeong et al., 2005; Srinivasan et al., 2008; Butcher et al., 2007, 2008), and also plays a role in social behavior including mate attraction (Srinivasan et al., 2008, 2012). Several studies have also demonstrated that ascaroside signaling is dependent on bacterial food supply. Golden and Riddle (1984) demonstrated that in C. elegans, food supply serves as a signal that acts antagonistically to the dauer pheromone by enhancing the recovery of nematodes from the dauer to their last larval stage.

Ascarosides are highly conserved across the phylum Nematoda as a common signaling system; however, each species has a unique ascaroside profile (Choe et al., 2012). Only recently has the research expanded to entomopathogenic nematodes (EPNs). For example, Kaplan et al. (2012) showed that asc-C5 is commonly found in both Steinernema and Heterorhabditis species. Additionally, asc-C5 was shown to play a role in the dispersal of Steinernema feltiae infective juveniles (IJs) (Kaplan et al., 2012). Noguez et al. (2012) found that ascaroside asc-C11-EA inhibits recovery of Heterorhabditis IJs into the fourth juvenile stage.

Entomopathogenic Steinernema nematodes have an obligate mutualistic relationship with Xenorhabdus bacteria (Boemare et al., 1993; Boemare, 2002). Xenorhabdus symbionts facilitate many processes of the nematodes' life cycle, including acting as a food source for the developing nematodes once inside an insect cadaver, and allowing the nematodes to recover from their IJ stage and grow and reproduce. Several studies have demonstrated that fitness of Steinernema nematodes is impacted by the symbiotic partners (Martens et al., 2003; Sicard et al., 2004; Murfin et al., 2015; McMullen et al., 2017). Specifically, it has been shown that non-cognate Xenorhabdus bacteria pairings resulted in total progeny production and virulence towards the nematodes' insect host (McMullen et al., 2017). Additionally, Steinernema IJs carry fewer bacterial cells when colonized by a non-cognate symbiont. However, until now, no research had been done to elucidate the role that Xenorhabdus symbionts may play in the production of ascarosides in adult stages of Steinernema species. Therefore, in this study we assessed the effect of cognate and non-cognate bacteria on the production of ascarosides in first-generation adults of Steinernema nematodes.

Two nematode species were considered in this study: Steinernema carpocapsae (Weiser) All (Sierra Biologics) strain and Steinernema feltiae (Filipjev) SN strain. Nematodes were reared in vivo utilizing last-instar larvae of the greater wax moth Galleria mellonella (Lepidoptera: Pyralidae) according to procedures described by Kaya and Stock (1997). Briefly, a nematode suspension of 1.8 ml (approximately 1000–2000 IJs ml⁻¹) was pipetted onto a 10 cm Petri dish lined with two discs of filter paper (Whatman no. 1). Ten G. mellonella larvae were added to the plate, which was covered and stored at 25°C in the dark. Upon death of the larvae, the cadavers were placed in a modified White trap according to procedures outlined by Kaya and Stock (1997). Once IJs emerged into the White trap, they were collected, rinsed with sterile distilled water and stored in tissue culture flasks at a concentration of 1000–2000 IJs ml⁻¹. Flasks were stored at 15±1°C in the dark until needed.

Four Xenorhabdus species – X. nematophila [symbiont of S. carpocapsae All strain (XnAll)], X. nematophila [symbiont of S. anatoliense (XnAna)], X. bovienii [symbiont of S. feltiae SN strain (XbfSN)] and X. bovienii [symbiont of S. puntauvense (Xbp)] – were considered in the experiments. Additionally, one non-symbiotic bacterium, Serratia proteamaculans was used in these experiments (Table 1).

Bacterial cultures were grown up from glycerol stocks in 15 ml centrifuge tubes containing 5 ml of Luria Bertani (LB) broth supplemented with 0.1% (w/v) pyruvate overnight for 20–24 h with agitation at 28±1°C (Stock and Goodrich-Blair, 2012). Bacterial cultures were standardized to an average optical density of approximately 0.8 according to procedures outlined by McMullen et al. (2017). An aliquot of 100 µl of the standardized bacterial culture was pipetted onto 5 cm lipid agar plates supplemented with 0.1% sodium pyruvate (w/v) and spread into a lawn. The culture was also streaked onto nutrient agar supplemented with 0.0025% (w/v) Bromothymol Blue, 0.004% (w/v) triphenyltetrazolium and 0.1% (w/v) sodium pyruvate (NBTA) to confirm phenotypic identity and verify the catalase profile (catalase-negative). Three plates per bacterial species were prepared for each experiment. Plates were stored upside down at 28±1°C for 2 days. Experiments were repeated twice.

Nematode and bacterial rearing conditions as well as the experimental setup followed procedures described by Roder and Stock (2018) with the exception of the size of the Petri dishes and the amount of bacterial culture plated.

Aposymbiotic (symbiont-free) nematodes were reared in vivo as described above. For S. feltiae, G. mellonella cadavers were dissected 3 days post-infection, whereas for S. carpocapsae, G. mellonella cadavers were dissected 3.5–4 days post-infection. Approximately 300 gravid (egg-bearing) females were collected with the assistance of an L-shaped needle and placed into a watch glass containing 10–15 ml of M9 buffer. Females were rinsed and subsequently ground in an axenizing NaOCl solution that facilitated dissolution of the females' bodies, allowing egg release (McMullen and Stock, 2014). Egg pellets were thoroughly rinsed with sterile distilled water and concentrated by centrifugation at 1600 rpm for 3 min (McMullen and Stock, 2014).

An inoculum of 100 µl of Steinernema egg suspension was pipetted onto either Petri dishes (10 cm diameter) with the tested bacteria or Petri dishes containing solely lipid agar as the negative control. There were two plates per nematode–bacterium combination per experimental setup. The egg aliquot contained an average of 1000–2000 eggs per plate for both S. feltiae and S. carpocapsae. Dishes were stored at 25±1°C for 18–24 h to allow nematode development from eggs. Plates were then removed from the incubator and maintained at room temperature (23±1°C) to allow the juveniles development. The experiment was repeated at least 3 times.

The time frame for collection of adults varied between 4 and 6 days, depending on the bacterial treatment. For ascaroside quantification, adult nematodes were harvested under sterile conditions in a laminar flow hood. Briefly, nematodes were rinsed off the plates with Ringer's solution and collected in a 15 ml centrifuge tube (BD Falcon^®). Plates from the same bacterial treatment were combined into one pool.

Nematodes were allowed to settle to the bottom of the tube and the supernatant was removed. Nematodes were rinsed 2–3 times until all debris was removed. An additional final rinse was done with 5–6 ml of Ringer's solution.

The nematode solution was homogenized by pipetting the suspension up and down. Three to six 100 µl aliquots were removed from the pool and dispensed onto a hemocytometer to quantify the total number of adults. The counts were utilized to determine the total number of adults in the sample. Triton X-100 was utilized as a surfactant to prevent nematodes from sticking to the pipette tips. Aliquots of 1000 S. carpocapsae adults were dispensed into new 15 ml centrifuge tubes (BD Falcon^®). For S. feltiae, because of the lower adult yield, aliquots of 300 adults collected from the XbfSN and Xbp conditions were collected and dispensed into new 15 ml centrifuge tubes. Because of the extremely low adult yields (100 or fewer) obtained when S. feltiae was reared with XnAll, the total amount of adults from this condition was pooled from multiple dishes. The supernatant of each sample was removed and the adults were suspended in 1 ml of Ringer's solution to soak.

The centrifuge tubes were incubated for 3 h while shaking at 180 rpm at 28±1°C. The supernatant from the samples was collected and placed into 1.5 ml microcentrifuge tubes. The samples were centrifuged for 5 min at 2000 rpm to allow any eggs that females may have released to pellet in the bottom of the tube. The supernatant was transferred to a 2.0 ml microcentrifuge tube and stored at −80°C until lyophilization.

Samples were centrifuged at 13,000 rpm for 5 min and the supernatant was transferred to a new 1.5 ml microcentrifuge tube and dried in a SpeedVac for 1 h. Following drying, 50 µl of methanol was added to extract ascarosides via sonication and vortexing. The microcentrifuge tubes were centrifuged again at 13,000 rpm for 5 min, and 10 µl of the supernatant was used for liquid chromatography mass spectroscopy (LC-MS).

LC-MS analysis was performed on a Phenomenex Luna 5 µm C18 100A (100×4.6 mm) column attached to an Agilent 1260 infinity binary pump and an Agilent 6130 single quad mass spectrometer with API-ES source, operating in dual negative/positive selected-ion monitoring (SIM) mode. A water (with 0.1% formic acid) and acetonitrile (with 0.1% formic acid) solvent gradient was used, holding at 5% acetonitrile for 5 min, ramping to 60% acetonitrile over 15 min, ramping to 100% acetonitrile, and holding at 100% acetonitrile for 4 min. Ascarosides were detected by LC-MS using the [M-H] ion. Quantification of ascarosides by LC-MS was done by generating a calibration curve utilizing synthetic standard ascarosides. Synthetic ascarosides (asc-C5, asc-C6, asc-C7, asc-C8, asc-C9, asc-C11, asc-C6-MK, IC-asc-C5) were used as standards for screening in LC-MS analysis. For comparative purposes, measurements for ascaroside concentration were standardized to 1000 adults for all treatments in both Steinernema species tested. All LC-MS analyses were performed in R.A.B.’s lab at the University of Florida. In this study, ascaroside molecules are reported based on their structure nomenclature.

Statistical analysis of ascarosides was performed in JMP software, version 13 (SAS). A mixed-effects ANOVA model (α=0.05) was considered for analysis. In the mixed-effects model, experimental setup was defined as a random effect and bacterial condition was defined as a fixed effect to account for any possible variability in setup. This was carried out through the fit model function, utilizing a restricted maximum likelihood (REML) method. A post hoc least squares means Tukey's honest significant difference (HSD) analysis was conducted to determine bacterial effects. Each ascaroside molecule was analyzed separately.

The tested nematode species were unable to successfully mature to adult stages when reared with non-symbiotic Serratia proteomaculans bacteria. Thus, the results provided below focus on data obtained from the three other bacterial culture conditions that considered different Xenorhabdus species and/or strains.

Four ascaroside molecules were quantifiably detected in S. carpocapsae first-generation adults: asc-C5, asc-C6, asc-C7 and asc-C11. Overall, asc-C5 was the most abundantly produced molecule, ranging in concentration from 10.09 pmol per 1000 adults when nematodes were reared with their native symbiont (XnAll) to 12.60 pmol per 1000 adults when they were cultivated with a non-cognate strain of X. nematophila (XnAna) (Fig. 1). Similarly, when the nematodes were reared with the non-cognate species X. bovienii (XbfSN), the amount of asc-C5 was higher (11.19 pmol per 1000 adults) than when the nematodes were reared with their cognate symbiont species (Fig. 1). However, none of the observed increments was significantly different. Contrastingly, production of asc-C6, asc-C7 and asc-11 was significantly lower than that observed for asc-C5, with concentrations below 1 pmol per 1000 adults for all bacterial strains tested. Notably, S. carpocapsae adults produced levels of asc-C11 ranging from minimally quantifiable when exposed to their native cognate symbiont (XnAll) to undetectable when reared on the non-cognate Xenorhabdus strain or species (Fig. 1).

Three ascaroside molecules were quantifiably obtained in S. feltiae adults: asc-C5, asc-C7 and asc-C11. There were no detectable amounts of asc-C6 in this species. Similar to findings for S. carpocapsae, asc-C5 was the most abundantly produced ascaroside molecule, ranging from 112.56 to 52.72 pmol per 1000 adults. Additionally, a slightly higher amount of this pheromone was detected when the nematodes were reared with the non-cognate Xbp strain, although this was not significantly different from amount produced by nematodes reared on their cognate strain (Fig. 2). Contrastingly, a significant reduction in asc-C5 was observed when the nematodes were reared with the non-native species XnAll (52.72 pmol per 1000 adults) (Fig. 2). asc-C7 was detected in small amounts (ranging from 5.34 to 8.09 pmol per 1000 adults) regardless of the rearing condition. Similar to asc-C5, a slightly higher amount of asc-C7 was detected when the nematodes were reared with the non-cognate strain (Xbp). asc-C11 was also produce in low amounts, ranging from 3.18 pmol per 1000 adults when nematodes were reared with the native symbiont XbfSN to 2.25 pmol per 1000 adults when cultivated with the non-cognate Xbp. asc-C11 was not detected when S. feltiae nematodes were reared with the non-cognate species XnAll.

Ascaroside molecules are known to take part in a variety of sex-specific and social behaviors including male attraction and aggregation in C. elegans (Srinivasan et al., 2008, 2012). In this study, we examined the ascaroside profiles of first-generation adults in two Steinernema spp. when reared in the presence of cognate and non-cognate Xenorhabdus species and/or strains and a non-symbiotic species, S. proteomaculans. Both Steinernema species considered in this study were unable to mature and/or reproduced with a non-symbiotic bacterium; thus, our results focus on their ascaroside production in the presence of cognate and non-cognate Xenorhabdus species and/or strains.

We showed quantitative and qualitative variation in ascaroside profiles in the tested Steinernema spp. For example, a larger amount of ascarosides was measured in S. feltiae than in S. carpocapsae. However, the latter species exhibited more diversity in the secreted pheromones. Specifically, while four ascaroside molecules (asc-C5, asc-C6, asc-C7, asc-C11) were identified in S. carpocapsae, only three (asc-C5, asc-C7 and asc-C11) were detected in S. feltiae.

In a previous study, Choe et al. (2012) showed that ascaroside profiles are unique to different nematode species. For example, the authors reported that asc-C6 was present in adults of S. carpocapsae as well as in adults and IJs of S. glaseri but was not present in IJs of either S. carpocapsae or S. riobrave. In this study, we demonstrated that asc-C6 was present in S. carpocapsae adults but was absent in S. feltiae adults. These findings suggest that production of this pheromone is both stage and species specific; however the precise function or role of this ascaroside molecule remains to be elucidated.

Bacteria, as a food source, have been shown to influence ascaroside production in C. elegans (Golden and Riddle, 1984; Kaplan et al., 2011). In relation to this, the present study showed that Xenorhabdus differentially affected each of the tested Steinernema spp. with respect to ascaroside production. For example, in S. carpocapsae, both non-cognate strain XnAna and non-cognate species XbfSN enhanced the production of asc-C5 and asc-C7. Similarly, in S. feltiae, the non-cognate Xenorhabdus strain Xbp favored production of asc-C5. Although these findings are interesting, from a statistical stand point they were not significant.

In S. carpocapsae, the non-cognate species XbfSN had no effect in the amount of ascaroside produced. However, in S. feltiae, the non-cognate species XnAll significantly impacted the amount of asc-5. This pheromone has been demonstrated to stimulate dispersal of Steinernema IJs (Kaplan et al., 2012); however, its role in the behavior of adult stages is still undetermined.

Choe et al. (2012) showed that adults of S. carpocapsae and IJs of S. riobrave produce asc-C11 in small quantities. In this study, we detected traces of asc-C11 in S. carpocapsae adults when they were reared with their cognate symbiont. However, this molecule was undetected when the nematodes were cultured with a non-cognate strain or species of Xenorhabdus. In S. feltiae, a larger amount of this pheromone was measured when the nematodes were cultivated with both native and non-native Xenorhabdus strains. asc-C11 was not quantified when the nematodes were cultured with a non-cognate Xenorhabdus sp. (XnAll).

Noguez et al. (2012) demonstrated that a chemically modified version of this ascaroside, ascaroside C11 ethanolamine (asc-C11-EA), induces IJ formation in the entomopathogenic nematode Heterorhabditis bacteriophora. Manosalva et al. (2015) also showed that asc-C11 is a major ascaroside secreted by plant-parasitic nematodes and that it plays a role in inducing plant defenses. However, the role of this pheromone molecule in entomopathogenic nematodes is not yet understood.

In summary, this study reports for the first time the production of ascaroside molecules in first-generation adult stages of Steinernema nematodes and expands on current knowledge on the diversity of ascaroside molecules produced by these nematodes, in the presence of cognate and non-cognate bacteria. Specifically, we showed that slightly larger amounts of secreted ascaroside molecules were observed when both Steinernema species were reared with non-cognate Xenorhabdus strains. Contrastingly, lower pheromone amounts were detected when the nematodes were reared in non-cognate Xenorhabdus species. Overall, Xenorhabdus had no effect on the quantity and quality of ascarosides produced in S. carpocapsae. However, in S. feltiae, Xenorhabdus had an effect in the recorded amounts of two pheromone molecules: asc-C5 and asc-C11. Understanding the interaction of ascaroside production and food supply will make a significant contribution not only to the unravelling of the life cycle of these entomopathogens but also to the optimization of culturing methods for mass production systems.

We would like to acknowledge Taylor Barrios, Aidan Foster, Grant Osborn and Dalaena Rivera for assisting with experiments and collection of data. We are also grateful to Fatma Kaplan (Pheronym) for providing training to A.C.R. on nematode ascaroside extraction.