Selective breeding of entomopathogenic nematodes for enhanced attraction to a root signal did not reduce their establishment or persistence after field release

We recently showed that the efficacy of an entomopathogenic nematode (EPN) as a biological control agent against a root pest could be enhanced through artificial selection. The EPN Heterorhabditis bacteriophora was selected for higher responsiveness towards (E)-β-caryophyllene (EβC), a sesquiterpene that is emitted by maize roots in response to feeding damage by the western corn rootworm (WCR). EβC is normally only weakly attractive to H. bacteriophora, which is one of the most infectious nematodes against WCR. By selecting H. bacteriophora to move more readily along a EβC gradient we obtained a strain that was almost twice more efficient in controlling WCR population in fields planted with an EβC-producing maize variety. However, artificial selection for one trait may come at a cost for other important traits such as infectiousness, establishment and/or persistence in the field. Indeed, infectiousness was slightly but significantly reduced in the selected strain. Yet, this apparent cost was largely compensated for by the higher responsiveness to the root signal. Here we show that the selection process had no negative effect on establishment and persistence of field-released EPN. This knowledge, combined with the previously reported results, attest to the feasibility of manipulating key traits to improve the efficacy of beneficial organisms.Key words: entomopathogenic nematodes, tritrophic interactions, artificial selection, biological control, Diabrotica virgifera virgifera, western corn rootworm, persistence, establishmentDiabrotica virgifera virgifera LeConte (Chrysomelidae: Coleptera, western corn rootworm, WCR) is a major well established pest of maize in the American Corn Belt and more recently also in Europe.¹ The larval stages of this beetle can cause significant damages to maize roots, leading to reduction of plant growth, deficiencies in nutrient and water uptake, lodging, increased susceptibility to water stress and reduced grain yield.² This combination of factors result in an estimated loss of one billion US dollars per year in the USA.³ The pest has been introduced in Europe in the early ''90s,⁴ and it is expected that at full establishment the costs resulting from WCR damages will be half a billion Euros.⁵ Several strategies are available to control this soil-dwelling pest, including crop rotation, pesticides and transgenic Bt maize, but WCR can readily evolve resistance to each of these methods.^6–8 This is why efforts have been invested in biological control alternatives.Entomopathogenic nematodes (EPN) show great promise as biological agents against WCR.⁹ Root-produced volatiles appear to play an important role in the recruitment of EPN^10–13 and one such volatile, (E)-β-caryophyllene (EβC), has recently been identified for maize roots¹⁴ and was found to be an ideal below-ground alarm signal.¹⁵ EPN efficacy can be improved by exploiting the ability of WCR-damaged maize roots to emit the attractant.¹⁴ Further studies have shown the importance of choosing the right species of nematodes.¹⁶ Among the EPN species tested against WCR, Heterorhabditis bacteriophora has proven to be one of the most virulent nematodes,¹⁷ but it barely responds to EβC.¹⁶ We therefore recently selected H. bacteriophora for higher responsiveness to EβC.¹⁸ In the field, the selected strain exhibited better abilities to control WCR larvae, but logically only in maize plots with plants that emitted EβC. However, previous studies have shown that enhancing beneficial traits through selective breeding can incur costs and negatively alter other traits in the selected strain.¹⁹ For EPN such trade-offs after selective breeding have also been reported, for instance resulting in reduced storage stability²⁰ or a lower capacity to kill their hosts.²¹ After selection for enhanced responsiveness to EβC response, we observed a small, but significant negative effect on infectiousness of the selected strains. However, this drawback was readily outweighed by the improved ability to locate hosts in the field.¹⁸Not only infectiousness is a crucial trait for the successful use of EPN in biological control: establishment and persistence in the field are of decisive importance as well. These traits vary with EPN species and are determined by biotic factors such as pathogens and predators²² or abiotic factors such as soil type,²³ humidity,²⁴ temperature²⁵ or pH.²⁴ But the main factor that is thought to determine long-term persistence in the field is the presence of available host insects.²⁵ In field trials in Hungary, three EPN species, H. bacteriophora, H. megidis and Steinernema feltiae, were released to test their control potential against WCR. They all persisted at least as long WCR were present in soil, during the same year.²⁶ There was no significant difference between the three species in the establishment or persistence. Yet, independent of timing of application, EPN populations dramatically decreased within five months after application. The authors²⁶ propose that this short persistence is due to the absence of suitable alternative hosts in intensively cultivated crop fields in Europe.To determine if the selection for enhanced responsiveness to EβC went at a cost for establishment and persistence we compared these key traits for the original and the EβC-selected stains. Using a metal auger (2 cm diam.; 20 cm high), 310 soil samples were dug out either two days (establishment) or 28 days (persistence) after EPN application. The soil was placed in plastic boxes (4.5 cm diam.; 60 cm high) and as previously described²⁶ Tenebrio molitor (Coleoptera: Tenebrionidae) larva was placed as bait in the boxes. Presence/absence of EPN was evaluated by visually checking T. molitor larvae for EPN infection. Soil samples from areas where no EPN were applied served as controls. No significant differences were found between the original and selected strain of H. bacteriophora strain (factor “strain”), neither in establishment after two days nor in persistence after 28 days (factor “time”) (Fig. 1, two-way ANOVA, Ftime_1,35 = 2.937, p = 0.097; Fstrain_2,35 = 10.359, p < 0.001; Ftime × strain_2,35 = 1.202, p = 0.315, statistical differences within factors were calculated using a Bonferoni post-hoc test). Hence, the selection of H. bacteriophora for a better response to EβC had no consequence for how the nematodes settled in the experimental fields. Future efforts to improve the effectiveness H. bacteriophora against WCR might also include selection for increased persistence in soil. This would allow lower application rates and could provide growers with an affordable and efficient control strategy against this voracious pest.Open in a separate windowFigure 1Establishment and persistence of the original and a selected strain of H. bacteriophora. The selected strain (squares) established and persisted as well as the original strain (diamonds). The triangles represent control samples from plots where no nematodes were released. Establishment (after two days) and persistence (after 28 days) was equal for both strains. Moreover, the number of soil samples containing EPN after 28 days was not significantly lower than after 2 days, independently of treatment. A few nematodes were detected in the control samples but again no differences over time were detected. Error bars indicate the SEM. Different lower-case letters indicate statistical differences within establishment (after 2 days) or persistence (after 28 days) (p <0.05).So far, manipulation of tritrophic systems in order to improve biological control has been largely theoretical.^27–29 We show here that for EPN this approach is realistic and that their responsiveness to root-produced foraging signals can be enhanced without significant costs for other relevant traits. It has also been shown that the emissions of the signals by the plants can be enhanced.³⁰ Combining these strategies opens new perspectives for the development of ecologically sound strategies in pest management.