Evaluation of Entomopathogenic Nematodes against Common Wireworm Species in Potato Cultivation

Wireworms (Coleoptera: Elateridae) are common insect pests that attack a wide range of economically important crops including potatoes. The control of wireworms is of prime importance in potato production due to the potential damage of the larvae to tuber quantity and quality. Chemical insecticides, the main control strategy against wireworms, generally fail to provide satisfactory control due to the lack of available chemicals and the soil-dwelling habits of the larvae. In the last decades, new eco-friendly concepts have emerged in the sustainable control of wireworms, one of which is entomopathogenic nematodes (EPNs). EPNs are soil-inhabitant organisms and represent an ecological approach to controlling a great variety of soil-dwelling insect pests. In this study, the susceptibility of Agriotes sputator Linnaeus and A. rufipalpis Brullé larvae, the most common wireworm species in potato cultivation in Türkiye, to native EPN strains [Steinernema carpocapsae (Sc_BL22), S. feltiae (Sf_BL24 and Sf_KAY4), and Heterorhabditis bacteriophora (Hb_KAY10 and Hb_AF12)] were evaluated at two temperatures (25 and 30 °C) in pot experiments. Heterorhabditis bacteriophora Hb_AF12 was the most effective strain at 30 °C six days post-inoculation and caused 37.5% mortality to A. rufipalpis larvae. Agriotes sputator larvae were more susceptible to tested EPNs at the same exposure time, and 50% mortality was achieved by two EPNs species, Hb_AF12 and Sc_BL22. All EPN species/strains induced mortality over 70% to both wireworm species at both temperatures at 100 IJs/cm2, 18 days post-treatment. The results suggest that tested EPN species/strains have great potential in the control of A. sputator and A. rufipalpis larvae.


Introduction
Wireworms, Agriotes spp. (Coleoptera: Elateridae), are one of the most destructive soil-dwelling insect pests that can cause severe economic losses in many crops including potato [1,2]. Wireworms generally live under soil during their larval development and feed on underground parts of plants such as seeds, roots, and tubers, leading to severe yield and tuber quality losses [3]. In addition to direct damage to tubers, the feeding holes of wireworm larvae on tubers predispose plants to subsequent secondary infections by other invertebrates and microbial pathogens [4]. Wireworm damage leads frequently to a drop in market value due to the visual quality of potatoes. In the absence of appropriate control

Source of Wireworms
Agriotes rufipalpis and A. sputator larvae were assembled from potato fields in different parts of Türkiye and identified based on the mitochondrial cytochrome c oxidase subunit I (COI) sequences in a previous study [31]. The collected larvae were brought individually to the laboratory in plastic containers (50 mL) containing autoclaved sandy loamy soil (20 g) with a slice of potato (Ø2 cm) and kept at 25 • C. The 4th and 5th larvae were separated according to their head width and observed for one week for any sign of infection [9,34,35]. Only healthy larvae were included in the pathogenicity bioassays. The pathogenicity of EPNs was tested against the mixed groups of 4th and 5th larval instars.

Evaluation of the Pathogenicity of EPNs in Pot Experiment
The effectiveness of EPNs was evaluated at two temperatures (25 and 30 • C) in pots (1.5 L) (Surface area 240 cm 2 ) including 1 kg autoclaved (121 • C for 60 min) sandy loamy soil and a slice of fresh potato (Ø2 cm). The soil used in our experiment was obtained from Bolu Abantİzzet Baysal University (Department of Seed Science and Technology) and consisted of 81% sand, 14% silt, and 5% clay. The organic matter content of the soil was 2.1% with a pH of 6.5. Prior to the inoculation of IJs, each pot was irrigated with 100 mL of distilled water to provide an adequate amount of moisture for IJs. One larva of 4th or 5th instar was placed into pots and allowed to move deeper in the soil profile. The IJs suspended in 5 mL distilled water were applied uniformly to the soil surface at the concentration of 25, 50, 100, and 150 IJs/cm 2 with the help of an automatic pipette (corresponding to 6000, 12,000, 24,000, and 36,000 IJs per pot, respectively). Then, the pots were maintained at 25 • C and 30 • C, 65% RH. Since moisture is one of the most important factors affecting the movement of IJs, the pots were irrigated with 50 mL of distilled water daily during the experiment. The larval mortality was checked at 6 and 18 days after treatment. Dead larvae that were transferred to white traps were observed under a stereomicroscope to confirm the nematode infection. There were four replicates of each treatment with ten larvae per replicate. All experiments were repeated twice on different dates.

Statistical Analysis
No mortality occurred in the control groups. Data pooled from two experiments were analyzed using IBM SPSS Statistics, Version 20.0 for Windows (SPSS Inc., Chicago, IL, USA). Prior to analyses, a normality test was performed and data were subjected to arcsine transformation. To determine significant differences among treatments, full-factorial model repeated-measures ANOVA was applied. The effects of the main factors (Nematode, Temperature, and IJs concen-trations) and their interactions were considered significant at α = 0.05. Post-hoc comparisons were performed using Tukey's multiple comparison test (p ≤ 0.05).

Results
The results revealed that the larvae of A. rufipalpis and A. sputator showed varying degrees of susceptibility to all tested EPN species and strains ( Figure 1). All main factors had a significant effect on the mortality rates of both wireworm species (Table 2). Increasing concentrations of IJs and exposure time generally led to higher mortalities in the A. rufipalpis larvae, and mortality rates ranged between 2.5 and 37.5% at both temperatures tested. The susceptibility of A. rufipalpis larvae to EPNs tended to increase at 30 • C, and three EPN species/strains induced mortality over 30% at 6 days after treatment (DAT). Heterorhabditis bacterophora AF12 was the most efficient strain at 6 DAT and yielded 37.5% mortality at the highest concentration (150 IJs/cm 2 ) ( Table 3).

Statistical Analysis
No mortality occurred in the control groups. Data pooled from two experiments were analyzed using IBM SPSS Statistics, Version 20.0 for Windows (SPSS Inc., Chicago, IL, USA). Prior to analyses, a normality test was performed and data were subjected to arcsine transformation. To determine significant differences among treatments, full-factorial model repeated-measures ANOVA was applied. The effects of the main factors (Nematode, Temperature, and IJs concentrations) and their interactions were considered significant at α = 0.05. Post-hoc comparisons were performed using Tukey's multiple comparison test (p ≤ 0.05).

Results
The results revealed that the larvae of A. rufipalpis and A. sputator showed varying degrees of susceptibility to all tested EPN species and strains ( Figure 1). All main factors had a significant effect on the mortality rates of both wireworm species (Table 2). Increasing concentrations of IJs and exposure time generally led to higher mortalities in the A. rufipalpis larvae, and mortality rates ranged between 2.5 and 37.5% at both temperatures tested. The susceptibility of A. rufipalpis larvae to EPNs tended to increase at 30 °C, and three EPN species/strains induced mortality over 30% at 6 days after treatment (DAT). Heterorhabditis bacterophora AF12 was the most efficient strain at 6 DAT and yielded 37.5% mortality at the highest concentration (150 IJs/cm 2 ) ( Table 3).      At the lowest concentration, only two H. bacteriophora isolates (Hb_KAY10 and Hb_AF12) were able to cause mortalities of over 40% at 18 DAT. A remarkable increase occurred in the mortality rates of A. rufipalpis larvae at 100 and 150 IJs/cm 2 concentrations at 30 • C, and all EPN species/strains induced mortality of over 80% (Table 4). All EPN species/strains were able to kill the larvae of A. sputator at tested IJ concentrations and temperatures except for S. feltiae BL24 and H. bacteriophora KAY10 which induced no mortality at 25 IJs/cm 2 at 25 • C. However, with increasing IJ concentrations, H. bacteriophora KAY10 yielded the highest mortality (47.5%) at 100 and 150 Ijs/cm 2 at 25 • C and 6 DAT followed by H. bacteriophora AF12 (45.0%) and S. feltiae KAY4 (42.5%). Although S. feltiae BL22 was the least efficient isolate at 25 • C, the efficiency of BL22 significantly increased at 30 • C and achieved the highest mortality (50.0%) at 6 DAT along with H. bacteriophora KAY10 (Table 5).

Discussion
In the present study, native EPN species were tested for biocontrol potential against the most abundantly collected wireworm species from potato fields which is a major concern for potato growers in Europe including Türkiye [31][32][33][34][35][36]. As compared to other host species of EPNs, wireworms generally exhibited lower susceptibility to nematode infection, and mortality generally occurred over a longer period of time [37]. For instance, Williams et al. [38] tested the efficacy of S. feltiae, S. carpocapsae, H. bacteriophora, and H. indica on the larvae of Melanotus communis (G.) at 100 Ijs/cm 2 concentration and the highest larval mortality did not exceed 15%. In another study, Forgia et al. [39] conducted a laboratory bioassay against Agriotes sordidus (Illiger) larvae in well plates at 2 Ijs/cm 2 concentration and reported 8.3 and 16.7% of mortality for S. carpocapsae and H. bacteriophora strains, respectively. Campos-Herrera and Gutiérrez [40] evaluated the pathogenicity of different strains of S. feltiae and S. carpocapsae in well plates at 250 Ijs/cm 2 and reported 9% mortality in larvae of A. sordidus 12 DAT which was induced by only one S. feltiae strain. In contrast to aforementioned studies, Toba et al. [41] reported 58% mortality in the 7-10th instars larvae of Limonius californicus (M.) when S. feltiae was applied at a concentration of 393 Ijs/cm 2 . In another study, Morton and Garcia-del-Pino [42] conducted a laboratory bioassay with different S. feltiae, S. carpocapsae, and H. bacteriophora strains against the 5th/6th larvae of A. obscurus, and mortality rates ranged between 17% and 35% for H. bacteriophora strains while the highest mortality (75%) was obtained with S. carpocapsae B14 strain at a concentration of 100 IJs/cm 2 . In the present study, all tested EPN species/strains were highly pathogenic to A. rufipalpis and A. sputator larvae at 18 DAT, and mortality over 70% was achieved by all EPN species/strains. The high virulence of tested EPNs in this study is in line with Ansari et al. [34], Morton and Garcia-del-Pino [42], and Sandhi et al. [43] who reported mortality between 50 and 75% in wireworm species. It is a well-known fact that the pathogenicity of EPNs on the same host species differs greatly among species and even strains [44,45]. Therefore, preliminary pathogenicity screening tests provide valuable information before evaluation of performance of EPNs [46]. In the present study, the most pathogenic EPN species that proved to be highly virulent on the larvae of G. mellonella in an earlier study were used [30]. In the aforementioned studies, different wireworm and EPN species were utilized in the bioassays which might be one possible reason leading to variations in the mortality rates. Earlier studies illustrated species-dependent nematode infection among a range of hosts [47,48]. Unsuitable EPN species-host combinations may be partially responsible for the differences in the mortality rates of wireworm larvae. In the current study, all EPN species and strains that were collected from potato fields infested with A. sputator and A. rufipalpis were used in the bioassays. It is reasonable to assume that tested EPN species/strains were predisposed to successfully infect and kill wireworm larvae. The co-existence of the tested EPNs within the host habitat may have helped to give additional positive responses to wireworm-derived cues and eventually to explain the high infection rates [47,49]. On the other hand, infection of EPNs requires successful penetration of IJs into the host body. Wireworms are considered to have strong morphological structures that help them to avoid or limit the penetration of IJs [50]. The differences in the morphological structures of wireworm species that function as physical barriers to EPNs might be another reason behind differences in the mortality rates. In addition, the immune systems of host species that detect the presence of microbial infection play a key role in the pathogenicity of IJs. Rahatkhah et al. [51] indicated that there could be a great variation in the recognition of IJs by the immune system of different wireworm species. Conversely, EPNs and their bacterial associates produce a large number of metabolites in the host hemocoel that exhibit immunosuppressant activity with varying levels of efficiency as well as toxicity to the host intestine [52][53][54]. A great variation in the chemical composition of secondary metabolites produced by different species and strains of Xenorhabdus and Photohabdus bacteria was also reported in earlier studies which may have contributed to the differences in the mortality of wireworm larvae [55,56]. Furthermore, the developmental stage of the host insect is one of the key factors affecting the pathogenicity of EPNs. Williams et al. [38] reported that smaller M. communis larvae showed higher susceptibility to EPNs compared to larger wireworm larvae. Morton and Garcia-del-Pino [42] reported 75% mortality against the 5th/6th instars larvae of A. obscurus which is in line with our study. The immune responses of insects show variation among developmental stages and late instars larvae may have more immune responses against pathogens [57,58]. Ebssa and Koppenhöfer [59] reported that the 4th and 5th instars of Agrotis ipsilon (H.) (Lepidoptera: Noctuidae) were the most susceptible stages to EPN species. In another study, Abdolmaleki et al. [58] stated that the 4th instar larvae of Pieris brassicae (L.) (Lepidoptera: Pieridae) demonstrated higher immune responses to bacterial associates of EPNs Photorhabdus temperata subsp. temperate than the 3rd instars. In the aforementioned studies, the efficacy of EPNs was tested against mixed instars of wireworm species and this might have also affected the effectiveness of EPNs. In addition, considering the number of larval instars of wireworm species (up to 13 instars), the 4th/5th instars larvae of A. sputator and A. rufipalpis may be the most susceptible stages against tested EPNs [35].
Environmental variables also have an influence on the effectiveness of EPNs, and the adaptation capability of EPNs to environmental factors varies greatly among species and strains [60,61]. Temperature and humidity are among the major environmental factors that can either enhance or diminish the survival, mobility, and virulence of EPNs [62][63][64]. In the present study, a significant increase was observed in the efficacy of H. bacteriophora strains, particularly at low concentrations at 18 DAT. In previous studies, the optimum temperatures were reported to range between 22 and 24 • C for S. feltiae and 14 and 35 • C for S. carpocapsae, while H. bacteriophora reported to perform better between 25 and 30 • C [65,66]. The higher performance of H. bacteriophora strains at low concentrations may be explained by the optimal temperatures needed by the strains to successfully infect their host.

Conclusions
In this study, all tested EPN species/strains were highly pathogenic to A. rufipalpis and A. sputator larvae, and mortality over 70% was achieved at 18 DAT by all EPN species/strains. The results obtained indicated that EPN species that were recovered from the potato fields where EPNs and wireworms co-exist have the potential to provide better control against wireworms. However, field evaluation of these EPN species and strains will provide better insights into the performance of EPNs. In field conditions, several wireworm species in different development stages may be present with varying susceptibility to EPNs. Therefore, combining several EPN species may help suppress the wireworm populations.  Data Availability Statement: Data generated in this study are available upon reasonable request to the corresponding author.