Compatibility of entomopathogenic nematodes with insecticides against the cabbage white butterfly, Pieris rapae L. (Lepidoptera: Pieridae)

Compatibility of entomopathogenic nematodes (EPNs) with insecticides is a crucial mainstay of integrated pest management (IPM) programs. This study was designed to evaluate the joint action of EPN species and insecticides when employed to deter 3rd and 4th larval instars of cabbage white butterfly, Pieris rapae L. (Lepidoptera: Pieridae) under laboratory conditions. EPNs [Steinernema carpocapsae (All strain), S. feltiae (Filipjev), Heterorhabditis bacteriophora (HP88), and H. bacteriophora (Ar-4)], at concentrations of 50, 100, and 125 IJs/larva, were tested with 3 insecticides (lambda-cyhalothrin, emamectin benzoate, and indoxacarb) at LC25 and LC50 values. Additionally, expression profiles of 2 detoxification genes (CYP6AE120 and PrGSTs1) when the 4th instar larvae were treated by H. bacteriophora (HP88) and lambda-cyhalothrin were examined. Data indicated that statistically significant mortality of 2 larval instars of P. rapae was observed in vitro among EPN species and pesticide concentrations. At concentration of 50 IJs/larva, LT50 values were 2.385 and 3.92 days for S. carpocapsae (All strain) and H. bacteriophora(Ar-4), respectively, on 3rd instar larvae; also, these values were 3.506 and 3.107 days for S. feltiae and H. bacteriophora (Ar-4), respectively, on 4th instar larvae Lambda-cyhalothrin was the most toxic insecticide, followed by emamectin benzoate and indoxacarb at LC25 and LC50, respectively. An additive effect was observed between EPN species with LC25 and LC50 of the tested insecticides, except for lambda-cyhalothrin at LC50 with H. bacteriophora (Ar-4), and indoxacarb, with all EPNs showing antagonistic effects on mortality of 3rd instar larvae after 3 days post-treatment. The interaction between the tested pesticides at LC25 and LC50 and EPN species, showed an additive effect, excluding lambda-cyhalothrin at LC25 with S. carpocapsae (All strain) and LC25 of indoxacarb with H. bacteriophora (Ar-4), which showed potentiation effects. The interaction of S. feltiae (Filipjev) with tested insecticides at LC50 exhibited an antagonistic effect on the mortality of 4th instar P. rapae larvae after 3 days post-treatment. The expression of both CYP6AE120 and PrGSTs1 was significantly up-regulated with lambda-cyhalothrin, followed by H. bacteriophora (HP88) compared to control. The findings suggested that combining EPNs and the pesticide concentrations can be a practical strategy for managing P. rapae and could pave the way to using new control technologies in protecting organic farm vegetables from lepidopteran pests.

Page 2 of 12 Aioub et al. Egyptian Journal of Biological Pest Control (2021) 31:153 Background Pieris rapae L. (Lepidoptera: Pieridae), known as cabbage white butterfly, is a major agricultural pest that causes severe losses in crop yields of Brassicaceae worldwide (Kingsolver 2000). P. rapae larvae consume cabbage leaves, reducing them to stalks (Xiang et al. 2018). It also contaminates plants by depositing copious feces on leaves (Reda et al. 2018). However, the widespread use of synthetic pesticides to control the pest not only harmful to the environment, human health, and non-target creatures (Fantke et al. 2012), but also induces resistance in many pests, seriously compromising the effectiveness of the pesticides (Qie et al. 2020). Control of P. rapae is becoming increasingly challenging due to its high reproductive capacity, broad temperature tolerance, concealed feeding of the larvae, and failure to succumb to numerous chemical pesticides. Consequently, biological control offers an attractive alternative to traditional insecticides El-Ashry et al. 2021). Integrated pest management (IPM) can be used to limit the harmful environmental effects of some chemicals, while achieving effective and long-term pest control (Kary et al. 2021). One category of biocontrol agents that appears to be compatible with a wide range of chemical pesticides is entomopathogenic nematodes (EPNs) (Kary et al. 2018).
EPNs from the Steinernematidae and Heterorhabditidae families are fatal a biological control agent that treat a wide range of economically significant insect pests (Laznik and Trdan 2014). EPNs penetrate the insect body and then kill their hosts by releasing symbiotic bacteria during the infective juvenile (IJ) stage (Dara 2017). EPNs can be as effective as chemical insecticides under ideal conditions, including appropriate temperature, humidity, and low UV exposure (Laznik and Trdan 2017). EPNs and other agrochemicals can be used simultaneously or sequentially to control certain pest species or life stages. The use of multiple control agents in combination can improve the efficacy of IPM techniques and provide a cost-effective and time-saving alternative for P. rapae management. When two control agents work separately on a target host and the toxicity of one component is unaffected by the other, their combined effects can either be additive, potentiating, or antagonistic (Robertson et al. 2017). Studies suggest that mixing low-impact insecticides or lower insecticide concentrations with biological control agents could improve the biological control agents' effectiveness, while reducing the negative effects of insecticides (El-Ashry and Ramadan 2021). For instance, the mixture of Steinernema carpocapsae and abamectin was used to effectively management of the potato tuber moth, Phthorimaea operculella Zeller (Kary et al. 2021), and Heterorhabditis indica, S. carpocapsae, and indoxacarb were effective together for the control of Spodoptera litura (Fab.) (Khan et al. 2021).
The objectives of this study were to evaluate the compatibility of EPNs and managing the 3rd and 4th larval instars of P. rapae under laboratory conditions and examined the expression profile of 2 genes in 4th instar larvae of P. rapae in response to different concentrations of EPNs and the tested pesticides.

Propagation of EPNs
EPN species used in all experiments were propagated on last instar larvae of G. mellonella. New cultures less than one week old were used and EPNs' cultures were renewed by placing 2 filter papers (Whitman No.1) in a 9-cm Petri dish. Five G. mellonella were placed in each Petri dish and the dish was covered. All EPNs' cultures were held at 25 °C ± 3 °C for 2 days in the laboratory; dead larvae were transferred to white traps (Kaya and Stock 1997). After 15-20 days, collected EPNs and their suspensions were adjusted to 1000 IJs/ml and stored in shallow distilled water in transfer flasks at 12 °C for up to 7 days prior to use.

Source of insects
Third and 4th larval instars of P. rapae were collected from cabbage fields in the Zagazig district, Sharqia Governorate, Egypt in March 2021. Larvae were placed in wooden cages (25 × 20 × 18 cm) containing leaves of local cabbage from the collection sites (Brassica oleraceae, Cruciferaceae) as a food source, and held in the laboratory at 25 ± 3 °C. Fresh leaves were provided regularly ad libitum; the developing larvae shifted from old to fresh leaves independently.

Virulence of EPNs with P. rapae larvae
Comparative infectivity of imported nematodes and local isolate Ar-4 of H. bacteriophora was measured under laboratory conditions (25 ± 3 °C) using Petri dish assays on 3rd and 4th larval instars. A 9-cm Petri dish lined with 2 filter papers (Whitman No.1) and approximately 50, 100 or 125 IJs/larva was used in 0.5 ml and each dish was seeded with 5 healthy 3rd or 4th instars of P. rapae larvae. Control Petri dishes received only distilled water. Each treatment was replicated 5 times and P. rapae mortality rate was recorded daily for 5 days. The percentage of mortality and median lethal time (LT 50 ) of nematodes were estimated. Dead larvae after one, two, three, four, and five days were observed and transferred to the white trap. Dead larvae were examined, using a dissecting microscope for the presence of tested EPN species. Percentage of mortality was calculated according to the following equation:

Bioassay of tested insecticides with P. rapae larvae
Emamectin benzoate, indoxacarb, and lambda-cyhalothrin were tested against 3rd and 4th larval instars of P. rapae in Petri dish assays. The effectiveness of each pesticide at concentrations of LC 25 and LC 50 was estimated. Each Petri dish was lined with 2 pieces of filter paper (Whitman No.1) and 5 healthy 3rd or 4th larval instars were introduced. Immediately, 2 discs (5-cm diameter) of local cabbage immersed in LC 25 and LC 50 of each insecticide concentration for one min and air-dried for 3 min were introduced to 3rd or 4th larval instars. Control Petri dishes received only 2 fresh cabbage discs washed in distilled water. As mentioned before, each treatment was replicated 5 times and P. rapae mortality percentage was recorded daily for 5 days.

Combine effect of a low EPNs' concentration with tested insecticides on P. rapae larvae
Combined effect of lambda-cyhalothrin, emamectin benzoate, and indoxacarb and 50 IJs/larva of tested EPNs on Mortality (%) = Number of dead larvae / Total number of larvae × 100 3rd and 4th larval instars were investigated. Five active larvae (3rd or 4th instars) were placed in each Petri dish (9-cm diameter) lined with moist filter paper and provided with 2 discs of cabbage leaves (5-cm diameter) rinsed in LC 25 and LC 50 of emamectin benzoate, indoxacarb and lambda-cyhalothrin for one min and left to air-dry for 3 min. These were sprayed by 0.5 ml of EPN suspension containing 50 IJs/larva. Petri dishes were immediately sealed tight and were incubated under laboratory conditions at 25 ± 3 °C. Control Petri dishes contained only 5 larvae for each instar provided with 2 fresh cabbage discs washed in distilled water. Each instar (3rd or 4th) of larvae treated with LC 25 and LC 50 of insecticides and EPNs was replicated 5 times and P. rapae mortality percentage was recorded daily for 5 days.
The joint action of the EPNs and pesticides was estimated according to Richer's (1987) formula: where E = expected effect of the combination and X and Y = mortality percentages resulting from X and Y, respectively.
The expected effect was compared to the actual effect obtained experimentally from the insecticide interaction mixture according to Mansour et al. (1966): Three categories were used to classify the co-toxicity factor effect. A co-toxicity factor ≥ + 20 was considered potentiation, while ≤ − 20 was considered as antagonism and − 20 to + 20 indicated an additive effect.

Gene expression
Expression profiles of 2 genes (CYP6AE120 belonging to Cytochrome P-450 and PrGSTs1 belonging to glutathione S-transferase) were determined according to Liu et al. (2018). These were investigated by RT-qPCR in different concentrations of the 2 best treatments (H. bacteriophora (HP88) and lambda-cyhalothrin) for controlling 4th instar larvae of P. rapae.

RNA extraction
RT-qPCR was used to investigate the expression profiles of CYP6AE120 and PrGSTs. RNA extraction from tissue samples was conducted using QIAamp RNeasy Mini kit (Qiagen, Germany, GmbH) when 30 mg of the sample was added to 600 µl RLT buffer containing 10 μl β-mercaptoethanol per 1 ml. For homogenization of samples, tubes were placed into adaptor sets, which were fixed into the clamps of a Qiagen tissue Lyser. Disruption was performed in a two-minute high-speed (30 Hz) shaking step. One volume of 70% ethanol was added to the cleared lysate, and the steps were completed according to the "purification of total RNA from animal tissues" protocol of the QIAamp RNeasy Mini kit (Qiagen, Germany, GmbH). One column DNase digestion was done to remove residual DNA. Primers were used in a 25-µl reaction containing 12.5 µl of the 2 × QuantiTect SYBR Green PCR Master Mix (Qiagen, Germany, GmbH), 0.25 µl of RevertAid Reverse Transcriptase (200 U/µL) (Thermo Fisher), 0.5 µl of each primer of 20 pmol concentration, 8.25 µl of water, and 3 µl of RNA template. Primers for RT-qPCR are listed in Table 1. The housekeeping gene (β-actin) was used as an internal reference to normalize target gene expression. The reaction was performed in a Stratagene MX3005P real-time PCR machine. The relative quantification of gene expression was computed using the 2 −ΔΔCt method (Yuan et al. 2006).

Statistical analysis
A complete randomized design was implemented in all experiments. Data were analyzed using ANOVA with CoStat version 6.45. Means were compared via Tukey's multiple range tests at P ≤ 0.05 probabilities. The median lethal time (LT 50 ) and median lethal concentrate (LC 50 ) values were calculated by probit analysis (Finney 1971) using AnalystSoft Biostat Pro V 5.8.4.3 software.
As shown in Table 4, H. bacteriophora (Ar-4) (All strain) was the most virulent nematode species with the 3rd and 4th larval instars with LT 50 values of 3.292 and 3.107 days, respectively. While this value was 2.763 days by S. carpocapsae (All strain) with the 3rd instar larvae.. Conversely, S. feltiae was less virulent to the 4th instar larvae with LT 50 value of 3.506 days.

Toxicity of tested pesticides to the 3rd and 4th larval instars of P. rapae
The respect LC 25 and LC 50 values were (1.52 and 2.48 ppm) for lambda-cyhalothrin, (2.30 and 4.82 ppm) for emamectin benzoate and (3.12 and 6.31 ppm) for indoxacarb at 3 days post treatment for the 3rd instar P. rapae larvae. Whereas for the 4th instar P. rapae larvae, the respect of these values were (1.92 and 3.58 ppm) for lambda-cyhalothrin, (2.88 and 5.16 ppm) for emamectin benzoate and (3.67 and 6.97 ppm) for indoxacarb at 3 days post treatment.
The 3rd and 4th larval instars of P. rapae displayed different mortality percentages and LT 50 values after treatment    (Table 5). The mortality of 4th instar larvae 3 days post-treatment at LC 25 and LC 50 were (40 and 80%) with lambda-cyhalothrin, (28 and 56%) with emamectin benzoate and (48 and 20%) with indoxacarb ( Table 6). As shown to Table 7, LT 50 value of 3.34 4 days at LC 25 of lambda-cyhalothrin revealed the most virulent pesticide on 3rd instar larvae, while for emamectin benzoate this value was 3.84 days on the 4th instar larvae. LT 50 values of indoxacarb were 4.22 and 4.12 days for 3rd and 4th larval instars, respectively. On the other hand, emamectin benzoate at LC 50 was the most virulent pesticide with LT 50 values of 2.61 and 2.74 for 3rd and 4th larval instars, respectively. Indoxacarb at LC 50 was the least virulent

Expression of PrGSTs1 and CYP6AE120 in larvae exposed to H. bacteriophora (HP88) and lambda-cyhalothrin
The expression of CYP6AE120 and PrGSTs1 was investigated using H. bacteriophora (HP88) at 50,100,125 IJs/ larva and lambda-cyhalothrin at LC 25 and LC 50 values to control the 4th instar larvae of P. rapae, respectively (Fig. 2). The result revealed that the expression of CYP6AE120 and PrGSTs1 was significantly up-regulated at all concentrations of H. bacteriophora (HP88) and lambda-cyhalothrin than the control. Moreover, the expression of the 2 tested genes caused significant changes in 4th instar larvae of P. rapae with lambdacyhalothrin than H. bacteriophora (HP88).

Discussion
Entomopathogenic nematodes (EPNs) are lethal insect parasites that live in soil and belong to the families Steinernematidae and Heterorhabditidae in the Phylum Nematoda. They have proven to be the most efficient biological control organisms for soil and above-ground pests (Laznik et al. 2010). In the present study, H. bacteriophora (HP88) was the most effective biological control nematode species, followed by H. bacteriophora (Ar-4). S. carpocapsae (All strain) at different concentrations was less effective, while S. feltiae was the least successful one on 3rd and 4th larval instars of P. rapae. The LT 50 value of H. bacteriophora (HP88) demonstrated that it was the most virulent nematode 3 days post-treatment, followed by H. bacteriophora (Ar-4), then S. carpocapsae (All strain). S. feltiae was the least virulent nematode according to its LT 50 value on 3rd and 4th instars P. rapae larvae. Reda et al. (2018) reported that the cumulative mortality of P. rapae increased as a result of S. carpocapsae infection. Moreover, S. carpocapsae and S. feltiae are known to have a role in controlling the 4th instar of Fig. 2 Relative expression levels of PrGSTs1 and CYP6AE120 in larvae exposed to three different concentrations (50 IJs/larva, 100 IJs/larva and 125 IJs/larva) of H. bacteriophora (HP88) (a), LC 25 and LC 50 values of lambda-cyhalothrin (b). The transcriptional level of each gene in H. bacteriophora or lambda-cyhalothrin-treated individuals was normalized relative to that in acetone-treated (control) individuals. *P < 0.05 indicated the relationship between CK (control) and other treatment (Tukey's post hoc test, p < 0.05) Page 10 of 12 Aioub et al. Egyptian Journal of Biological Pest Control (2021) 31:153 P. rapae larvae (Salem et al. 2007). In this respect, the authors found that the larvae of P. rapae and Agrotis ipsilon were very sensitive to the nematode and both insects died 3-5 days after inoculation. Furthermore, Saleh (1995) found in a bioassay test that the Heterorhabditis tayserae nematode at concentrations of 5-100 IJs/larva of P. rapae induced 30-100% and 55-100% mortality within 24 and 48 h., respectively. Additionally, larvae of P. rapae infected with S. feltiae exhibited mortality 3 days after exposure ranging from 75 to 97.5% (Wu and Chow 1989). S. carpocapsae (All strain) and S. carpocapsae S2 were more virulent to the 2nd larval instar of P. rapae than the 5th instar, but Hemidesmus indicus SAA2 and H. bacteriophora HP88 were more virulent heterorhabditids to the 5th larval instar than the 2nd one (Salem et al. 2007). A similar result was noted by Belair et al. (2003) who demonstrated a significant relationship between the mortality of P. rapae and the time of their exposure to IJs. Obtained data indicated that the LC 25 and LC 50 values of lambda-cyhalothrin, emamectin benzoate, and indoxacarb all reduced the numbers of 3rd and 4th larval instars of P. rapae. Nevertheless, lambda-cyhalothrin is a better treatment than the other pesticides. This may be due to its different of mode of action from the other tested pesticides. Toxicological interactions of insecticides with EPNs can cause the final toxicity of the mixture to decrease (antagonism), can add to toxicity (additive) and increase toxicity (synergism or potentiation) with some chemicals (Capinera 2008). The findings revealed the additive effect was observed between EPN species with LC 25 and LC 50 values of lambda-cyhalothrin and emamectin benzoate on the mortality of 3rd instar P. rapae mortality, except that lambda-cyhalothrin at LC 50 with H. bacteriophora (Ar-4) exhibited an antagonistic effect 3 days post-treatment. An antagonistic effect was observed with LC 50 of indoxacarb. An additive effect was observed with all tested insecticides at LC 25 with EPN species, except for lambda-cyhalothrin with S. carpocapsae (All strain) and indoxacarb with H. bacteriophora (Ar-4) that showed potentiation effects on 4th larval instar mortality of P. rapae 3 days post-treatment. The interaction of the tested pesticides at LC 50 with EPNs showed an additive effect on 4th larval instar mortality of P. rapae 3 days post-treatment, excluding the interaction of S. feltiae with tested insecticides at LC 50 which exhibited an antagonistic effect. This result may be because some pesticides have different effects on nematodes due to the differentiation of nematodes and insect physiology and feeding patterns (Özdemir et al. 2020). The pattern of interaction was dependent on nematode species (Zhang et al. 2020). This result confirms the work of Fetoh et al. (2009) who mentioned that the mortality numbers of A. ipsilon increased when treated with a mixture of S. carpocapsae and indoxacarb. In addition, S. carpocapsae (All strain) in combination with LC 50 of indoxacarb was much more effective against Tuta absoluta than by using it alone (Sabry et al. 2016). Combined applications of insecticides and EPNs (H. bacteriophora and S. feltiae) allow for a reduction in the rates of chemical pesticide and EPNs usage (Kary et al. 2021). There is compatibility of S. feltiae with pesticides to reduce the number of T. absoluta (Garcia-del-Pino et al. 2013). Additionally, an antagonistic effect of S. feltiae with dichlorvos, abamectin, and azadirachtin was detected when the nematodes were applied immediately after insecticide application (Amizadeh et al. 2019). Another study reported that oxamyl increased S. carpocapsae efficacy against Agrotis segetum synergistically (Amin 1999). Vashisth et al. (2013) reported that combinations of chemical agent with nematodes resulted in synergistic levels of insect mortality.
In this study, lambda-cyhalothrin caused significant changes in expression of CYP6AE120 and PrGSTs1 compared to H. bacteriophora (HP88), suggesting that there might be a different mode of action between pesticides and EPNs. In addition, many employing functional expression assays have found that resistance to pesticides can be achieved through elevated expression of pesticide-metabolizing enzymes such as members of cytochrome P450 and glutathione-S-transferases groups (Chung et al. 2007). Liu et al. (2018) reported that 4 genes (PrGSTe3, PrGSTs1, PrGSTs2, and PrGSTs4) belonging to glutathione-S-tranferase were expressed predominantly in different tissues of P. rapae in response to abamectin. Furthermore, CYP6AE120 exhibited an up-regulated expression level in P. rapae when exposed to chlorantraniliprole. Therefore, these genes are potential candidates involved in the detoxification of the tested insecticides.

Conclusions
This study indicated that all the EPNs induced high mortality in the P. rapae larvae and gave promising results. Although heterorhabditid species appeared to be more virulent than steinernematid ones, generally, there were statistically insignificant differences between the EPN species. The local species H. bacteriophora (Ar-4 strain and the imported species, H. bacteriophora (HP88 strain) achieved a high mortality 3 days post-treatment of the 3rd and 4th larval instars of P. rapae when combined with the LC 25 and LC 50 values of insecticides (lambda-cyhalothrin, emamectin benzoate, and indoxacarb) using an appropriate concentration (50 IJs/larva). Hence, H. bacteriophora (HP88 strain) and H. bacteriophora (Ar-4 strain) have the potential to be used as a biocontrol agent for the integrated management of valued vegetable crops that