Additive interactions of some reduced-risk biocides and two entomopathogenic nematodes suggest implications for integrated control of Spodoptera litura (Lepidoptera: Noctuidae)

Higher volumes of conventional and novel chemical insecticides are applied by farmers to control resistant strains of armyworm (Spodoperta litura) in Pakistan without knowing their risks to the environment and to public health. Ten reduced-risk insecticides were tested for their compatibility with two entomopathogenic nematodes (EPNs); Heterorhabditis indica and Steinernema carpocapsae against S. litura. The insecticide emamectin benzoate was highly toxic (LC50 = 2.97 mg/l) against 3rd instar S. litura larvae when applied alone whereas, novaluron and methoxyfenozide were the least toxic (LC50 = 29.56 mg/l and 21.06 mg/l), respectively. All the insecticides proved harmless against the two EPNs even 96 h after treatment. Indoxacarb, flubendiamide and spinetoram produced the greatest mortalities (72–76%) of S. litura larvae after 72 h when applied in mixtures with H. indica. Lowest mortalities (44.00 ± 3.74% and 48.00 ± 2.89) were observed for mixtures of H. indica with methoxyfenozide and chlorfenapyr, respectively. The positive control treatments with both EPNs (S. carpocapsae and H. indica) produced > 50% mortality 96 h after treatment. For insecticide mixtures with S. carpocapsae, only indoxacarb produced 90% mortality of larvae, whereas, indoxacarb, flubendiamide, emamectin benzoate, and spinetoram produced 90–92% mortality of larvae when applied in mixtures with H. indica. Additive interactions (Chi-square < 3.84) of EPN mixtures with reduced volumes of reduced-risk insecticides suggest opportunities to develop more environmentally favorable pest management programs for S. litura.

Insects. Wax worms, G. mellonella, were collected from infested beehives, and resulting adults were released in plastic jars, measuring 5 cm diameter and 30 cm depth. These jars were provided with cotton swabs soaked with a 10% honey solution for moth feeding and folded card sheets for oviposition. Newly emerged larvae were reared on a wheat-based semi-natural diet, in the laboratory following the mass-rearing technique reported by Khan et al. 28 and Ashfaq et al. 29 .
Although extensive field surveys were performed to collect S. litura larvae, we were only successful in collecting insects from the research fields (Lat. 31.437778°; Long. 73.063611° and Lat. 31.390556°; Long. 73.018056°) of the University of Agriculture, Faisalabad. Upon reaching the laboratory, larvae were reared on washed and dried, fresh leaves of cotton, Gossypium hirsutum L., and kept in an insect rearing chamber at 26 ± 2 ºC, 75 ± 10% RH, and 12:12 LD. Field collected adults and those emerging from pupae in the laboratory were kept in well-ventilated cages (30 × 30 × 30 cm) and provided a cotton swab dipped in a sucrose-solution (100 g/l), also containing vitamin solution (20 ml/l) and methyl 4-hydroxybenzoate (2 g/l) for adult feeding 3,8,12 . The homogenous lots of insects were used for bioassays with EPNs and insecticides.
Insecticide formulations. Ten insecticides with novel modes of action were selected for evaluation along with the selected EPN species against S. litura. All of the selected insecticides are registered for use against Spodoptera spp. in Pakistan 30 . Further details about the recommended dose rates, active ingredients, mode of action, target pests, and suppliers/distributors in Pakistan are provided in Table 1. The test concentrations of each formulated insecticide were prepared in distilled water and were used in all bioassay tests 1,3,8,31 . Toxicity of insecticides to S. litura. Preliminary bioassays were performed to estimate the range of insecticides' lethal concentrations killing 5-95% of 3rd instar larvae of S. litura. Groups of 10 larvae were used for each insecticide treatment, with insecticide solutions applied using a leaf dip method 1,3,8 . A single leaf disc (5 cm diameter) dipped in each insecticide concentration for 10 s and dried on a paper towel, was placed on moist filter papers to avoid desiccation in Petri plates. Each insecticide concentration was repeated five times and leaves were treated separately at different time with new insecticide solution of same concentration prepared on each occasion. The leaves dipped in distilled water alone served as the untreated check in all replicates. All treated and untreated larvae were kept in an insect growth chamber at 26 ± 2 ºC, 75 ± 10% RH, and 12:12 LD. Mortality was determined 96 h after treatment applications. Larvae were considered dead when no movement of appendages was seen upon touching with a dissecting needle. Toxicity of insecticides and EPN mixtures to S. litura. The toxicity of ten selected insecticides and the EPN mixtures was evaluated in S. litura using previously reported methodology 7,12,26 . Plastic containers (150 ml) were provided with wet sand (10%, w/w) and five grams of insect diet. To the sand and diet in each container was further applied a solution containing the median lethal concentration (LC 50 ) of each insecticide (  EPNs. In the case of bioassays with insecticides and S. litura, mortality scores were corrected using Abbott's formula 33 . The corrected mortality values were further subjected to Probit analysis and the median lethal concentrations (LC 50 ) were calculated using Polo Plus software Version 1.0 (LeOra Software LLC). The lethal concentrations of various insecticide treatments were reported with significant differences when 95% confidence intervals (CIs) were non-overlapping 34 . Data for the toxicity of the test chemical insecticides against the selected EPNs were analyzed statistically through Minitab 18 Software and the means were compared for significance using Tukey's HSD test (P < 0.05).

Toxicity of insecticides to
The mean mortality scores obtained in the combined treatments of insecticides and EPNs were also subjected to ANOVA and means were compared for significance as stated above. However, comparisons of the expected versus observed mortality of S. litura was performed through a binomial test and the interaction between insecticides and EPN mixtures was described as additive, antagonistic or synergistic 7,12,20 . Expected larval mortality in each treatment was estimated as: where, P E symbolizes mortality (expected) in mixture treatments of EPNs and insecticides, P O denotes mortality of S. litura in the control treatment, P 1 symbolises the mortality of S. litura due to insecticide alone and P 2 represents the mortality (observed) in the treatment with EPNs alone.
A chi-square (χ 2 ) value was calculated by using the formula where L o displays the observed numbers of live larvae in the treatment, L e denotes the expected number of live larvae in the treatment, D o is the number of observed dead larvae and D e is the number of expected dead larvae in the treatment.
We tested the hypothesis of treatment interactions by using a 3.84 value of Chi-square (df = n − 1; and P = 0.05). Synergy was assigned to the insecticide and EPN treatments when observed mortality (P o ) was higher than the expected mortality (P e ) and the χ 2 was higher than 3.84. The interaction was considered additive when the χ 2 value < 3.84 and antagonism was defined as treatment mixtures where observed mortality (P o ) was less than expected mortality (P e ) and the χ 2 was higher than 3.84 7,12,20 .
Ethical statement. As no human or mammalian subjects were involved in this research, hence, no ethics approvals were required for this study. However, we adapted all the standard bioassay procedures while conducting the experiments.

Results
Toxicity of insecticides to S. litura. The selected insecticides were tested for their toxicity against the S. litura larvae to obtain their relevant median lethal concentrations (LC 50 ). These median lethal values were further used in treatment mixtures with the EPN spp. The results of Probit analyses are displayed in Table 2. The insecticide emamectin benzoate was found to be highly toxic (LC 50 = 2.97 mg/l) against 3rd instar S. litura larvae. The highest median lethal concentration (LC 50 = 29.56 mg/l) was determined for the insecticide novaluron which was followed by methoxyfenozide (LC 50 = 21.06 mg/l), whereas the toxicity scores of chlorfenapyr (LC 50 = 17.16 mg/l), spinosad (LC 50 = 14.78 mg/l) and indoxacarb (LC 50 = 10.92 mg/l) did not differ significantly from each other because of overlapping confidence intervals. The insecticides chlorantraniliprole, flubendiamide and lufenuron also proved very toxic against 3rd instar S. litura larvae and exhibited lower LC 50 values (5.50, 7.41 and 7.85 mg/l, respectively).

Toxicity of insecticides to EPNs.
A series of conventional mortality bioassay tests were performed to determine the toxic effects of the field recommended dose rates of ten insecticides against EPNs and the results are displayed in Table 3. We found significant variation among all the tested insecticides against both EPN species (S. carpocapsae and H. indica) after different exposure times (P < 0.05). All the tested insecticides proved to be harmless against the two tested EPN species (S. carpocapsae and H. indica) based on less than 20% mortality, even after 96 h of treatment exposure, and less than 10% mortality in both EPN spp., after 72 h. The highest mortality (16.27 ± 0.9) of H. indica was recorded after 96 h with chlorfenapyr, followed by spinosad (14.89 ± 1.1), flubendiamide (13.37 ± 1.0), spinetoram (13.22 ± 1.0) and indoxacarb (13.21 ± 1.0). Almost the same insecticides viz., chlorfenapyr, spinosad, indoxacarb, spinetoram and flubendiamide proved to be harmless against S. carpocapsae, causing mortality ranging from 11.00 to 14.33% after 96 h of treatment.

Toxicity of insecticides and EPN mixtures to S. litura. Toxicity of selected insecticides and EPN
mixtures were recorded after 72 and 96 h of treatment application. In our study, insecticides proved to be an important factor in S. litura mortality, along with the EPNs. The results revealed significant variations in the toxicity levels of all the treatments and exhibited considerable differences in the mortality of S. litura larvae in all insecticide mixtures with S. carpocapsae (F = 86.15, P < 0.001, df = 11, R-sq. = 95.21%) and H. indica (F = 73.67, P < 0.001, df = 11, R-sq = 91.79%) after 72 h of treatment application (Table 4).

S. litura mortality (%) and interactive response between insecticides and EPNs mixtures
All the insecticide and EPN mixtures exhibited additive interactions indicating that both insecticides and EPNs contributed significantly to S. litura larval mortality (Table 5). However, we did not observe any synergistic or antagonistic interactions between insecticides and EPNs even after 96 h of treatment. The positive controls with both EPN spp., produced more than 50% mortality after 96 h (S. carpocapsae = 51.00 ± 1.91% and H. indica = 62.00 ± 2.58%). In the case of insecticide mixtures with S. carpocapsae, indoxacarb produced 90% mortality of S. litura larvae, whereas, indoxacarb, flubendiamide, emamectin benzoate, and spinetoram produced 90-92% mortality when applied in mixtures with H. indica. The two insecticides viz., methoxyfenozide and chlorfenapyr produced minimum mortalities (64-68%) with EPN mixtures even after 96 h.

Discussion
Insecticides are most widely used by farmers to control several key pests of agricultural significance. Many insect pests including armyworms, Spodoptera spp., have been reported to develop resistance against the insecticides used in pest management programs 1,2,8 . Combining certain efficient techniques like biological control agents, especially EPNs, and insecticides can help address the challenges in management of certain pest insects in agriculture 7,12,13,21,25,35,36 .
As mentioned earlier, S. litura has developed resistance to many of the chemical insecticides used for its management in many countries. In the present study, emamectin benzoate was found to be highly toxic to 3rd instar S. litura larvae, whereas, novaluron and methoxyfenozide were the least toxic amongst the tested insecticides. We could not find any report of resistance development in Spodoptera spp. against novaluron; however, it has been reported to methoxyfenozide, a benzylhydrazide growth regulator (IGR) through leaf-dip bioassays 1 . Chlorfenapyr, spinosad, and indoxacarb were moderately toxic against 3rd instar larvae compared to the other tested insecticides.
Chlorantraniliprole, flubendiamide, and lufenuron also proved very toxic against 3rd instar S. litura larvae. Low to very low resistance was reported for Pakistani strains of the armyworm against chlorantraniliprole, flubendiamide, spinetoram, and spinosad 2 . Since the registration of lufenuron in Pakistan in 1996 30 , it gained popularity for successful control of the armyworm, however, moderate to high lufenuron resistance was reported in both S. litura and S. exigua in Pakistan 2,5 .
Before evaluating the efficacy of insecticide and EPN mixtures against S. litura, the toxic effects of the selected insecticides were first estimated against the two EPN spp., S. carpocapsae, and H. indica to appraise their compatibility with the tested insecticides. Our results indicate that all insecticides were relatively harmless or least toxic to the nematodes with mortalities less than 20% after 96 h of exposure. The insecticides caused less than 10% mortalities in both EPNs after 72 h of exposures, which encourages the integrated use of insecticides and EPNs as tank mixes are not usually prepared more than 24 h before spraying.
Studies have shown that delayed exposures of certain insecticides at higher concentrations can cause variable mortality in both EPN species (S. carpocapsae and H. indica) 12 , however, the insecticides remained harmless even after longer treatment exposition. Low EPN mortality was reported with certain other insecticides including chlorpyrifos 37 , whereas some conventional organophosphates including chlorphenvinphos and dichlorvos and their mixtures were reported to cause high mortality of Heterorhabditis amazonensis. Conversely, deltamethrin, Table 5. Toxicity of selected insecticides and nematode mixtures against S. litura after 96 h of treatment application. Sc S. carpocapsae, Hi H. indica. *Means followed by the same letter within a column do not differ significantly at 5% probability. a Median lethal concentrations (LC 50 ) of insecticides and 500 IJs of each EPN specie were applied in each treatment mixture. www.nature.com/scientificreports/ and chlorpyrifos-ethyl did not reduce the survival of that EPN species 26,38 . S. carpocapsae was also reported to be more susceptible to deltamethrin compared to H. bacteriophora and H. indica 25 . Similarly, some commercial formulations of fipronil, malathion, cypermethrin, imidacloprid, chlorantraniliprol, and azadirachtin were also reported to cause no harm to survival or infectivity of entomopathogenic nematodes 35 . In nematodes, the involvement of butyrylcholinesterase as a front line of defense was proposed because it may protect acetylcholinesterase from an early attack by its inhibitors 39 . Results of our study and some other reports suggest that several EPNs are compatible with many insecticides and hence can be used in combinations or mixtures in pest management programs [11][12][13]21,38 .

S. litura mortality (%) and interactive response between insecticides and EPN mixtures
Synergistic, additive, and antagonistic interactions have been reported when some conventional and novel chemical insecticides were tested in combination with S. carpocapsae and H. indica after 48 h of treatment exposures. However, with increased exposure time to 96 h, the interactions of EPNs and insecticides turned antagonistic 25 . The increased exposure period to the EPNs alone and in combination with insecticides have proven to increase the mortality of insect larvae 11,12,38 . Similarly, an increase in rates of H. indica and S. glaseri, resulted in higher mortalities in S. litura under lab conditions, whereas H. indica demonstrated greater efficacy in greenhouse-pot-culture conditions 14 . Increased rates of H. indica and S. carpocapsae also produced greater mortality of S. litura larvae after 48 and 96 h of exposure 12,24 .
Combinations of H. bacteriophora with chlorantraniliprole and imidacloprid demonstrated synergistic or additive interactions in second-instar larvae of Holotrichia oblita and produced faster mortality than the EPNs or insecticide alone. Combination of S. carpocapsae with chlorantraniliprole or spinetoram has been suggested as a least toxic control strategy against the fall armyworm, S. frugiperda 15 .
It can be inferred from our results that all insecticides proved toxic alone and when applied in mixtures with EPNs against S. litura larvae. We suggest that more EPNs should be investigated for their compatibility with insecticides against certain insect pests of agricultural economic significance.

Conclusions
Several reports describe the development of insecticide resistance in the armyworm S. litura against many insecticides included in pest management programs in many countries. Growers often apply huge volumes of insecticide to control resistant strains of pests without knowing their risks to the environment and to public health. Efforts are being made to investigate alternative methods to be used either alone or in combination with chemical insecticides for mitigation of this notorious pest. Entomopathogenic nematodes are among useful options for inclusion in IPM programs, used alone or in combination with chemical insecticides. The experiments conducted in the present study on compatibility of two EPN spp. with reduced doses of some novel reducedrisk chemical insecticides show significant potential for armyworm management. EPN mixtures with reduced volumes of insecticides will also help in lowering the risks of environmental pollution and to public health.

Data availability
Statistically nalysed data are presented in the manuscript. Observations recorded during the experiments are acquiesced to the Department of Entomology, University of Agriculture, Faisalabad, Pakistan and hence, the consent of authorities is required for obtaining the data.

Scientific Reports
| (2021) 11:1268 | https://doi.org/10.1038/s41598-020-79725-w www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.