Introduction

The standardization of methods to assess insecticide toxicity is required to test the susceptibility of target species or the active ingredient of interest in a repeatable manner. An insecticide is a stable, chemically defined substance, while test subjects, like insects, are biological organisms that commonly exhibit varied responses to the insecticides to which they are exposed. Therefore, results from insecticide trials need to be carefully analyzed since the response of the insect to the active ingredients can be affected by numerous biological traits (e.g., Potin et al. 2022). Thus, experiments testing insecticide toxicity need to consider sources of biological variation, including for example the target guild (phytophagous pest versus carnivorous predators or parasitoids), pest feeding habits (leaf chewing versus phloem feeding), and the mode of action of the test insecticide (contact versus stomach poisons).

Neonicotinoid insecticides are marketed for the control of pests in agricultural, forest, and urban environments. Commonly used neonicotinoid insecticides comprise five different active ingredients: imidacloprid, thiamethoxam, acetamiprid, dinotefuran, and clothianidin, all of which offer broad-spectrum control of insect pests. Neonicotinoids are formulated for foliar application, as seed dressing, incorporation into baits, and soil drenching via irrigation. This wide option of recommendation has raised the likelihood of non-target effects and environmental contamination (Goulson 2013; Pisa et al. 2015; Frank and Tooker 2020). These possibilities have attracted attention from scientific and nonscientific communities who are interested in assessing the non-target impacts of neonicotinoids (Wood and Goulson 2017; Mace et al. 2022).

Hippodamia convergens and Coccinella septempunctata (both Coleoptera: Coccinellidae) collected from citrus orchards in San Bernardino, California, U.S., showed high survival rates to dry residues of label (1X) and two times (2X) label doses of thiamethoxam, a commonly used insecticide that kills a wide range of pests and natural enemies (Rodrigues et al. 2013; Barbosa et al. 2016). These data suggested a high tolerance for both lady beetle species to thiamethoxam. Additionally, strains of H. convergens from the southeast U.S. have become resistant to the synthetic pyrethroid, lambda-cyhalothrin, and the organophosphate, dicrotophos (Rodrigues et al. 2013; Barbosa et al. 2016). In the laboratory, strains of Propylaea japonica (Thunberg) (Coleoptera: Coccinellidae) and C. septempunctata, field-collected from field sites in China exhibited resistance to imidacloprid (Tang et al. 2015) and thiamethoxam (Cheng et al. 2022), respectively. Thiamethoxam is used to control phloem-feeding insect pest species in various crops, including citrus orchards in California U.S., where applications target pest psyllids, aphids, mealybugs, and other pests (Boina and Bloomquist 2015; Mace et al. 2022). These applications of thiamethoxam in agricultural areas expose coccinellids to insecticide residues on plant foliage that may lead to resistance development.

Different methodologies have been used to measure insecticide toxicity, and assessing the efficacy of dry insecticide residues on filter paper, including neonicotinoids, is one commonly used approach (Magalhães et al. 2007; Dagg et al. 2019; Rehman et al. 2019; Ribeiro et al. 2021). Therefore, the main objective of this study was to elucidate the effects of dried neonicotinoid and pyrethroid residues using different substrates. Laboratory bioassays including filter paper, were conducted against an important natural enemy species, field-collected convergent lady beetles, H. convergens, and adults of southern green stinkbug, Nezara viridula L. (Hemiptera: Pentatomidae), a common pest species maintained as a laboratory colony. Surprisingly, bioassays with dry neonicotinoid residues on filter paper failed to cause mortality in cohorts of experimental insects. This unexpected finding resulted in the development and execution of experiments presented here that sought to verify this surprising result.

Material and methods

Sources of insects used in experiments

The first insecticide bioassay testing toxicity of thiamethoxam and cyfluthrin was conducted with field-collected adults of two coccinellid species, H. convergens and C. septempunctata. Adults were collected from a conventionally managed citrus orchard (var. navel orange), in San Bernardino, California, U.S. (34.087314 N, − 117.109077 W), and transferred to the laboratory. Two field collections were performed: April 15 and 26, 2022. Adult coccinellids were fed citrus aphids, Toxoptera citricada (Kirkaldy) (Hemiptera: Aphididae) harvested from infested citrus branches at the same field site coccinellids were collected from. Aphid-infested citrus branches were maintained in water using 1L plastic containers held inside bugdorm cages [model 6S610 (MegaView Science Co., Ltd., Taiwan)]. Field-collected coccinellids that were maintained on aphids were used in bioassays within 5 days of field collections.

Because of unexpected results from bioassays assessing the toxicity of dry thiamethoxam residues on filter paper (see below), a laboratory colony of H. convergens was established using field-collected adults with the intent of producing sufficient adults for use in experiments. Eggs laid by field-collected females were kept in a temperature and light controlled room at 25 °C and a 14:10 h (L:D) photoperiod in the Insectary and Quarantine Facility at the University of California, Riverside (IQF-UCR). Larvae were reared in ventilated 170-mL cylindrical vials (No. 50–50, Thornton Plastic Co., Salt Lake City, UT), and adults were maintained in ventilated 500-mL plastic containers for egg collection. Both larvae and adults maintained in colonies were fed ad libitum California lilac aphid, Aphis ceanothi Clarke (Hemiptera: Aphididae) collected from infested flowers of red yucca, Hesperaloe parviflora (Torr.) J.M. Coult. (Asparagales: Asparagaceae), found growing across the University of California Riverside campus.

A colony of N. viridula was maintained at the IQF-UCR at 25 °C and a 14:10 h (L:D) photoperiod. Both nymphs and adults were fed fresh green bean pods and raw redskin peanuts, and potted cowpea plants (Vigna unguiculata (L.) Walp (Fabales: Fabaceae). Adult pentatomids were used in bioassays to assess the toxicity of dried residues of test insecticides against a common widespread phytophagous pest.

Insecticides

In the first round of bioassays, thiamethoxam and cyfluthrin were tested against H. convergens and C. septempunctata at rates used to treat citrus for infestations of Asian citrus psyllid (ACP), Diaphorina citri Kuwayama (Hemiptera: Liviidae), a key citrus pest in California (Table 1). A second round of bioassays assessed the residual toxicity of four neonicotinoid insecticides, acetamiprid, dinotefuran, imidacloprid, thiamethoxam, and the pyrethroid, bifenthrin, a highly toxic contact insecticide (Snodgrass et al. 2005) against H. convergens and N. viridula. All experimental insecticides were tested using commercial formulations and label doses that were diluted to appropriate concentrations with deionized water (Table 1).

Table 1 Neonicotinoid and pyrethroid insecticides, label doses, dry residue methods, and total number of Hippodamia convergens and Nezara viridula adults evaluating three different dry residue methods for toxicity assessment
Full size table

Evaluation of toxicity of dried insecticide residues on filter paper (bioassay #1)

Determining the susceptibility of H. convergens and C. septempunctata collected from citrus orchards mortality rates of adult coccinellids was exposed to thiamethoxam and cyfluthrin residues applied at the label dose (1X) and two times the label dose (2X) were compared to mortality rates for their respective control groups (0X), that were not exposed to insecticide residues. We tested thiamethoxam at equivalent label doses (1X and 2X) compared to control groups (0X), using 49, 44, and 40 adult H. convergens and 51, 50, and 28 adult C. septempunctata. The cyfluthrin bioassay was conducted with 48, 50, and 30 H. convergens and 60, 51, and 30 C. septempunctata adults, respectively. Similar to thiamethoxam, we confined the insects to dry residues of cyfluthrin at doses equivalent to the recommended label doses (1X and 2X) and control groups. Each treatment had five replications, each consisting of 7–10 beetles.

For this filter paper bioassay, glass Petri dishes (15-cm diameter) were lined with filter paper of the same diameter and treated with 2.7 mL of insecticide-prepared solutions. The 2.7 mL volume was based on the recommended spray volume standardized for 10,000 sq m. The 2.7 mL was delivered to the filter paper using graduated pipette (100–1000 mL, Pipetemanä, Middleton, WI) and applied in a spiral pattern that started at the margin of the filter paper and ended in the center. The insecticide solution was allowed to dry at room physical conditions (temperature of 25 °C and ≈ 60% relative humidity) for 2 h after which field-collected adult coccinellids of unknown age were introduced into the test arena. A 1.5-mL microcentrifuge tube containing a 20% (v/v) solution of honey and water was provided as a food source for test insects. Mortality rates for insects exposed or not exposed (i.e., control groups) to dry insecticide residues were tallied 48 h after test insects were placed in their respective Petri dishes. Insects were considered dead if after being placed on its dorsum an individual was unable to turn upright and immediately walk. Mean percent mortality rates were calculated as ([# dead coccinellids /# treated coccinellids]*100).

Evaluating three different dry residue methods for toxicity assessment (bioassay #2)

The above experimental design testing the toxicity of dry thiamethoxam residues on filter paper towards field-collected coccinellids failed to induce mortality (see Results bioassay #1). Therefore, a second set of bioassays was carried out to evaluate different ways of obtaining dry neonicotinoid and one pyrethroid residues for testing residual toxicity towards adult coccinellids and adults of a phytophagous pest species, N. viridula. The intent of this second set of bioassays aimed to identify a technique that exposed test insects to insecticide residues that were bioactive. Consequently, three different techniques were used to obtain dry insecticide residues for use in bioassays: (1) insecticide solutions were applied to filter paper in a manner analogous to bioassay #1 (see above), (2) 2.7 mL of experimental insecticide solutions were applied directly to the bottom of a glass Petri dish (inert surface), and (3) insecticides were applied to green leaf discs (9-cm diameter) that were cut from mature leaves harvested from an Oroblanco® grapefruit tree. Leaf disks were dipped in insecticide solutions for 10–20 s. For all methods, the insecticide solution was allowed to dry at room physical conditions previously cited for ≈ 2 h before being exposed to the insects.

Insecticide solutions used in these bioassays were prepared using the maximum label dose recommended for controlling N. viridula by diluting the commercial formulation in deionized water (Table 1). Four neonicotinoids (imidacloprid, thiamethoxam, acetamiprid, and dinotefuran) and one pyrethroid (bifenthrin) were tested (Table 1). The pyrethroid was selected because of its previously reported toxicity as dry residues on filter paper to N. viridula (Stará et al. 2011; Fernandes et al. 2016). Details of commercial formulations, label doses, and the total number of insects assayed are summarized in Table 1. Adult coccinellids were provided honey water (20:80 v/v) as a carbohydrate source and N. viridula were provisioned with one raw redskin peanut and two fresh sweet corn kernels as food. The number of individual insects used in trials is provided in Table 1. Assessment of insect mortality and calculation of percent mortality were determined in the same manner as Bioassay # 1.

Determining if dried neonicotinoid residues on filter paper have bioactivity (bioassay #3)

This bioassay was conducted to determine if neonicotinoid toxicity was rendered inactive when applied filter paper because the insecticide was adsorbed to (1) the filter paper or (2) adsorbed to the glass bottom of the Petri dish. To test these two possibilities, two commercial neonicotinoid insecticides representing soluble (SC) and granule (WG) formulations, Assail® 30% SC (acetamiprid) and Actara® 25% WG (thiamethoxam), were prepared at the recommended label rate and tested (Table 1). The susceptibility of adult N. viridula to neonicotinoid insecticides applied to filter paper was assessed.

Filter papers (paper qualitative 1, 90 mm ∅, Cat No. 1001 090 Whatman™) were placed individually on the bottom of glass Petri dishes of the same diameter. As described in bioassay #2, 1 mL of the insecticide solution was applied to the filter paper. This resulted in the two test treatments: (1) insecticide residues adsorbed to the filter paper or (2) insecticide residues adsorbed to the inert glass surface (i.e., the bottom of the Petri dish). After application, the insecticides were left to dry at 25 °C and ≈ 60% relative humidity for ≈ 2 h. The treated filter paper or glass plates were used to test the following six inactivation scenarios:

  1. 1)

    The upper surface of treated filter paper inactivated dried residues (this known lack of toxicity was demonstrated in bioassay #2 [see Results]).

  2. 2)

    Bioactive dry residues formed on bottom of the glass Petri dish underneath the insecticide-treated filter paper after leaching through the paper. Removal of the filter paper from the Petri dish 2 h after application exposed test insects to lethal dry residues.

  3. 3)

    Bioactive dry residues formed on the underside of the treated filter paper and by turning the treated filter paper over lethal residues would be exposed to test insects.

  4. 4)

    The filter paper inactivated all insecticidal activity and lethal dry residues would result from direct application of insecticide to the glass bottom of the Petri dish.

  5. 5)

    Incorporation of nonionic surfactant adjuvant, 0.2% Tween® 20 (Agdia Inc., Elkhart, IN), to test insecticides would result in lethal dry residues on upper side treated filter.

  6. 6)

    Bioactive insecticide residues were adsorbed to the filter paper and could be recovered by soaking the treated filter paper in water and this water, when applied to an inert glass surface and allowed to dry would have insecticidal activity. To test this adsorption hypothesis, filter paper treated with 2.7 mL of insecticide that had dried for 2 h, was placed into a clean glass Petri dish and 5 mL of deionized water was added. The plate was gently shaken to move the water over the filter paper for approximately 90 s. The filter paper was then removed and discarded. The water in the Petri dish was left to dry for 2 h. Even though the dry residues obtained by soaking insecticide-treated filter paper in water could have lowered the concentration of insecticide, mortality of N. viridula was anticipated if this treatment had bioactivity. High rates of target pest mortality are observed using rates of bifenthrin lower than recommended label doses (Torres et al. 2022).

  7. 7)

    Control treatments were consisted of test insects and in a Petri dish lined with untreated filter paper.

Each treatment was replicated five to six times and each bioassay used 6–8 adult N. viridula 5–8 days of age. Mortality rates of N. viridula were recorded 48 h after placement in test arenas.

Statistical analyses

All statistical analyses were performed in R (R Core Team 2020) to compare the recorded variable mortality rate among treatments for all bioassays. The generalized linear mixed-effects models (GLMM) with binomial distributions were performed. The models were constructed using the glmer function from the lme4 package (Bates et al. 2014). The significance of the individual components and their interactions were assessed using the likelihood ratio test (LRT) with the “ANOVA” function in the car package (Fox and Weisberg 2018) (α = 0.05). Covariates and model fit were assessed by plotting confidence intervals (CI) around the observed values and best model fit was determined based on CI values, with smaller CI values indicating a superior model fit to data. Confidence intervals were calculated using the bootMer function in the lme4 package (Bates et al. 2014).

Results

Evaluation of toxicity of dried insecticide residues on filter paper (bioassay #1)

Field-collected adult H. convergens and C. septempunctata exhibited mortality that varied based on the deviance analysis as a function of tested insecticide concentrations (F2, 2 = 16.60, p < 0.0001), lady beetle species (F1, 2 = 23.14, p = 0.0001), and the interaction between insecticide concentrations and lady beetle species (F2, 2 = 8.02, p = 0.0027). Overall, mortality rates were low when exposed to dry thiamethoxam residues on filter paper, except for the 40% mortality observed for C. septempunctata exposed to 2X the field rate (Fig. 1A). The mortality of H. convergens was only 0% and 6.9% when exposed to thiamethoxam-treated filter paper with 1X and 2X the label rate (Fig. 1A) a mortality rate that was lower when to compared C. septempunctata. Mortality rates of C. septempunctata were higher at 15.6% and 40% when exposed to thiamethoxam-treated filter paper with 1X and 2X the label rate, respectively (Fig. 1A).

Fig. 1
figure 1

Mortality of field-collected lady beetle adults exposed to dry residues of thiamethoxam (A) and cyfluthrin (B) obtained on filter paper treated with 0X, 1X, and 2X the manufacturer-recommended label dose. Bars bearing different capital letters indicate differences in mortality as a function of insecticide concentrations for the same predator species, while small letters indicate difference between predator species for the same insecticide concentration. All tests were conducted at α = 0.05

Full size image

When adult H. convergens and C. septempunctata were exposed to dry residues of cyfluthrin on filter paper at 1X and 2X label dose rates, significant effects of insecticide concentration (F2, 2 = 37.72, p < 0.0001), lady beetle species (F1, 2 = 10.92, p = 0.0029), and interaction of these factors (F2, 2 = 5.3, p = 0.0123) on mortality rates were observed. In comparison to thiamethoxam, higher toxicity was observed for both coccinellid species with increasing cyfluthrin concentration, with an average mortality of 48.3% and 70% for H. convergens, and 21.9% and 40.9% for C. septempunctata, at 1X and 2X concentrations, respectively (Fig. 1B). Additionally, cyfluthrin residues were significantly more toxic to H. convergens than C. septempunctata at both tested concentrations (Fig. 1B).

Evaluating three different dry residue methods for toxicity (bioassay #2)

The exposure of adult H. convergens and N. viridula to four neonicotinoid insecticides (i.e., imidacloprid, thiamethoxam, acetamiprid, and dinotefuran) and the pyrethroid, bifenthrin, resulted in analogous mortality rates for both test species across test insecticides and dried residues (H. convergens: F4, 19 = 71.72, p < 0.0001, Fig. 2; and N. viridula: F4, 19 = 131.72, p < 0.0001, Fig. 3).

Fig. 2
figure 2

Mortality of adult Hippodamia convergens exposed to dry insecticide residues of four neonicotinoids and the synthetic pyrethroid, bifenthrin, via bioassays using insecticide dipped leaves, inert glass plates (i.e., bottoms of glass Petri dishes), and filter paper disks as dry insecticide residue substrates. A significant interaction effect of insecticide vs exposure method was detected (F12, 19 = 91.74, p < 0.0001), and uppercase letters compare the same exposure method among insecticides, while lower case letters compare exposure methods for the same insecticide. Different letters of the same case indicate significant differences at α = 0.05

Full size image
Fig. 3
figure 3

Mortality of adult Nezara viridula exposed to dry insecticide residues of four neonicotinoids, and the synthetic pyrethroid, bifenthrin, via bioassays using insecticide dipped leaves, glass plates (i.e., bottoms of glass Petri dishes), and filter paper disks as residue substrates. A significant interaction effect of insecticide vs exposure method was detected (F12,19 = 130.89, p < 0.0001), and uppercase letters compare the same exposure method among insecticides, while lowercase letters compare exposure methods for the same insecticide. Different letters of the same case indicate significant differences at α = 0.05

Full size image

A high level of mortality (near 100%) for adults of both species with bifenthrin was observed using all three methods of obtaining insecticide dry residues: leaf dip, inert glass surface of the Petri dish, and treated filter paper. Likewise, near 100% mortality for adult H. convergens and N. viridula was observed when insects were exposed to dry neonicotinoid residues on leaf discs and inert glass surfaces indicating a high significance among tested methods (H. convergens: F3, 19 = 2199.33, p < 0.0001; N. viridula: F3, 19 = 2884.43, p < 0.0001). Conversely, dry residues obtained on filter paper treated with each of the four tested neonicotinoids caused < 10% mortality of H. convergens (Fig. 2) and zero mortality of N. viridula adults (Fig. 3) resulting in a significant interaction effect of insecticides and exposure methods for H. convergens (F12, 19 = 91.74, p < 0.0001) and for N. viridula (F12, 19 = 130.89, p < 0.0001). Furthermore, mortality rates for test insects exposed to dry neonicotinoid residues on filter paper did not differ to those observed in the control groups for both species (Figs. 2 and 3).

Determining if dried neonicotinoid residues on filter paper have bioactivity (bioassay #3)

Dried residues of acetamiprid and thiamethoxam on the upper and lower surface of filter paper disks failed to exhibit toxicity against adult N. viridula (Table 2). The residues obtained from acetamiprid and thiamethoxam on the bottom of the glass Petri dish underneath treated filter paper caused mortality of test insects that averaged, 26.6% and 50% of adult N. viridula, respectively. This result differed significantly from untreated control groups (mean mortality 8.8% for acetamiprid; F = 97.72, df = 6, p < 0.0001; and 2.8% for thiamethoxam; F = 42.58, df = 6, p < 0.0001) (Table 2). The addition of the surfactant Tween 20 to the insecticide solutions increased the residual activity of test insecticides on the upper surface of the filter paper. However, observed mortality rates were still low at 28.9% and 38.1% for thiamethoxam and acetamiprid, respectively (Table 2).

Table 2 Mortality (% ± SE) of adult Nezara viridula, exposed to acetamiprid or thiamethoxam dry residues on filter paper and the bottom of glass plates
Full size table

The dry residues of neonicotinoids, acetamiprid, and thiamethoxam, on the inert glass surface of the Petri dish either applied directly or as dried reinstate obtained from treated filter paper, were highly active, resulting in 100% mortality of test insects (Table 2).

Discussion

The low mortality rates of field-collected adult H. convergens and C. septempunctata and colony reared adult N. viridula that were exposed to dry neonicotinoid residues on filter paper was an unexpected finding as it was anticipated that these insecticides would cause high levels of mortality of test insects. When applied to filter paper, it appears that the active ingredients adsorbed to the filter paper as exposure of the top and underside of insecticide-treated filter paper, without the addition of the adjuvant Tween 20, did not result in significant mortality of test insects. Adding Tween 20 increased partially the efficacy of applications to filter paper, possibly by enabling active ingredients to remain exposed on the treated surface of the filter paper. However, the addition of an adjuvant did not completely ameliorate reduced toxicity of test insecticides when applied to filter paper. Insecticide that managed to leach through the filter paper and dried on the glass bottom of Petri dishes exhibited toxicity towards N. viridula attaining 26.6% and 42.8% mortality with acetamiprid and thiamethoxam, respectively. In strong contrast, when insecticide solutions or rinsate from soaking treated filter paper in water were applied to the bottoms of glass Petri dishes and left to dry, insecticide residue was present to kill 100% of test insects.

The residual control and non-target effects of neonicotinoid insecticides for target pests and natural enemies from foliar application of neonicotinoids have been assessed in the laboratory using dry insecticide residues on different substrates, including plant foliage (Torres and Ruberson 2004; Chen et al. 2023), inert surfaces like glass (Barros et al. 2018; Řezáč et al. 2019), and filter paper (Magalhães et al. 2007; Dagg et al. 2019; Ribeiro et al. 2021). However, from results presented here, it is clear that evaluating the efficacy of dried neonicotinoid residues on filter paper significantly reduces mortality rates of test insects, which may result in incorrect conclusions about the bioactivity of tested active ingredients. The likely retention of acetamiprid and thiamethoxam within the filter paper suggests that active ingredients, when dry, adsorbed to fibers within the filter paper. As demonstrated, these compounds can be solubilized in water, and once removed from the filter paper, the dried residues are toxic to test insects. Whatman® filter papers, the filter paper substrate used in our experiments, are manufactured with an α-cellulose content > 98% (SigmaAldrich®). Interestingly, nanoformulations of thiamethoxam have been made using cellulose nanocrystals as the carrier due to the affinity of this active ingredient with cellulose. This cellulose nanocrystal formulation of thiamethoxam results in enhanced performance against target pests when compared to non-cellulose nanocrystal commercial formulations (Elabasy et al. 2020). Neonicotinoids bound to cellulose fibers within a paper substrate have potential as a novel approach for delivering insecticides to cellulose-consuming insect pests like termites, which could potentially reduce exposure of toxic compounds to the environment and non-target species (Acda 2007; Rasib and Wright 2018).

The efficacy of different classes of insecticides have been routinely tested by evaluating the toxicity of dry insecticide residues on filter paper toward target insect species. The dried residue-filter paper method worked well for the two pyrethroids, cyfluthrin and bifenthrin, that were evaluated in this study. Additionally, this filter paper approach has been used to evaluate the toxicity of other types of pyrethroids, in addition to organophosphates, carbamates, spinosyns, and other classes of insecticide (Haverty and Wood 1981; Haverty and Dell 1984; Santolamazza-Carbone and Fernández de Ana-Magán 2004; Cloyd and Dickinson 2006; Hussain et al. 2015; Ribeiro et al. 2021). For neonicotinoids, however, the results presented here strongly indicate that dried residues are not retained on the surface of filter paper and likely soak into the paper and when dry bind to paper fibers (i.e., cellulose). Literature data support neonicotinoid cellulose adsorption process as spontaneous and endothermic in nature (Yuan et al. 2019). Consequently, it is recommended that efficacy of dried neonicotinoid residues towards target test insects be tested on green leaves or inert surfaces and filter paper bioassays should not be used.