An update of the Worldwide Integrated Assessment (WIA) on systemic insecticides. Part 2: impacts on organisms and ecosystems

Honeybees (Apis mellifera)

Since the publication of the WIA document on the effects of neonicotinoid insecticides and fipronil on non-target invertebrates, research on this matter has continued. Lundin et al. (2015) provide a systematic review of research approaches, evaluating 268 publications on bees in general (honeybees, bumblebees, solitary bees). Another overview of scientific advances in the field of neonicotinoids and pollinators was made by Godfray et al. (2015). Van der Sluijs and Vaage (2016) reviewed the implications of the present pollinator crisis for global food security and concluded that it threatens global and local food security, can worsen the problems of hidden hunger, erodes ecosystem resilience, and can destabilize ecosystems that form our life support system. They call for an international treaty for global pollinator stewardship that simultaneously addresses its key drivers: creation and restoration of floral and nesting resources, a global phase out of prophylactic use of neonicotinoids and fipronil, improvement of test protocols in authorization of agrochemicals (see Sánchez-Bayo and Tennekes, 2017 for the changes that are needed), and restoration and maintenance of independence in regulatory science.

In the paragraphs below, results of recent studies with regard to honeybees (Apis mellifera) are listed, considering effects in vivo (field and semi-field situation) and in vitro (laboratory experiments).

Sublethal effects on memory, behavior, and locomotion

Karahan et al. (2015) found that feeding honey bees with field-realistic doses of imidacloprid (0.36 to 7.20 ng/bee) negatively affected the number of foraging trips, number of foragers returning, and flowers visited. Roat et al. (2014) found changes in the brain proteome of Africanized honeybees for doses of 10 pg fipronil per day during 5 days. Concentrations of several brain proteins involved in detoxification, glycolysis, and cell growth were altered, possibly leading to memory and learning impairment and to a reduced life span. Zaluski et al. (2015) also used Africanized honeybees in their study about effects of fipronil on colony development and bee motoric control and behavior. Treated adult bees (1/500th of the LD50) bees showed reduced motor activity and became lethargic, while treated colonies showed a reduction in egg laying and larval numbers.

Tan et al. (2015) investigated the effect of imidacloprid on adult bee memory and learning behavior by feeding total doses of 0.24 ng to larvae of A. cerana. They found that long-term memory, but not short-term or larval survival, was affected by the treatment. Also using Apis cerana in an earlier study, these authors found that trained exposed bees foraged less and had a lower avoidance of predators (i.e., Asian hornet Vespa velutina) (Tan et al. 2014). Wright et al. (2015) used a choice assay with imidacloprid and thiamethoxam influencing olfactory memory. They found that low acute doses affect olfactory memory negatively, with this effect being greater than the effect on memory. Doses of imidacloprid (11.25 ng/bee), clothianidin (2.5 ng/bee), and thiacloprid (1.25 mg/bee) given to trained forager bees resulted in less successful returns and a lower ability to navigate (Fischer et al. 2014). Effects on learning and memory were also found by Mengoni Goñalons and Farina (2015) who fed sublethal doses of imidacloprid to young bees. They postulate that impaired memory and sensitivity to rewards of individual bees affects colony performance.

Peng and Yang (2016) found a reduction of mushroom bodies in parts of the brain responsible for olfactory and visual processing. At the molecular level, interaction between odor binding proteins and imidacloprid has been studied in A. cerana by Li et al. (2015a), who found that presence of imidacloprid decreased the affinity of a specific odor binding protein and a flower volatile.

A 24-h exposure of adult bees to imidacloprid, dinotefuran, clothianidin, and thiamethoxam at sublethal field-realistic doses (0.323 to 0.481 ng/bee) resulted in behavioral changes. Bees walked less and groomed more (Williamson et al. 2014). Blanken et al. (2015) used flight cages to determine effects of imidacloprid (about 6 ng/mL, weekly feeding of 660 mL in a 13-week period) on flight capacity of forager bees, in combination with differential Varroa destructor mite loads of the bee donor colonies. Their results showed an interaction between physiological stressed caused by Varroa and imidacloprid, with imidacloprid possibly affecting the body mass of bees and lower body mass causing decreased flight capacity. An interesting finding is that of Kessler et al. (2015). Their data generated by choice assays (sucrose laced with imidacloprid or thiamethoxam versus plain sucrose) suggest that bees prefer solutions with imidacloprid and thiamethoxam. Another study investigating effects on food consumption found that thiamethoxam decreased bees’ response to higher sucrose concentrations (Démares et al. 2016). Alkassab and Kirchner (2016) exposed winter bees to sublethal doses of clothianidin and measured behavioral effects. Chronic exposure to 15 ppb was found to significantly affect long-term memory. Both deltamethrin and acetamiprid were used in retrieval assays (conditional proboscis response) by Thany et al. (2015). Their results showed that retrieval was impaired at lower doses of acetamiprid compared to deltamethrin.

Papach et al. (2017) present the first evidence of impaired learning and memory in adult bees that were fed thiamethoxam (0.6 ng/bee) during the larval stage. Colony survival critically depends on successful learning and memory. Chronic larval exposure to sublethal doses of this neonicotinoid resulted in alterations of associative behavior in adults. Similar delayed effects on learning and memory following larval exposure have been reported for other neonicotinoids such as imidacloprid (these studies are reported in the WIA 2015 study).

Effects of sublethal doses of thiacloprid on social interactions and network structure established by a group of honeybee worker individuals has been quantified in a study by Forfert and Moritz (2017) using experimental groups. Bees fed with thiacloprid (0.17 and 0.80 μg thiacloprid in 20 μL 2.7 M sucrose solution) significantly reduced their network centrality, but they nevertheless exchanged more food to other group members, which resulted in a dilution of the contaminated food. The authors argue that although thiacloprid may act as a general perturbator of social network structure, it still may play a role in the dynamics of disease transmission in the colony if pathogens are transmitted via food exchange.

Using flight mills, Tosi et al. (2017) found that flight activity (duration and distance) was increased after a single sublethal dose and decreased (duration, distance, velocity) after 1–2 days of chronic exposure.

To understand how neonicotinoids affect behavior and immunity at the molecular level, Christen et al. (2016) looked at transcriptional regulation of 8 genes in caged honeybees fed with field-realistic concentrations of acetamiprid, clothianidin, imidacloprid, and thiamethoxam. They found downregulation of transcription of two genes involved in memory and increased transcription of the gene responsible for vitellogenin, the latter possibly affecting foraging behavior. A follow-up study confirmed these results and looking at effects of binary mixtures of acetamiprid, clothianidin, imidacloprid, and thiamethoxam on memory and vitellogenin gene transcription, found smaller effects of mixtures opposed to single substance application on gene regulation (Christen et al. 2017).

Sublethal effects on immunity and metabolism

Gene expression profiles in honeybee midgut showed that insecticide treatments (imidacloprid or fipronil) had no impact on detoxifying genes but led to a significant downregulation of immunity-related genes, suggesting a possible immunotoxicity of neonicotinoid and phenylpyrazole insecticides under chronic exposure (Aufauvre et al. 2014). This study also showed that N. ceranae + fipronil and N. ceranae + imidacloprid combinations do not systematically lead to a synergistic effect on honeybee mortality. Brandt et al. (2016) found that imidacloprid, thiacloprid, and clothianidin caused reduced hemocyte density, encapsulation response, and antimicrobial activity after a relatively short exposure (24 h) to field-realistic concentrations. Looking specifically at the interaction of thiacloprid and the pathogens Nosema ceranae and black queen cell virus, Doublet et al. (2014) found that thiacloprid increased the viral load of larvae and so negatively affected larval survival, as well as aggravating the effect of Nosema on adult mortality. A similar study by Gregorc et al. (2016) combined exposure to Nosema ceranae and thiamethoxam and showed no synergistic effects of the two. Reviews of the relation between nicotinoid pesticides and honeybee disease can be found in Sánchez-Bayo and Desneux (2015) and Sánchez-Bayo et al. (2016b).

Badawy et al. (2015) measured the effects of oral and topical application of four pesticides (acetamiprid, dinotefuran, pymetrozine, pyridalyl) on detoxifying enzyme activity (acetylcholinesterase, carboxylesterase, glutathione-S-transferase and polyphenol oxidase). They found dinotefuran to be the most toxic, pyridalyl second and acetamiprid/pymetrozine the least toxic. Carboxylesterase and glutathione-S-transferase were able to detoxify low doses of acetamiprid, pymetrozine, and pyridalyl but not dinotefuran. Böhme et al. (2017) were feeding pollen containing mixtures of pesticides at field-realistic (sublethal) doses to determine synergistic effects, as exposure to multiple substances through pollen is common but little studied. They found that larval weight was higher and acini diameters of the hypopharyngeal glands of nurse bees were smaller in the experimental group. Renzi et al. (2016) also looked at hypopharyngeal glands and found that dietary exposure to thiamethoxam was associated with smaller acini and lower total protein content of bee heads.

Exposure to thiamethoxam was also found to alter thermoregulation in individual bees, with effects dependent on ambient temperature and dose (Tosi et al. 2016). At higher temperatures (33 °C), body temperature of exposed bees increased, whereas lower temperatures (22 °C) lead to lower body temperature 60–90 min post treatment. In both exposed groups, body temperatures were lower than control group the following day.

An interesting finding was done by Rittschof et al. (2015), who investigated aggressive behavior of honeybees as a result of early-life social experience, using acetamiprid as a stressor to identify effects on the immune system. Their results found that aggressive bees had less immunosuppressive effects of acetamiprid than less aggressive bees.

Sublethal effects on reproduction

Sublethal effects on honeybee reproduction were not mentioned in the original WIA article on invertebrates (Pisa et al. 2015) but might be of considerable importance, as specific effects on, for example, sperm viability and queen mating success might directly affect population numbers. Williams et al. (2015) found that queens exposed to clothianidin and thiamethoxam had larger ovaries and reduced quality and quantity of sperm stored in the spermatheca. Very low doses of imidacloprid, alone and in combination with the parasite Nosema ceranae, were found to increase activity of detoxifying enzymes and decrease survival of queens (Dussaubat et al. 2016).

Drones that were raised in semi-field and laboratory conditions and exposed to fipronil through feeding showed a decrease in quantity of spermatozoa and increased mortality of spermatozoa (Kairo et al. 2017). This confirmed earlier research by the same authors had shown that queens inseminated with sperm of fipronil exposed drones had less and less viable spermatozoa stored in their spermatheca (Kairo et al. 2016). They found that several pesticides, among them fipronil, imidacloprid, and thiamethoxam, reduced sperm viability (in vitro sperm assay). Effects on drones were also found by Straub et al. (2016), who reported reduced drone life span as well as decreased sperm quality (low quantity of spermatozoa, reduced viability by 40%). Number of newly emerged adults and drone body mass was unaffected. Sublethal dose of imidacloprid (2 ppb) decreased also sperm viability by 50% 7 days after treatment in another study (Chaimanee et al. 2016).

Wu-Smart and Spivak (2016) fed small (1500–7000 bees) colonies with different doses of imidacloprid (0, 10, 20, 50, and 200 ppb) in syrup for 3 weeks to investigate its effect on queen productivity. They observed a decrease in egg laying rate and queen motility associated with exposure, as well as negative effects on foraging, hygienic behavior of worker bees, and on colony development in all treated colonies. Independent of colony size, number of eggs laid per 15 min was reduced by approximately 50% by 10 ppb imidacloprid compared to control. These findings demonstrate that chemical exposure may affect sperm quality in the spermatheca of honey bee queens, queen fecundity, threatening the reproductive success and survival of the colony.

An interesting study on honeybee reproductive metabolism was done by Wessler et al. (2016). They looked at the effect of thiacloprid and clothianidin on the secretion of acetylcholine by the hypopharyngeal gland. Acetylcholine is a key compound of larval food and royal jelly. Release of acetylcholine and its presence in larval food decreased by 80% after 4 weeks of exposure to high doses of both neonicotinoids. Field-realistic doses (200 ppb for thiacloprid, 1 to 10 ppb for clothianidin) lowered acetylcholine in larval food and showed negative effects on brood development.

Sublethal effects due to ontogenic exposure

Residue analyses of pollen, honey or bee wax revealed the presence of a cocktail of multiple insecticides accumulating at the same time (Bonmatin et al. 2015; David et al. 2016; Krupke and Long 2015; Mullin et al. 2010; Daniele et al. 2017; Giorio et al. 2017 this special issue). However, relatively few investigations have focused on the sublethal effects of pesticides on the honeybee brood.

It has been clearly shown that rearing brood in contaminated combs causes delayed development of larvae and emergence as well as a shortened adult life span (Wu et al. 2011). An additive interaction between black queen cell virus (BQCV) and thiacloprid on host larval survival was also observed (Doublet et al. 2014). A recent study by López et al. (2017) demonstrated a synergistic interaction when larvae are exposed to sublethal doses of dimethoate or clothianidin in combination with Paenibacillus larvae, the causative agent of American foulbrood (AFB). It is evident that the cellular response of larvae to individual and combined stressors allows for unmasking previously undetected sublethal effects of pesticides on colony health (Giorio et al. 2017, this special issue).

Bee larvae that were fed sublethal doses of thiamethoxam by Tavares et al. (2015a) showed condensed cells and early cell death in the optical lobe part of the brain, as well as dose-dependent effects on development speed and body size.

By exposing a hive to imidacloprid, Yang et al. (2012) discovered that honey bee larvae fed with a sublethal dose of imidacloprid still completed their development into adult bees, but they did so with a decreased olfactory learning ability. This impairment occurred with a dose that could be as little as 0.04 ng per larva. These results demonstrate that sublethal dosages of imidacloprid given to the larvae affect the subsequent associative ability of the adult honeybee workers. Peng and Yang (2016) further revealed the effect of sublethal doses of imidacloprid on the neural development of the honeybee brain by immune-labeling synaptic units in the calyces of mushroom bodies. This not only links a decrease in olfactory learning ability to abnormal neural connectivity but also provides evidence that imidacloprid damages the development of the nervous system in regions responsible for both olfaction and vision during the larval stage of the honeybee.

To reveal the potential spectrum of sublethal effects of imidacloprid exposure in the larval stage, Wu et al. (2017) measured changes in global gene expression in the heads of newly emerged adults. They found that multiple physiological changes could be induced by the sublethal exposure to imidacloprid, affecting detoxification, immunity, sensory processing, neuron development, metabolism, mitochondria, and synthesis of royal jelly.

Direct lethality of neonicotinoids to wild bees

Around 2000 bee species are known from Europe, with 400 of these classified as endemic (Nieto et al. 2014). The biology, behavior, and ecology of each of these species differ from those of honeybees, for example, some bees ingest pollen for transport (e.g., Hylaeus sp.), which might provide much greater exposure than carrying pollen in corbiculae. Consequently, extrapolating from the limited toxicological data available for 19 bee species to the effects of neonicotinoids on the wider European fauna is fraught with difficulties given the wide variation in relative sensitivity, ecology, and behavioral traits. Conversely to the results of Cresswell et al. (2012) who exposed bumble bees and honey bees to high doses, current data suggests that wild bees are equally to slightly less sensitive to neonicotinoids compared to honeybees when considering direct mortality (e.g., Sánchez-Bayo et al. 2017). However, care must be taken when considering individual bee species, genera, and families, as different taxonomic groups may show consistently different individual-level sensitivity. Most European wild bees are smaller than honeybees and there is the potential for them to be more sensitive on a basis of a few ng/bee exposure. In general, continuing to use honeybee neonicotinoid sensitivity metrics is likely to be a reasonable proxy measure for the direct sensitivity of the wild bee community to neonicotinoids (Arena and Sgolastra 2014), but further work is needed in this area to cover the wide range of bee species present in agricultural environments.

In large parts of Asia, the ecological niche of Apis mellifera is occupied by the similar but distinct sister species A. cerana. As agriculture has intensified and pesticide use increased strongly, effects can be expected on this bee species but little toxicological research has been conducted so far. The study of Yasuda et al. (2017) addresses this knowledge gap. They used the subspecies A. cerana japonica to determine LD50 values for acute contact toxicity for commonly used pesticides. Of the neonicotinoid group, dinotefuran proved to be most toxic (1.4 ng/bee), followed by thiamethoxam (2.4 ng/bee), clothianidin (3.4 ng/bee), imidacloprid (3.6 ng/bee), and acetamiprid (278 ng/bee). This LD50 for fipronil was determined at 2.5 ng/bee. The authors note that A. cerana is generally more sensitive to pesticides and that results obtained for A. mellifera cannot be generalized to A. cerana.

Arena and Sgolastra (2014) conducted a meta-analysis comparing the sensitivity of wild bees to pesticides relative to the sensitivity of honeybees. This analysis combined data from 47 studies covering 53 pesticides from six chemical families with a total of 150 case studies covering 18 bee species (plus A. mellifera). The authors calculated a sensitivity ratio (R) between the lethal dose for species a (A. mellifera) and for species s (other than A. mellifera), where R = LD50a/LD50s. A ratio of over 1 indicates that the other bee species is more sensitive to the selected pesticides than A. mellifera and vice versa. There was high variability in relative sensitivity ranging from 0.001 to 2085.7, but across all pesticides a median sensitivity of 0.57 was calculated, suggesting that A. mellifera was generally about two times more sensitive to pesticides than other bee species. In the vast majority of cases (95%), the sensitivity ratio was below 10.

Combining data for all neonicotinoids (acetamiprid, imidacloprid, thiacloprid, and thiamethoxam) and for both acute contact and acute oral toxicity, nine studies covering nine bee species (plus A. mellifera) were found. These studies showed a median sensitivity ratio of 1.045 which is the highest median value of all the analyzed pesticide chemical families. The most relatively toxic neonicotinoids to other bees were the cyano-substituted neonicotinoids acetamiprid and thiacloprid as these pesticides exhibit lower toxicity to honeybees than the nitro-substituted neonicotinoids imidacloprid and thiamethoxam.

In 2013, the EU installed a partial ban on imidacloprid, clothianidin, thiamethoxam, and fipronil while allowing continued use of acetamiprid and thiacloprid. Searching for studies about effects of the banned compounds including both acute contact and acute oral toxicity, 12 studies covering 10 bee species (plus A. mellifera) were found. These studies showed a median sensitivity ratio of 0.957 which is close to the calculated sensitivity ratio for all neonicotinoids. Studies on Bombus terrestris consistently report a lower sensitivity ratio between 0.005 and 0.914, median 0.264. Bombus terrestris is widespread in Europe and is the most commonly used non-Apis model system for assessing the effects of neonicotinoids on wild bees. Differences in bee body weight have been proposed to explain these differences, with sensitivity to pesticides inversely correlated with body size (Devillers et al. 2003). However, this has not been consistently demonstrated and other mechanisms have been suggested such as species-level adaptation to feeding on alkaloid-rich nectar (Cresswell et al. 2012). With the limited data available, Arena and Sgolastra (2014) could not comment on the strength of these claims and further experiments are needed.

Spurgeon et al. (2016) calculated various toxicity measures of clothianidin on honeybees, the bumblebee species B. terrestris and the solitary bee species Osmia bicornis. Acute oral toxicity 48, 96, and 240 h LD50s for honeybees were 14.6, 15.4, and 11.7 ng/bee, respectively. For B. terrestris, the corresponding values were 26.6, 35, and 57.4 ng/bee, respectively. For O. bicornis, the corresponding values were 8.4, 12.4, and 28.0 ng/bee, respectively. These findings are generally in line with the findings of Arena and Sgolastra (2014), with B. terrestris less sensitive than A. mellifera at all time points and O. bicornis less sensitive at 240 h.

Sgolastra et al. (2017) calculated relative sensitivity to clothianidin to these same three species over a range of time periods from 24 to 96 h. The highest LD50 values were obtained after 24 h for A. mellifera and B. terrestris and after 72 h for O. bicornis. At these time points, O. bicornis was the most sensitive of the three species, with LD50 measurements of 1.17 ng/bee and 9.47 ng/g, compared to 1.68 ng/bee and 19.08 ng/g for A. mellifera and 3.12 ng/bee and 11.90 ng/g for B. terrestris. These results are in line with the values calculated by Spurgeon et al. (except for the 240 h values), with decreasing sensitivity in the order of O. bicornis > A. mellifera > B. terrestris. Together, these studies support the position that small bodied species show greater sensitivity to neonicotinoids.

Czerwinski and Sadd (2017) found detrimental interactions of imidacloprid exposure and bumblebee immunity. Adult workers of Bombus impatiens received 6-day pulses of either low (0.7 ppb) or high (7 ppb) field-realistic doses of imidacloprid. This was followed by an assay to test immunity and survival following a nonpathogenic immune challenge. The results showed that high-dose imidacloprid exposure reduces constitutive levels of phenoloxidase, an enzyme involved in melanization. Hemolymph antimicrobial activity initially increases in all groups following an immune challenge, but while heightened activity is maintained in unexposed and low imidacloprid dose groups, it is not maintained in the high exposure dose bees, although exposure had ceased 6 days prior. When imidacloprid exposure was followed by an immune challenge, a significantly decreased in survival probability was observed relative to control bees and those only immune challenged or imidacloprid exposed. A temporal lag for immune modulation and combinatorial effects on survival suggest that resource-based trade-offs may, in part, contribute to the detrimental interactions. These findings are particularly relevant because such impairment of the immune system at field-realistic exposure to neonicotinoids is likely to have health consequences for pollinators that in real life often face multiple stresses of sublethal neonicotinoid exposure and pathogens. It also raises a broader question whether impairment of the immune system by neonicotinoids is limited to insects or whether it also affects other non-target species that are exposed.

Baron et al. (2017) provides the first evidence of impacts of thiamethoxam on the ovary development and feeding of spring-caught wild queens of four bumblebee species: Bombus terrestris, B. lucorum, B. pratorum, and B. pascuorum. In a laboratory experiment testing the impacts of field relevant doses (1.87–5.32 ppb) of thimethoxam, they found that 2 weeks of exposure to the higher concentration of thiamethoxam caused a reduction in feeding in two out of four species, suggesting species-specific anti-feedant, repellency, or toxicity effects. The higher level of thiamethoxam exposure resulted in a reduction in the average length of terminal oocytes in queens of all four species. Further, the authors highlight that the discovery of species-specific effects on feeding has significant implications for current practices and policy for pesticide risk assessment and use.

Stingless bees (Apidae: Meliponini) are pan-tropical eusocial bees that are important pollinators for wild plants and crops (Barbosa et al. 2015). Little research on exposure and toxicology has been done for this diverse and abundant clade that is under pressure of habitat loss and intensification of agriculture. Lima et al. (2016) provide an overview of general agrochemical stressors on stingless bees.

Of the available studies involving neonicotinoids or fipronil, several indicate that the species studied are more sensitive to certain pesticides than A. mellifera and that results and testing procedures cannot be generalized. Topical LD50 (2.41 ng/bee 24 h, 1.29 ng/bee 48 h) and oral LC50 (2.01 ng/μL 24 h, 0.81 ng/μL 48 h) values for Melipona scutellaris for imidacloprid were lower than those of A. mellifera (Costa et al. 2015). Lourenco et al. (2012) found that for fipronil topical LD50 (0.6 ng/bee 48 h) and oral LC50 (0.011 ng/μL 48 h) were also lower than that of the honeybee. Rosa et al. (2016) found decreased larval survival feeding field-realistic doses (0.004 to 4.375 ng/larva) of thiamethoxam to Scaptotrigona depilis larvae in vitro. Low doses of fipronil (0.27 ng/bee topical, 0.24 ng/bee oral) affected brain morphology by apoptosis or necrosis of mushroom bodies of Scaptotrigona postica (Jacob et al. 2015), comparable to its effect on mushroom bodies of A. mellifera (Roat et al. 2014). Tomé et al. (2012) also found effects of imidacloprid on mushroom bodies and behavior in Melipona quadrafasciata and showed that imidacloprid impaired respiration and flight activity in this species. Valdovinos-Núñez et al. (2009) compared the toxicity of different pesticides for three stingless bee species (Melipona beechei, Trigona nigra, Nannotrigona perilampoides) and found neonicotinoids (imidacloprid, thiamethoxam, and thiacloprid) to be more toxic than permethrin and diazinon.

Synergistic effects of additional pesticides with neonicotinoids

Sgolastra et al. (2017) investigated the interaction between clothianidin and the ergosterol biosynthesis inhibiting (EBI) fungicide propiconazole in three bee species, A. mellifera, B. terrestris, and O. bicornis. Each species was administered a LD10 dose of clothianidin (0.86, 1.87, and 0.66 ng/bee, respectively, a non-lethal dose of propiconazole (7 μg/bee) and a combination of the two treatments. Bees were then observed for a 96-h period and mortality quantified. Some synergistic effects were recorded. In A. mellifera, mortality was significantly higher for the combined dose in the first two time periods (4 and 24 h). Mortality in B. terrestris for the combined dose was only significantly higher in the first time period, after 4 h. However, in O. bicornis, exposure to the combination of clothianidin and propiconazole resulted in significantly higher mortality at all time points.

Spurgeon et al. (2016) conducted similar experiments to Sgolastra et al., investigating the effect of a combination of clothianidin and propiconazole on A. mellifera, B. terrestris, and O. bicornis. In order to calculate an LD50, clothianidin concentrations were varied and propiconazole concentrations were held at zero, a low dose and a high dose. The low dose was taken from the EFSA Panel on Plant Protection Products (EFSA 2012) reported environmental concentrations, and the high dose was 10 times the low dose to represent a plausible worst-case scenario. Mortality was quantified over 48, 96, and 240 h. For A. mellifera, clothianidin LD50s with and without propiconazole were always within a factor of 2, with no clear negative trend at higher propiconazole concentrations. For B. terrestris, clothianidin LD50s with propiconazole were between 1.5- to 2-fold lower. For O. bicornis, clothianidin LD50s with propiconazole was up to 2-fold lower with a negative trend as propiconazole concentrations increased. Spurgeon et al. concluded that the clothianidin and propiconazole combination had no to slight synergy for A. mellifera and slight to moderate synergy for B. terrestris and O. bicornis.

In an additional trial, Thompson et al. (2014) demonstrated that the dose of fungicide applied is a key factor determining neonicotinoid toxicity using propiconazole and thiamethoxam mixtures. The authors argue that their low rates of significant synergies between neonicotinoids and fungicides was because of their lower, more field-realistic fungicide doses of 161–447 ng/bee compared to 10,000 ng/bee used by Iwasa et al. (2004), an early study demonstrating this interaction. The values of 161–447 ng/bee were calculated as realistic worst-case exposures based on approved application rates for UK crops. In a study of pesticide residues in pollen collected by B. terrestris in the UK, David et al. (2016) found concentrations of DMI fungicides up to 84 ppb, while Sanchez-Bayo and Goka (2014) report residues of propiconazole in honeybee-collected pollen up to 361 ppb. At the latter concentration a bee would need to consume approximately 28 g of pollen to receive the dose used in the Iwasa et al. (2004) study, which is not realistic. However, data are lacking demonstrating true field-realistic exposure rates to fungicides for free flying bees.

Overall, these studies support the position that neonicotinoids can act synergistically with fungicides, increasing their lethality to bees. However, the dose rate of both neonicotinoids and fungicides, time of exposure, neonicotinoid and fungicide chemical class, and length of time after exposure are all important explanatory factors affecting this relationship. The concentration of fungicide used in laboratory studies appears to be the most important factor determining synergistic lethality. Fungicides are regularly sprayed during the period when flowering crops are in bloom under the assumption that these compounds are safe for bees. Further work is needed in this area to establish realistic levels of chronic exposure to fungicides for free flying bees in order to assess the likely impact of neonicotinoid/fungicide synergies on bee populations.

Studies to date have only examined pairwise interactions between pesticides. It is clear that bees and other non-target organisms inhabiting farmland are routinely exposed to far more complex cocktails of pesticides than any experimental protocol has yet attempted to examine (e.g., David et al. 2016; Giorio et al. 2017 this special issue). A major challenge for scientists and regulators is to attempt to understand how chronic exposure to complex mixtures of neonicotinoids, fipronil, and other chemicals affects wildlife, this with or without other natural stressors (infectious agents, parasitism) and adverse abiotic conditions.

Population-level effects of neonicotinoids on wild bees

Nothing was known about the population-level effects of neonicotinoids on wild bees in 2014. As a managed domesticated species, population trend data are available for honeybees, but not for wild bees. One study has attempted to investigate the impact of neonicotinoids on wild bee population trends. Woodcock et al. (2016) used an incidence dataset of wild bee presence in 10 × 10 km grid squares across the UK. The dataset is comprised of bee sightings by amateur and professional entomologists and is probably the most complete national bee distribution database currently available. Sixty-two wild bee species were selected and their geographic distance and persistence over an 18-year period between 1994 and 2011 was calculated. Neonicotinoid seed-treated oilseed rape was first used in the UK in 2002, and so the authors calculated spatially and temporally explicit information describing the cover of oilseed rape and the area of this crop treated with neonicotinoids. The 62 species were split into two groups—species that foraged on oilseed rape (n = 34) and species that did not (n = 28). Species persistence across this time period was then compared with expected neonicotinoid exposure. Over the 18-year period, wild bee species persistence was significantly negatively correlated with neonicotinoid exposure for both the foraging and non-foraging group, with the effect size three times larger for the oilseed rape foraging group. Overall, the study suggests that bee species were more likely to disappear from areas with a high exposure to neonicotinoids as measured by the amounts applied as seed dressings to oilseed rape and that this trend was more pronounced for species known to forage on oilseed rape. While more work is needed, this is a major correlational study that suggests a link between levels of neonicotinoid exposure and bee community persistence at a national scale.

Rundlöf et al. (2015) conducted an extensive field trial of the effects of clothianidin-treated oilseed rape on wild bees. Sixteen oilseed rape fields separated by at least 4 km were selected across southern Sweden and were paired on the basis of similar landscape composition. In each pair, one of the fields was randomly selected to be sown with oilseed rape treated with 10 g clothianidin/kg of seed and the other field was sown without a neonicotinoid seed treatment. Twenty-seven cocoons of the solitary bee O. bicornis (15 male, 12 female) were placed out alongside each field a week before the oilseed rape began to flower, and six colonies of B. terrestris were placed alongside each field on the day the oilseed rape began to flower. The O. bicornis placed adjacent to treated oilseed rape showed no nesting behavior and did not initiate brood cell construction. O. bicornis adjacent to untreated fields showed nesting behavior in six of the eight fields studied. Bumblebees placed next to treated oilseed rape showed reduced colony growth and reproductive output. Bumblebee colonies were collected and frozen when new queens began to emerge, with this happening between the 7th of July and 5th of August depending on each colony. The number of queen and worker/male cocoons present was counted. At the point of freezing, colonies placed next to treated oilseed rape fields had significantly fewer queen and worker/male cocoons present.

Sterk et al. (2016) performed a similar field experiment to Rundlöf et al. Two areas of 65 km² in northern Germany were selected in which the only flowering crops comprised of winter-sown oilseed rape. In one area, the oilseed rape was treated with the same seed coating used by Rundlöf et al. of 10 g clothianidin/kg seed. The other area was an untreated control. In each area, ten B. terrestris colonies were placed at each of six localities. Colonies were left adjacent to oilseed rape between April and June, covering its main flowering period. After this the colonies were moved to a nature reserve. No differences were found in colony weight growth, number of workers produced, or reproductive output as measured by the production of new queens.

It is interesting to note that the latter field studies, using the same neonicotinoid seed dressing, found markedly different results. The major difference is that while Rundlöf et al. used spring-sown oilseed rape, Sterk et al. used winter-sown oilseed rape. The length of time between sowing and peak flowering is much greater for winter-sown oilseed rape (mid-August to May) than for spring-sown oilseed rape (April/May to mid-June). As such, there is more time for neonicotinoids to degrade, and for them to leach into soil and water for winter-sown oilseed rape, reducing the amount of active ingredient available to be taken up by the crop. Indeed, the mean loads of clothianidin in the Rundlöf et al. study were 13.9 ppb in honeybee pollen, and 5.4 (bumblebee) and 10.3 (honeybee) ppb in nectars, whereas those in the German study were 0.50–0.97 ppb in honeybee pollen, 0.88 in bumblebee pollen, and 0.68–0.77 ppb in honeybee nectar (Rolke et al. 2016). Such a difference as revealed by exposure to the insecticide for honeybees (14–27 times less for pollen and 13–15 times less for nectar in the latter study) could explain the difference in reported colony growth and number of gynes and drones produced, since concentrations of clothianidin in the food of bees below 1 ppb are not supposed to produce any effect that were measured (Piiroinen et al. 2016). An additional difference is that in the Sterk et al. (2016) study, colonies were moved to a nature reserve consisting of forests, lakes, and heathland after the flowering period of oilseed rape ended. The quality of available foraging area at this nature reserve is likely to have been of both a higher quality and quantity than what was available in a conventional agricultural landscape and is not typical of the experience of a bumblebee colony located in such a landscape that will have to continue foraging there after crops such as oilseed rape cease flowering. In addition, a major problem with the experimental design of Sterk et al. is that only one treated and one control area were used, so there is no true site-level replication, as opposed to Rundlöf et al. who used eight treated and eight control fields. All these differences in experimental design highlight the difficulty of developing a single experimental design that may answer risk assessment questions for every potentially affected species. It also highlights the importance of evaluating the experimental design in terms of resulting data quality when considering the differences in results between Rundlöf et al. (2015) and Sterk et al. (2016).

Only one study is available that looked at the impact of neonicotinoids on the reproductive success of a solitary bee in controlled conditions. Sandrock et al. (2014a) established laboratory populations of O. bicornis, a solitary stem nesting bee. Bees were fed on sugar solution treated with 2.87 ppb thiamethoxam and 0.45 ppb clothianidin along with untreated pollen. There was no impact of neonicotinoids on adult female longevity or body weight. However, treated bees completed 22% fewer nests over the course of the experiment. Nests completed by treated bees contained 43.7% fewer total cells and relative offspring mortality was significantly higher, with mortality rates of 15 and 8.5% in the treated and untreated groups, respectively. Overall, chronic neonicotinoid exposure resulted in a significant reduction in offspring emergence per nest, with treated bees producing 47.7% fewer offspring. These results suggest that exposure to these low-level, field-realistic doses of neonicotinoids (< 3.5 ppb) did not increase adult mortality but did have sublethal impacts on their ability to successfully build nests and provision offspring.

Colony-level impact on bumblebees

Laycock et al. (2014) fed microcolonies of four B. terrestris workers thiamethoxam-treated sugar solution at a range of concentrations up to 98 ppb. Pollen was not treated with thiamethoxam. Sugar solution consumption was significantly reduced at the 39 and 98 ppb treatments. Worker mortality was only increased at the highest dose of 98 ppb. Worker oviposition failure was only significantly higher at the 39 and 98 ppb treatments, with no significant differences seen between the lower concentration treatments between 0 and 16 ppb.

Scholer and Krischik (2014) exposed greenhouse queenright colonies of Bombus impatiens to imidacloprid- and clothianidin-treated sugar syrup at concentrations of 0, 10, 20, 50, and 100 ppb for 11 weeks. Queen mortality was significantly increased at 6 weeks for the 50 and 100 ppb treatments and at 11 weeks for the 20 ppb treatment for both clothianidin and imidacloprid. Surprisingly, no significant impact was found on numbers of workers or new queens produced, though this was in part because very low numbers of new queens were produced across all treatments (average of four per colony). Colonies in treatments above 10 ppb imidacloprid and 20 ppb of clothianidin gained significantly less weight over the course of the study.

Cutler and Scott-Dupree (2014) placed B. impatiens colonies adjacent to maize fields during pollen shed in Ontario, Canada. Four neonicotinoid-treated conventional and four untreated organic fields were used. Colonies were placed adjacent to each field on the first day of major pollen shed. Colonies were left for 5–6 days and then transported to an area of semi-natural habitat for 30–35 days, after which they were frozen. Colonies placed next to treated maize produced significantly fewer workers than those placed next to organic farms. All other metrics (colony weight, honey and pollen pots, brood cells, worker weight, male and queen numbers and weights) were not significantly different. However, bumblebees collected less than 1% of their pollen from maize and neonicotinoid residues in collected pollen were very low (mean of 0.4 ppb) for bees foraging adjacent to treated fields and always below the LOD (0.1 ppb) for bees adjacent to organic fields. Given that it is well-known that bumblebees collect very low volumes of maize pollen, the relevance of this study is unclear.

FERA (2013) also conducted a field trial with B. terrestris colonies placed out adjacent to oilseed rape treated with either clothianidin, imidacloprid or an untreated control. Colonies were allowed to forage freely for 6–7 weeks while the oilseed rape flowered and then were moved to a non-agricultural area to continue developing. The initial aim was to measure colony growth and development across these three treatments and compare this with neonicotinoid concentrations collected from food stores within the nests, but the study was criticized for a number of methodological problems such as variable placement date and initial colony size, lack of site-level replication, and contamination of control colonies with neonicotinoid residues during the experiment. The study was ultimately not published in a peer reviewed journal but it came to the conclusion that there was no clear relationship between bumblebee colony success and neonicotinoid concentrations. Goulson (2015) reanalyzed the FERA data using linear models and retaining two colonies excluded in the original study as outliers, but which do not meet the statistical definition of this term. This reanalysis showed that the concentration of clothianidin in nectar (range 0 to 0.28 ppb) and the concentration of thiamethoxam in pollen (range 0 to 1.6 ppb) significantly and negatively predicted both colony weight gain and production of new queens. Very similar findings emerged from the recent large field trial of Woodcock et al. (2017) who exposed B. terrestris colonies to oilseed rape fields treated with either clothianidin, thiamethoxam or controls at field sites in the UK, Germany, and Hungary. Total neonicotinoid levels in the range 0 to 8 ppb in colony food stores negatively predicted colony reproductive output.

Most research on neonicotinoids and bees has focused on the three compounds subject to the EU moratorium. Thiacloprid is considered to be less dangerous to bees, since it has a much higher acute LD50. As a result, it is sometimes sprayed on crops or trees at or near flowering, potentially exposing bees to much higher doses than they would obtain from neonicotinoids applied as seed dressings. Ellis et al. (2017) placed B. terrestris colonies adjacent to raspberry crops that had been sprayed with thiacloprid following normal farming practice, and compared these to control nests placed next to unsprayed raspberry crops. Exposed colonies were more likely to die, grew more slowly, and produced 46% fewer reproductive than control colonies. This study strongly argues that thiacloprid should not be regarded as safe to bees.

Studies produced since 2014 have advanced our knowledge in several key areas. Laboratory studies have continued to demonstrate negative effects of neonicotinoids on bumblebee reproductive output at generally high concentrations, with the lowest sublethal effects on reproductive output detected at 10 ppb. Field studies using bumblebees demonstrate that exposure to neonicotinoid-treated flowering crops can have significant impacts on colony growth and reproductive output depending on the levels exposed to, with crop flowering date relative to sowing and availability of uncontaminated forage plants likely to explain variation in the detected residues among the available studies. Our understanding of the impact on solitary bees is much improved with the findings of Sandrock et al. (2014a) suggesting substantial impacts on solitary bee reproductive output at field-realistic concentrations of 3.5 ppb. Field studies demonstrating this under real-world conditions are limited with the work of Rundlöf et al. (2015) and Woodcock et al. (2017), demonstrating no nest-building activity at the neonicotinoid treatment sites.

Feltham et al. (2014) exposed B. terrestris colonies to sugar solution treated with 0.7 ppb and pollen treated with 6 ppb of imidacloprid for 2 weeks. Colonies were then placed out in an urban area in Scotland. The foraging workers from each nest were then monitored for a further 4 weeks. There was no difference in the length of time spent collecting nectar or the volume of nectar collected between workers from treated and control colonies. However, treated workers collected significantly less pollen, bringing back 31% less pollen per time unit to their colonies. Treated workers also collected pollen less frequently, with 41% of foraging bouts collecting pollen versus 65% for control workers; a decline of 24%.

Gill and Raine (2014) performed an experiment where B. terrestris colonies were exposed to sugar solution treated with 10 ppb of imidacloprid while also having access to forage freely outside. Colonies and individual worker bumblebees were studied over a 4-week period. In common with their previous findings (Gill et al. 2012), imidacloprid-treated workers initiated significantly more foraging trips across all 4 weeks of the experiment. The authors note that this is likely driven by an acute individual-level response in the first weeks (neonicotinoids acting as a neural partial agonist, increasing desire to forage) and by a chronic colony-level response in the latter part of the experiment, with treated colonies allocating a higher proportion of workers to pollen collection. Pollen foraging efficiency of treated workers decreased as the experiment progressed with the smallest collected pollen loads recorded in week four, suggesting a chronic effect of imidacloprid on pollen foraging ability. It is not clear whether this is as a result of individual performance deteriorating or new emerging workers having been exposed for a greater period of time.

Stanley et al. (2015a) exposed B. terrestris colonies to 2.4 or 10 ppb thiamethoxam-treated sugar solution for 13 days. Colonies were then moved to pollinator exclusion cages where they were allowed to forage freely on two varieties of apple blossom. Bees from colonies exposed to 10 ppb spent longer foraging, visited fewer flowers and brought back pollen on a lower proportion of foraging trips compared to bees from control colonies. Stanley and Raine (2016) also exposed B. terrestris colonies to 10 ppb thiamethoxam sugar solution for a 9- to 10-day period. At this point, colonies were moved to a flight arena provisioned with two common bird’s-foot trefoil (Lotus corniculatus) and one white clover (Trifolium repens) plants. Worker bees were individually released and their interaction with the flowers was recorded. Significantly more treated workers displayed pollen foraging behavior compared to control workers. However, control workers learned to handle flowers efficiently after fewer learning visits.

Arce et al. (2016) placed B. terrestris nests out in an area of parkland for a 5-week period while also supplying them with sugar solution treated with 5 ppb of clothianidin. The volume of sugar solution provided was estimated to be half that which colonies typically consume over the course of the experiment. No pollen was provided, so workers had to forage for this and to make up the shortfall in nectar resources. In contrast to the previous papers, only subtle changes to patterns of foraging activity and pollen collection were detected. There was no clear difference in colony weight gain between treatments or number of brood individuals. However, by the end of the experiment, treated colonies contained fewer workers, drones, and gynes when compared with control colonies.

Switzer and Combes (2016) studied the impact of acute imidacloprid ingestion on sonicating behavior of B. impatiens. Sonicating is a behavior whereby a bumblebee lands on a flower and vibrates loudly to shake pollen loose from anthers. Bumblebee workers were fed a dose of 0, 0.0515, 0.515, or 5.15 ng of imidacloprid in 10 μL of sugar solution. These are equivalent to concentrations of 0, 5.15, 51.5, and 515 μg/L (~ppb), with the highest volume consumed equivalent to 139% of the honeybee acute LD50, a moderate proxy for bumblebees, as bumblebees are generally less sensitive than honeybees. Bees were then allowed to forage from tomato (Solanum lysopersicum) plants and sonicating behavior was observed. At the lowest dose of 5.15 μg/L of imidacloprid, no impact was found on wingbeat frequency, sonication frequency or sonication length. No analysis could be made for higher doses, as bees in these treatments rarely resumed foraging behavior after ingesting imidacloprid. Given the neonicotinoid concentrations used in this study and sample size problems, it is difficult to draw many conclusions other than that levels of exposure above 50 μg/L impair bumblebee pollen foraging behavior.

Overall, these studies suggest that exposure to neonicotinoids in nectar at concentrations between 0.7–10 ppb can have sublethal effects on the ability of bumblebees to collect pollen at both the individual and colony level. This shortfall in pollen and subsequent resource stress is a plausible mechanism to explain diminished colony growth and production of sexuals in the absence of increased direct worker mortality. Given that concentrations as high as 10 ppb are at, but within, the upper limit of what bumblebees are likely to experience in the field, it is likely that wild bumblebees exposed to neonicotinoids in contemporary agricultural environments suffer from a reduced ability to collect pollen, with a subsequent impact on their reproductive output.

Effects on target pests

Fipronil induced Drosophila S2 cell apoptosis in vitro experiments (Zhang et al. 2015). This side effect occurs through caspase-dependent mitochondrial pathways and appears to coincide with a decrease in the mitochondrial membrane potential and an increase in reactive oxygen species. Other authors have shown significant increases in tumor frequencies on wing cells of Drosophila melanogaster, suggesting that this insecticide is mutagenic and carcinogenic in somatic cells of this fruit fly (de Morais et al. 2016a).

Wild strains of Drosophila melanogaster are rather resistant to imidacloprid with acute LD50s > 1304 μM (> 333.8 ppm) for both females and males (Charpentier et al. 2014). However, the same study has shown lethal effects of imidacloprid on chronically exposed D. melanogaster during 8 days: 27% of females died at 3.91 nM and 28% of males at 39.1 nM. The latter concentrations were several orders of magnitude below chronic LC50s of 18 and 45 μM for females and males, respectively. Moreover significant sublethal effects have been demonstrated on mating and fecundity at very low exposure concentrations (mating: both genders exposed at 0.391 nM; fecundity: females exposed at 3.91 nM), i.e., in the 0.1–1 ppb range of concentration.

Stimulated reproduction of the green peach aphid (Myzus persicae) by exposure to sublethal doses of imidacloprid had been reported previously (Yu et al. 2010). This hermetic effect undermines the effectiveness of the insecticide in controlling the target pest, and it seems to be accompanied by a complex pattern of up- and downregulation of genes during exposure. A recent study suggests that such an effect is passed on to the second generation, although there is some adaptability to low doses of the insecticide (Ayyanath et al. 2014). In another study, the soybean aphid (Aphis glycines) showed significantly higher reproduction rate when treated at sublethal doses of imidacloprid (0.05 mg/L) than in non-treated controls (Qu et al. 2015). However, other sublethal doses (0.1 and 0.2 mg/L) caused slower juvenile development, shorter reproductive periods, and reduced adult lifespan and fecundity, indicating that the threshold for hormetic responses is rather low. Stimulatory reproductive effects have also been observed with exposure of males of the Neotropical brown stink (Euschistus heros) to imidacloprid, but not with exposure of the females (Haddi et al. 2016).

A commercial mixture of a pyrethroid (b-cyfluthrin) and a neonicotinoid (imidacloprid) produced behavioral sublethal effects on bed bugs (Cimex lectularius) such as reduced locomotion, reduced feeding, and difficulties in host-finding that resulted in good control of the bugs by preventing their dispersal (Crawley et al. 2016). However, the bugs have already developed resistance to at least four neonicotinoids, acetamiprid, dinotefuran, imidacloprid, and thiamethoxam in several North American cities (Romero and Anderson 2016).

Efficacy of three neonicotinoids (acetamiprid, imidacloprid, and thiamethoxam) on controlling sand termites (Psammotermes hypostoma; Isoptera) has been demonstrated by Ahmed et al. (2015), the effect of all three lasting up to 60 days. Dembilio et al. (2015) also studied the lasting effect of imidacloprid applied by crown spray or stipe injection on palms to control the red palm weevil (Rhynchophorus ferrugineus; Coleoptera). Complete control (100%) was achieved after 45 days of stipe injections at 4–10 mL, corresponding to 2 g a.i./plant. Spray application used larger volumes that stipe injections, were less efficient and resulted in larger losses of the insecticide by washoff from the palm fronds into the surrounding environment. In Brazil, plantations of eucalypts were treated by immersing the seedlings in a solution containing fipronil (0.4%) to prevent attack by termites, as it has been shown to be protective for 56 days (dos Santos et al. 2016). A comparison of termites’ diversity between treated and untreated plots did not show significant differences, although treated plots tended to have fewer species (Silva et al. 2016). The authors indicated that “any effect [of fipronil] is masked by the effects of the plantation itself”, since both treated and untreated plantations of this tree had significantly less termite diversity than native savanna forests or regeneration forests.

Baits laced with fipronil are being used to control the expansion of the invasive Argentine ant (Linepithema humile) in Japan. Although the main super colonies of this species appear to be very susceptible to this insecticide, the treatment was also damaging for all other local ant populations. Fipronil bait treatments, therefore could lead to significant impacts on the local arthropod biodiversity (Hayasaka et al. 2015). Hydrogels containing 750 ppb thiamethoxam killed 50% of the forager ants in 3 days, while baiting with 1500 and 750 ppb provided 100% mortality of workers and queens within 8 days in the laboratory trials. These concentrations were lower than the ones required to control ants in highly infected areas as reported by Rust et al. (2015). At sublethal levels, imidacloprid may have different effects in red imported fire ants (Solenopis invicta) depending on the concentrations used; for example, concentrations of this insecticide in sugar water at 0.01 ng/L are attractive to the ants and increase their digging activity, whereas concentrations higher than 0.25 ng/L suppress their water consumption, digging and foraging behaviors (Wang et al. 2015e). At the latter concentration, newly mated queens reduced their brood tending ability, while the time to larval emergence was delayed significantly and no pupae or adult workers were produced (Wang et al. 2015d).

Effects on butterflies

Mulé et al. (2017) presented a systematic review of the effects of chemical insecticides on four common butterfly families: Lycaenidae, Nymphalidae, Hesperiidae, and Papilionidae. Only one study in their sample (Krischik et al., 2015) looked at the effects of a neonicotinoid (imidacloprid) on butterflies (Danaus plexippus and Vanessa cardui) illustrating a huge data gap. The systematic review concludes that the use of all the insecticides studied (dichlorvos, imidacloprid, malathion, naled, permethrin, and resmethrin) cause negative effects on the most common butterfly families, such as reduced survival rate, feeding interruption, and alteration of oviposition behavior.

Effects on natural enemies of pests

Compared to the research on pollinators, few studies on the toxic effects and population impacts of neonicotinoids and fipronil on other arthropods have been published in the past 2 years. Research in this case has been limited to beneficial insects used in biological control or integrated management programs (IPM), building upon the already known negative effects of these chemicals (Pisa et al. 2015). Recent studies have centered on the newly developed compounds (Giorio et al. 2017, this special issue), while the previous literature focused mainly on imidacloprid.

Effects on non-target soil organisms

Imidacloprid and thiacloprid are highly toxic to springtails (Collembola spp.), with LC50s of 0.44 and 9 mg/kg dry soil, respectively, for Folsomia candida. In multigenerational tests with this species, imidacloprid showed consistently high toxicity through three generations, whereas toxicity of thiacloprid was reduced in the second and third generation. The authors suggest that the higher persistence of imidacloprid in soil compared to that of thiacloprid could be the reason for this differential toxicity in time (van Gestel et al. 2017).

The only microcosm study available for terrestrial arthropods using imidacloprid was done by Uhl et al. (2015). The experimental setup consisted of a tritrophic system: strawberry plants, a ground cricket (Nemobius sylvestris) and a web spider (Pisaura mirabilis). Strawberry leaves were treated at two different rates that reflected a typical dosage within the crop (2.4 g/m²) and low exposure in field margins and forests (0.24 g/m²). The treatments were sublethal, as cricket mortalities were low and evenly distributed among treatments and controls. However, crickets showed significantly less mobility and feeding behaviors with the high treatment, while both treatments resulted in significantly lower weight and thorax growth. The high treatment also increased the predation of crickets by the spider, while spiders tended to move more under such circumstances. However, and surprisingly, the low treatment resulted in higher survival of the crickets than in the controls. Overall, herbivory was reduced and predation increased at sublethal concentrations of imidacloprid, suggesting possible impacts through trophic interactions.

Significant changes in invertebrate community composition were observed immediately after spray applications of fipronil for locust (Chortoicetes terminifera) control in Queensland, Australia, though the richness and abundance of species caught in pan and pitfall traps were not significantly affected. The changes in species composition for the flying insects (pan traps) persisted for up to 79 days after spraying operations, whereas those for the ground-dwelling invertebrates (pitfall traps) lasted up to 189 days. The authors of this field study explained that a long drought period that occurred during their 2-year monitoring may have influenced the slow recovery of the invertebrate populations (Walker et al. 2016). The composition of arthropod communities was not significantly affected over time in another study that used fipronil sprays for locust control in New South Wales, Australia (Maute et al. 2017a). However, springtails, mites, beetles, crickets, psocopterans, and dipterans experienced short-term decreases in abundance. The highest reductions were observed in two ant species, one of which did not recover for longer than a year. Because arthropod abundance and community assemblage changed over the 2-year study, both in control and treatment sites, changes in patterns of local rainfall over the study period were larger than changes in abundance due to pesticide treatment (Maute et al. 2017a). The same authors found no impacts on wood-eating termites activity, measured as consumption of wooden baits, or species composition but such termites are soil dwellers, are scarce in arid Australia and may not have been exposed to the sprays (Maute et al. 2016). Equally, litter decomposition carried out by soil microbial communities was not affected by the fipronil sprays (Maute et al. 2017b).

The few species of earthworms used in toxicology tests are more tolerant to neonicotinoids than other soil invertebrates. However only few species have been studied and Eisenia fetida, the species on which most studies were performed is an epigean (i.e., surface-dwelling) compost worm and is not a common species in forest or agricultural soil. The acute toxicities (LC50s in mg/kg soil) of five neonicotinoids to the earthworm Eisenia fetida after 14 days exposure were determined as 4.34 for nitenpyram, 3.05 for imidacloprid, 2.69 for acetamiprid, 2.68 for thiacloprid, and 0.93 for clothianidin (Wang et al. 2015b). The authors reported that exposures in the range 0.8–2.0 mg/kg also reduced the fecundity of this species between 39.5 and 84% depending on the compounds, while causing significant disruption of epidermal and midgut tissues. The median number of hatched cocoons (EC50) for earthworms exposed to imidacloprid was determined as 0.92 mg/kg soil, and its lowest observed effect concentrations (LOECs) for hatchability, AChE activity, growth, and DNA damage were 0.02, 0.1, 0.5, and 0.5 mg/kg soil, respectively (Wang et al. 2015c). Mixtures of imidacloprid and lambda-cyhalotrin appear to have antagonistic toxicity in these earthworms (Wang et al. 2015f). Other authors have demonstrated the low toxicity of the new compound guadipyr to Eisenia fetida, as concentrations of this insecticide in soil below 100 mg/kg in did not affect the growth or the reproduction output. Only increases in enzymatic activities of superoxidase dismutase and catalase were observed in the first few days of exposure, returning to normal levels afterwards (Wang et al. 2015a). The 48-h LC50 for acute toxicity of nitenpyram to the earthworm Pheretima posthuma was determined at 0.29 mg/kg soil (Hussain et al. 2017).

A study of the nematode communities in soils of corn fields treated with clothianidin, either as granules or in seed-coating, found significant differences in species richness and diversity compared to untreated control fields on two separate years. The main driver in community composition of nematodes was the year and month of sampling, but clothianidin treatments reduced the diversity of species significantly, with up to 5 species out of 36 being absent, even if total abundances between treatments and controls were statistically similar (Čerevková et al. 2017).

The enantiomers of fipronil appear to have different toxicity to the earthworm Eisenia fetida, with S-fipronil having a higher subchronic toxicity and bioaccumulation potential than R-fipronil (Qin et al. 2015). Weight reductions after 28 days exposure to concentrations in soil ranging 50–1000 mg/kg were 23–53% for the R-fipronil and 38–62% for the S-fipronil enantiomers. Residue accumulation in the earthworms reached a peak after 10 days exposure and then declined, following the dissipation pattern of the initial fipronil racemate in the soil. In tissues, fipronil, fipronil sulfone, and fipronil sulfide were detected, with bioaccumulation factors of 0.5–0.75 and a biological half-life in the range 1.5–2.1 days (Qin et al. 2015).

Studies on the effect of systemic pesticides on soil organisms are limited to a few species and the chosen species may not be the most ecologically meaningful. Therefore, the true impact of systemic pesticides on soil organisms and associated functions remains an important knowledge gap.

Effects on aquatic invertebrates

A comprehensive review of the acute and chronic toxicity of neonicotinoids to 49 species of aquatic insects and crustaceans, spanning 12 invertebrate orders, indicates that differences in sensitivity among aquatic invertebrate species range several orders of magnitude (Morrissey et al. 2015). More than two thirds of the data refer to imidacloprid, for which acute LC50s range from 4 μg/L in the most susceptible insect orders (Ephemeroptera, Trichoptera, and Diptera) to values exceeding 44,000 μg/L in the most tolerant cladoceran crustaceans. It is unfortunate that the standard species used in regulatory assessments, namely Daphnia magna, has a typical LC50 of about 100,000 μg/L, and this caused regulators to underestimate risks for some time. The study recommended ecological thresholds for neonicotinoids in water at concentrations below 0.2 μg/L for short-term acute exposures and 0.035 μg/L for long-term chronic exposures to avoid ecological impacts on aquatic invertebrate communities (Morrissey et al. 2015). Some of these impacts were described in a previous review (Pisa et al. 2015), and links between the toxic effects at the individual species, populations, and communities levels with impacts on aquatic and terrestrial ecosystems have been published more recently (Sánchez-Bayo et al. 2016a).

The decline of dragonflies and damselflies (Odonata) in Japan since the early 1990s has been blamed on the introduction of systemic insecticides in that country (Jinguji and Uéda 2015), but solid evidence was lacking. Impacts of these insecticides applied to rice seedlings in nursery-boxes at recommended rates were tested on the dragonfly Sympetrum frequens: none of the 50 nymphs introduced in experimental lysimeters treated with either imidacloprid or fipronil survived after 1 month, and only 13% remained in those treated with dinotefuran. Although the rate of adult emergence in the dinotefuran treatment was similar to that in the controls, the average head width of the dragonflies in this treatment was significantly narrower (Jinguji and Uéda 2015). In a similar study using paddy mesocosms treated with either clothianidin, fipronil, or chlorantraniliprole, the abundance of Odonata species in the clothianidin and specially in the fipronil-treated paddies was very low compared to controls and the other treatment (Kasai et al. 2016). Plankton species also declined in the clothianidin and chlorantraniliprole treatments right after the applications, but they recovered when their initial concentrations decreased to minimal levels. Previous mesocosm studies in Japan had showed the toxicity of imidacloprid to Odonata species when applied at the recommended rates to rice paddies (10 kg/ha). However, the study by Kobashi et al. (2017) also showed compensation among the predatory insects: while populations of Crocothemis servilia mariannae and Lyriothemis pachygastra nymphs were significantly reduced, those of Orthetrum albistylum speciosum increased slightly throughout the 5-month experimental period. Large decreases in the abundance of a common predator, Notonecta triguttata, were also observed, and the Guignotus japonicus disappeared, the effects on both species resulting from a delayed but measurable chronic toxicity (Kobashi et al. 2017). Other authors have determined the 48-h LC50 for clothianidin in North American dragonflies in the range 865 to 1245 μg/L (Miles et al. 2017).

Mayflies (Ephemeroptera) comprise other insect taxa very susceptible to neonicotinoids. The acute and chronic toxicity (28 days exposure) of thiamethoxam, thiacloprid, and imidacloprid to Cloeon dipterum, and their seasonal variability was studied by van den Brink et al. (2016). Thiacloprid was twice as toxic to the winter generation as the other two neonicotinoids, whereas both acute and chronic toxicity of imidacloprid to the summer generation was much higher than to the winter one. Camp and Buchwalter (2016) demonstrated that the higher susceptibility of 6 species of aquatic insects to imidacloprid during summer is due to the higher water temperatures during that season: the time-to-effect for sublethal impairment and immobility was significantly decreased with increasing temperature from 15 to 25 °C because the intake of the toxicant and metabolism also increased accordingly. For Cloeon dipterum, lethal median concentrations (LC50s) of the studied neonicotinoids (thiamethoxam, thiacloprid, and imidacloprid) decreased by a factor of 3 to 6 times between 24 and 96 h of exposure in either season (van den Brink et al. 2016). The same result was found with the lotic mayfly Isonychia bicolor exposed to imidacloprid (Camp and Buchwalter 2016). Moreover, chronic exposures of C. dipterum resulted in LC50s of 0.30 μg/L for thiacloprid, 0.32 μg/L for imidacloprid and 0.8 μg/L for thiamethoxam, the latter LC50s being 270, 800 and 100 times lower than their respective ones at 24 h (van den Brink et al. 2016). Also, LC50s for acute exposure of the freshwater amphipod Gammarus kischineffensis to thiamethoxam dropped from 75.6 μg/L at 24 h to 3.7 μg/L at 96 h, that is a 20-fold decrease in concentration in 4 days to achieve the same mortality effect (Uğurlu et al. 2015). These studies confirm the delayed and extreme chronic toxicity of neonicotinoids to aquatic organisms.

New toxicity data of aquatic predatory insects are now available for clothianidin (Miles et al. 2017). The 48-h LC50 for the aquatic beetle Graphoderus fascicollis (Dytiscidae) was determined at 2 μg/L, which indicates the susceptibility of this species compared to that of four species of water bugs (LC50 range 56–805 μg/L) and three species of dragonflies (LC50 range 865–1245 μg/L). The water bug Belostoma flumineum displayed a dose-dependent reduction in feeding rate after exposure to sublethal concentrations of clothianidin. The authors also carried out a mesocosm study to investigate the effect of three concentrations of clothianidin (0.6, 5, and 352 μg/L) in the arthropod communities. Predatory invertebrates experienced significant mortality with increasing levels of the insecticide in water, concomitant with increases of their prey up to 50% at the highest concentration, indicating a top-down trophic cascade in community abundance (Miles et al. 2017).

In laboratory tests with larvae of Chironomus dilutus, the 14-day LC50s for imidacloprid, clothianidin, and thiamethoxam were 1.52, 2.41, and 23.60 μg/L, respectively. However, the 40-d EC50s for adult emergence under exposure to the same chemicals were 0.39, 0.28, and 4.13 μg/L, respectively. This indicates that sublethal concentrations that prevent emergence of this key wetland species are between 4 and 9 times lower than those that cause mortality to the larvae (Cavallaro et al. 2017). Exposure of Chironomus riparius larvae to various mixtures of pyrethroid (deltamethrin and esfenvalerate) and neonicotinoid insecticides (imidacloprid and thiacloprid) at 50% of their known LC50s showed sometimes additive and other times antagonistic effects on survival (Kunce et al. 2015). In the case of the amphipod Hyalella azteca, combined exposure to imidacloprid and cyfluthrin resulted in mortality ratios of 1.7 to 2.7 higher than either insecticide alone, indicating greater than additive toxicity (Lanteigne et al. 2015).

Available toxicity data for amphipods indicate that these detritivores of organic material are less susceptible to neonicotinoids than insect larvae by one order of magnitude or more (Morrissey et al. 2015). However, such differences tend to be species specific. For example, recent studies have shown that the amphipod Gammarus fossarum is more susceptible than the caddisfly Chaetopteryx villosa when exposed to three neonicotinoids (imidacloprid, thiacloprid, and acetamiprid), either alone or in mixtures (Englert et al. 2017). Furthermore, the same study found that combined exposure of these shredder species to neonicotinoid residues in water and in food (tree leaves) had more negative impacts on their survival than direct exposure to contaminated water alone. Exposure of benthic organisms to residual neonicotinoids in water is already widespread in Europe: 47% of the 19 amphipods (Dikerogammarus spp.) collected from the Danube river in eastern Germany had residues of thiacloprid at levels 0.1–0.39 ppb (wet body weight) (Inostroza et al. 2016).

Mollusks are known to be quite tolerant of neonicotinoids (Morrissey et al. 2015), with water snails (Physa acuta and Helisoma trivolvis) having no mortality after exposure to 327 mg/L clothianidin for 2 days (Miles et al. 2017). It appears that the mechanism of toxicity of these insecticides on these organisms is different, as neonicotinoids inhibit the nACh receptors in the pond snail (Lymnaea stagnalis) instead of acting agonistically (Vehovszky et al. 2015). At environmentally relevant concentrations (0.1 to 100 μg/L), imidacloprid downregulated the production of many fatty acids in the snails, while the levels of polyamines, spermidine and putrescine increased, indicating neuron cell injury. Cholinergic gene expression was also increased as the snails tried to overcome imidacloprid binding to the nAChRs (Tufi et al. 2015). Exposure of the snails to mixtures of neonicotinoids and other pesticides found in water from agricultural areas in the Netherlands showed more disturbed metabolic pathways than exposures to individual chemicals (Tufi et al. 2016). Toxicity tests of a suite of neonicotinoids to the Ramshorn snail (Planorbella pilsbryi) showed 7-day LC50s of ≥ 4000 μg/L for imidacloprid, clothianidin, and thiamethoxam, respectively, whereas the 28-day LC50s were ≥ 182 μg/L for all three insecticides. However, growth and biomass are more sensitive endpoints than mortality for this species, with EC50s ranging from 33.2 to 122.0 μg/L (Prosser et al. 2016). Similarly, the 48-h LC50s for glochidia of the wavy-rayed lampmussel (Lampsilis fasciola) were ≥ 456 μg/L for all seven neonicotinoids tested by these authors. Thus, the aquatic studies to date confirm that neonicotinoids pose less of a hazard to mollusks compared to non-target aquatic insects.

Despite the above findings, a probabilistic risk assessment that compared the available acute and chronic toxicity data of neonicotinoids to the current levels of contamination of these insecticides in surface waters of the USA, concluded that “the aquatic invertebrate community is unlikely to be adversely affected by acute or chronic exposure to imidacloprid resulting from currently registered uses of imidacloprid in the United States” (Aslund et al. 2017). The study was funded by Bayer CropScience and three private environmental consulting firms. By contrast, a review of 29 surveys in nine countries found “that 81% (22/27) and 74% (14/19) of global surface water studies reporting maximum and average individual neonicotinoid concentrations respectively, exceeded the thresholds of 0.2 and 0.035 μg/L” for protection of 95% of aquatic species (Morrissey et al. 2015).

Studies in the natural environment

Laboratory studies suggest that neonicotinoids and fipronil can kill or harm vertebrates, and inferences can be made from laboratory studies adopting field-realistic conditions. But there are only a few studies of the impact of these systemic insecticides on vertebrate wildlife in their natural environment, involving environmentally relevant concentrations. While several were reviewed in Gibbons et al. (2015), others have been published since. Turaga et al. (2016) examined the crops of nearly 100 wild-caught quail (northern bobwhite, Colinus virginianus, and scaled, Callipepla squamata) but did not find any neonicotinoid-treated seeds; however, the authors note that this study was undertaken in an area with limited use of neonicotinoids. Bro et al. (2016) detected concentrations of up to 67 and 8.5 ppb of thiamethoxam/clothianidin and fipronil (plus fipronil sulfone), respectively, in grey partridge (Perdix perdix) eggs. While the impact of these concentrations on egg viability was unknown, they were probably too low (Table 2 in Gibbons et al. 2015; Table 3 here) to have had any deleterious effects. By contrast, Lopez-Antia et al. (2016) estimated a potential mean daily intake by individual red-legged partridges of 23.4 and 41.7 mg/kg of imidacloprid and fipronil, respectively, from eating treated seeds left on the surface at sowing. Such a daily dose of imidacloprid exceeds that which causes sublethal effects in red-legged partridges, while the dose of fipronil exceeds the species LD50. Millot et al. (2015) reported that about 10% of the grey partridges found in their study fields that were exposed to thiacloprid, subsequently died, even though few direct impacts of other pesticides were detected. Finally, Millot et al. (2017) suggested that in France, mortality due to poisoning by imidacloprid-treated seeds was at least likely in 70% of wildlife mortality incidents reported during 1994–2014. Lopez-Antia et al. (2016) conclude that the use of pesticide-treated seeds presents an unacceptable risk to farmland birds.

Several impacts other than from direct poisoning have been reported. In the Netherlands, Hallmann et al. (2014) found that local population trends of insectivorous birds were more negative in areas with higher surface water concentrations of imidacloprid; populations declined when concentrations exceeded 20 ng/L. While this study was correlative, rather than experimental, and took no account of co-occurring pesticides (Vijver and van den Brink 2014), the effects remained after correcting for other potential drivers of change in bird populations. Given that bird numbers declined at exceptionally low imidacloprid concentrations, the most likely impact of this neonicotinoid was indirect, by reducing their invertebrate food supply, although other mechanisms could not be ruled out.

Two further studies have examined the impact that fipronil sprayed to control either locusts (Maute et al. 2016) or yellow crazy ants, Anoplolepis gracilipes (Stork et al. 2014) had on reptile and bird populations. While no impact was detected on local populations of 23 reptile and 3 bird species, fipronil spraying led to a drop in numbers of one bird species (Christmas Island imperial pigeon, Ducula whartoni) and a short-term increase in another (Christmas Island white-eye, Zosterops natali). The rise in white-eye numbers was probably an indirect, food chain effect caused by a temporary increase in the number of moribund insects as prey. However, the pigeon being a frugivore, its decline may have been due to a direct toxic impact on its reproduction.

Summary of impacts on vertebrates

In the past 3 years, further evidence has highlighted negative impacts of the neonicotinoids imidacloprid and clothianidin, and fipronil (a phenyl pyrazole) on vertebrate wildlife. All three insecticides exert a wide range of deleterious sublethal effects in the laboratory, with imidacloprid, for example, altering the regulation of genes important in rat brain development and disrupting the spatial memory of echo-locating bats. These sublethal effects are often detected at concentrations substantially lower than those causing lethal effects. Except in the most extreme cases, however, aquatic vertebrates are very unlikely to be exposed to lethal concentrations of these two neonicotinoids in their natural environment, although sublethal effects of imidacloprid cannot be ruled out and is expected in some cases. By contrast, some recorded environmental concentrations of fipronil may be sufficiently high to harm fish.

Since the earlier review, new evidence has emerged suggesting that terrestrial vertebrates can be exposed to high concentrations of neonicotinoids by ingesting treated seeds left on the surface at sowing. One study estimated that the daily doses of insecticides ingested by red-legged partridges could cause sublethal (imidacloprid) or even lethal (fipronil) effects (Lopez-Antia et al. 2016). A second concluded that mortality due to poisoning by imidacloprid-treated seeds was at least likely in 70% of reported wildlife mortality incidents (Millot et al. 2017). Evidence of the indirect, food chain effects of these insecticides remains rare, though one correlative study found that some local populations of insectivorous birds declined more in areas with higher surface water concentrations of imidacloprid, possibly as a consequence of reduced invertebrate food supplies (Hallmann et al. 2014).