Honey bees (Apis mellifera) and other pollinator populations are declining worldwide, and the reasons remain controversial. Based on laboratory testing, fungicides have traditionally been considered bee-safe. However, there have been no experimental tests of the effects of fungicides on colony health under field conditions, and limited correlational data suggests there may be negative impacts on bees at levels experienced in the field. We tested the effects of one of the most commonly used fungicides on colony health by feeding honey bee colonies pollen containing Pristine® (active ingredients: 25.2% boscalid, 12.8% pyraclostrobin) at four levels that bracketed concentrations we measured for pollen collected by bees in almond orchards. We also developed a method for calculating per-bee and per-larva dose. Pristine® consumption significantly and dose-dependently reduced worker lifespan and colony population size, with negative health effects observed even at the lowest doses. The lowest concentration we tested caused a 15% reduction in the worker population at an estimated dosage that was three orders of magnitude below the estimated LD₁₅ values for previous acute laboratory studies. The enhanced toxicity under field conditions is at least partially due to activation of colonial nutritional responses missed by lab tests. Pristine® causes colonies to respond to perceived protein malnutrition by increasing colony pollen collection. Additionally, Pristine induces much earlier transitioning to foraging in individual workers, which could be the cause of shortened lifespans. These findings demonstrate that Pristine® can negatively impact honey bee individual and colony health at concentrations relevant to what they experience from pollination behavior under current agricultural conditions.

During the last years, declines in honey bee colonies are being registered worldwide. Cholinergic pesticides and their extensive use have been correlated to the decline of pollinators and there is evidence that pesticides act as neuroendocrine disruptors affecting the metabolism of neuropeptides. However, there is a big absence of studies with quantitative results correlating the effect of pesticide exposure with changes on neuropeptides insects, and most of them are conducted under laboratory conditions, typically with individual active ingredients. In this study, we present an analytical workflow to evaluate pesticide effects on honey bees through the analysis of (neuro)peptides. The workflow consists of a rapid extraction method and liquid chromatography with triple quadrupole for preselected neuropeptides. For non-target analysis, high resolution mass spectrometry, multivariate analysis and automatic identification of discriminated peptides using a specific software and protein sequence databases. The analytical method was applied to the analysis of target and non-target (neuro)peptides in honey bees with low and high content of a wide range of pesticides to which have been exposed in field conditions. Our findings show that the identification frequency of target neuropeptides decreases significantly in honey bees with high concentration of pesticides (pesticide concentrations ≥ 500 μg kg⁻¹) in comparison with the honey bees with low content of pesticides (pesticide concentrations ≤ 20 μg kg⁻¹). Moreover, the principal component analysis in non-target search shows a clear distinction between peptide concentration in honey bees with high level of pesticides and honey bees with low level. The use of high resolution mass spectrometry has allowed the identification of 25 non-redundant peptides responsible for discrimination between the two groups, derived from 18 precursor proteins.

The usage of inadequately processed industrial waste water (WW) can lead to strong soil alkalinity and soil salinization of agricultural fields with negative consequences on soil properties and biota. Gypsum as a soil amendment to saline-sodic soils is widely used in agricultural fields to improve their soil physical, chemical and hence biological properties. This study aimed at analysing the effects of intensive WW irrigation on the structure and composition of soil-dwelling arthropods on spinach fields (Spinacia oleracea L.) in a West African urban vegetable production system. We used gypsum as a soil amendment with the potential to alleviate soil chemical stress resulting in a potentially positive impact on soil arthropods. A total of 32 plots were established that showed a gradient in soil pH ranging from slight to strong soil alkalinity and that were irrigated with WW (n = 12) or clean water (CW; n = 20), including eight plots into which gypsum was incorporated. Our study revealed a high tolerance of soil-dwelling arthropods for alkaline soils, but spinach fields with increased soil electrical conductivity (EC) showed a reduced abundance of Hymenoptera, Diptera and Auchenorrhyncha. Arthropod abundance was positively related to a dense spinach cover that in turn was not affected by WW irrigation or soil properties. Gypsum application reduced soil pH but increased soil EC. WW irrigation and related soil pH affected arthropod composition in the investigated spinach fields which may lead to negative effects on agronomical important arthropod groups such as pollinators and predators.

The worldwide decline of pollinators is of growing concern and has been related to the use of insecticides. Solitary bees are potentially exposed to many insecticides through contaminated pollen and/or nectar. The kinetics of these compounds in solitary bees is, however, unknown, limiting the use of these important pollinators in pesticide regulations. Here, the toxicokinetics (TK) of chlorpyrifos (as Dursban 480 EC), cypermethrin (Sherpa 100 EC), and acetamiprid (Mospilan 20 SP) was studied for the first time in Osmia bicornis females at sublethal concentrations (near LC₂₀ₛ). The TK of the insecticides was analysed in bees continuously exposed to insecticide-contaminated food in the uptake phase followed by feeding with clean food in the decontamination phase. The TK models differed substantially between the insecticides. Acetamiprid followed the classic one-compartment model with gradual accumulation during the uptake phase followed by depuration during the decontamination phase. Cypermethrin accumulated rapidly in the first two days and then its concentration decreased slowly. Chlorpyrifos accumulated similarly rapidly but no substantial depuration was found until the end of the experiment. Our study demonstrates that some insecticides can harm solitary bees when exposed continuously even at trace concentrations in food because of their constant accumulation leading to time-reinforced toxicity.

Recent studies have shown that sub-lethal doses of herbicides may affect plant flowering, however, no study has established a direct relationship between the concentrations of deposited herbicide and plant flowering. Here the aim was to investigate the relationship between herbicide spray drift deposited on non-target plants and plant flowering in a realistic agro-ecosystem setting. The concentrations of the herbicide glyphosate deposited on plants were estimated by measuring the concentration of a dye tracer applied together with the herbicide. The estimated maximal and average deposition of glyphosate within the experimental area corresponded to 30 g glyphosate/ha (2.08% of the label rate of 1440 g a.i./ha) and 2.4 g glyphosate/ha (0.15% label rate), respectively, and the concentrations decreased rapidly with increasing distance from the spraying track. However, there were not a unique relation between distance and deposition, which indicate that heterogeneities of turbulence, wind speed and/or direction can strongly influence the deposition from 1 min to another during spraying. The effects of glyphosate on cumulative flower numbers and flowering time were modelled using Gompertz growth models on four non-target species. Glyphosate had a significantly negative effect on the cumulative number of flowers on Trifolium pratense and Lotus corniculatus, whereas there were no significant effects on Trifolium repens, and a positive, but non-significant, effect on number of flowers on Cichorium intybus. Glyphosate did not affect the flowering time of any of the four species significantly. Lack of floral resources is known to be of major importance for pollinator declines. The implications of the presented results for pesticide risk assessment are discussed.

Tropospheric ozone is a highly oxidative pollutant with the potential to alter plant metabolism. The direct effects of ozone on plant phenotype may alter interactions with other organisms, such as pollinators, and, consequently, affect plant reproductive success. In a set of greenhouse experiments, we tested whether exposure of plants to a high level of ozone affected their phenological development, their attractiveness to four different pollinators (mason bees, honeybees, hoverflies and bumblebees) and, ultimately, their reproductive success. Exposure of plants to ozone accelerated flowering, particularly on plants that were growing in autumn, when light and temperature cues, that commonly promote flowering, were weaker. Simultaneously, there was a tendency for ozone-exposed plants to disinvest in vegetative growth. Plant exposure to ozone did not substantially affect pollinator preference, but bumblebees had a tendency to visit more flowers on ozone-exposed plants, an effect that was driven by the fact that these plants tended to have more open flowers, meaning a stronger attraction signal. Honeybees spent more time per flower on ozone-exposed plants than on control plants. Acceleration of flower production and the behavioural responses of pollinators to ozone-exposed plants resulted in retained reproductive fitness of plants pollinated by bumblebees, honeybees and mason bees, despite the negative effects of ozone on plant growth. Plants that were pollinated by hoverflies had a reduction in reproductive fitness in response to ozone. In a natural setting, acceleration of flowering by ozone might foster desynchronization between plant and pollinator activities. This can have a strong impact on plants with short flowering periods and on plants that, unlike wild mustard, lack compensatory mechanisms to cope with the absence of pollinator activity in the beginning of flowering.

Apis mellifera is a pollinator insect model in pesticide risk assessment tests for bees. However, given the economic and ecological importance of stingless bees such as Melipona scutellaris in the Neotropical region, as well as the lack of studies on the effect of insecticides on these bees, toxicity tests for stingless bees should be carried out to understand whether insecticides affect both species of bees in the same manner. Thus, the present study quantified the differential sensitivity of the bees M. scutellaris and A. mellifera to the oral ingestion of the insecticide thiamethoxam by determining the mean lethal concentration (LC₅₀), mean lethal time (LT₅₀), and their effect on the insecticide target organ, the brain. The results showed that the stingless bee is more sensitive to the insecticide than A. mellifera, with a lower LC₅₀ of 0.0543 ng active ingredient (a.i.)/μL for the stingless bee compared to 0.227 ng a.i./μL for A. mellifera. When exposed to a sublethal concentration, morphological and ultrastructural analyses were performed and evidenced a significant increase in spaces between nerve cells of both species. Thus, A. mellifera is not the most appropriate or unique model to determine the toxicity of insecticides to stingless bees.

Cavity nesting bees are proficient and important pollinators that can augment or replace honey bee pollination services for some crops. Relatively little is known about specific pesticide concentrations present in cavity nesting insect reed matrices and associated potential risks to cavity nesting bees. Nesting substrates (Phragmites australis reeds in bundles) were deployed in an agriculturally intensive landscape to evaluate colonization and agrochemical exposure among cavity nesting pollinators over two consecutive field seasons. Composition of insect species colonizing reeds within nest bundles varied considerably; those placed near beef cattle feed yards were dominated by wasps (93% of the total number of individuals occupying reed nest bundles), whereas nest bundles deployed in cropland-dominated landscapes were colonized primarily by leaf cutter bees (71%). All nesting/brood matrices in reeds (mud, leaves, brood, pollen) contained agrochemicals. Mud used in brood chamber construction at feed yard sites contained 21 of 23 agrochemicals included in analysis and >70% of leaf substrate stored in reeds contained at least one agrochemical. Moxidectin was most frequently detected across all reed matrices from feed yard sites, and moxidectin concentrations in nonviable larvae were more than four times higher than those quantified in viable larvae. Agrochemical concentrations in leaf material and pollen were also quantified at levels that may have induced toxic effects among developing larvae. To our knowledge, this is the first study to characterize agrochemical concentrations in multiple reed matrices provisioned by cavity-nesting insects. Use of nest bundles revealed that cavity nesting pollinators in agriculturally intensive regions are exposed to agrochemicals during all life stages, at relatively high frequencies, and at potentially lethal concentrations. These results demonstrate the utility of nest bundles for characterizing risks to cavity nesting insects inhabiting agriculturally intensive regions.

The ecological and economic contributions of pollinator bees to agricultural production have been threatened by the inappropriate and excessive use of pesticides. These pesticides are often applied in areas with ecological peculiarities (e.g., the Neotropical savannah-like region termed as Cerrado) that were not considered during the product development. Here, we conducted field experiments with melon (i.e., Cucumis melo L.) plants cultivated under Brazilian Cerrado conditions and evaluated the impacts of botanical (i.e., neem-based insecticide) and synthetic (i.e., the pyrethroid insecticide deltamethrin and the fungicides thiophanate-methyl and chlorothalonil) pesticides on the flower visitation rates of naturally occurring pollinator bees. Our results revealed that both honey bees (i.e., Apis mellifera L.) and non-Apis bees visited melon flowers and the intensity of bee visitation was moderately correlated with yield parameters (e.g., number of marketable fruits and fruit yield). Pesticide treatments differentially affected bee species. For instance, Plebeia sp. bees were not affected by any pesticide treatment, whereas both A. mellifera and Halictus sp. bees showed reduced visitation intensity after the application of deltamethrin or neem-based insecticides. Fungicide treatment alone did not influence the bee's visitation intensity. Deltamethrin-treated melon fields produced significantly lighter marketable fruits, and the melon yield was significantly lower in melon fields treated with the neem-based insecticide. Thus, our findings with such pollinator bees reinforce the idea that field applications of botanical pesticides may represent as risky as the applications of synthetic compounds, indicating that these alternative products should be submitted to risk assessments comparable to those required for synthetic products.

The widespread cultivation of transgenic Bt cotton makes assessing the potential effects of this recombinant crop on non-target organisms a priority. However, the effect of Bt cotton on many insects is currently virtually unknown. The plant bug Adelphocoris suturalis is now a major pest of cotton in southern China and the beetle Haptoncus luteolus is one of the most ancient cotton pollinators. We conducted laboratory experiments to evaluate the toxicity of the Bt cotton varieties ZMSJ, which expresses the toxins Cry1Ac and Cry2Ab, and ZMKCKC, which expresses Cry1Ac and EPSPS, on adult A. suturalis and H. luteolus. No significant increase in the mortality of either species was detected after feeding on Bt cotton leaves or pollen for 7 days. Trace amounts of Cry1Ac and Cry2Ab proteins could be detected in both species but in vitro binding experiments found no evidence of Cry1Ac and Cry2Ab binding proteins. These results demonstrate that feeding on the leaves or pollen of these two Bt cotton varieties has no toxic effects on adult A. suturalis or H. luteolus.