The investigation of honey bee pesticide poisoning incidents in Czechia

Highlights

• Acute poisoning incidents in relation to recently treated crops were investigated.

• Spraying during rapeseed flowering was the main confirmed bee poisoning cause.

• Some sites were heavily contaminated with several pesticides due to tank mixes.

• Some sites, although agricultural, had only moderate levels of pesticides.

• Chlorpyrifos and thiacloprid (banned in the EU since 2020) were major contaminants.

Abstract

Honey bees are major pollinators of crops with high economic value. Thus, bees are considered to be the most important nontarget organisms exposed to adverse effects of plant protection product use. The side effects of pesticides are one of the major factors often linked to colony losses. Fewer studies have researched acute poisoning incidents in comparison to the study of the sublethal effects of pesticides. Here, we compared pesticides in dead/dying bees from suspected poisoning incidents and the suspected crop source according to government protocols. Additionally, we analyzed live bees and bee bread collected from the brood comb to determine recent in-hive contamination. We used sites with no reports of poisoning for reference. Our analysis confirmed that not all of the suspected poisonings correlated with the suspected crop. The most important pesticides related to the poisoning incidents were highly toxic chlorpyrifos, deltamethrin, cypermethrin and imidacloprid and slightly toxic prochloraz and thiacloprid. Importantly, poisoning was associated with pesticide cocktail application. Almost all poisoning incidents were investigated in relation to rapeseed. Some sites were found to be heavily contaminated with several pesticides, including a reference site. However, other sites were moderately contaminated despite agricultural use, including rapeseed cultivation sites, which can influence the extent of pesticide use, including tank mixes and other factors. We suggest that the analysis of pesticides in bee bread and in bees from the brood comb is a useful addition to dead bee and suspected crop analysis in poisoning incidents to inform the extent of recent in-hive contamination.

Introduction

The honey bee, Apis mellifera L., is of great importance to humankind in many different ways. Honey bees are considered the most important pollinators of cultural plants and agricultural crops (Klein et al., 2007), but they have additional value in producing unique bee products, such as honey, beeswax, venom, propolis, pollen, and royal jelly, which are important for the colony and are commonly used by humans (Schmidt, 1997; Erban et al., 2019a). Additionally, beekeeping is a part of cultural heritage worldwide (Tautz, 2008). However, beekeeping can endanger substantial colony losses (Neumann and Carreck, 2010) since if beekeepers lose a large portion of managed colonies, the renewal of apiaries can be demanding and frustrating. Honey bees are exposed to various environmental stressors, and pesticide or plant protection product (PPP) exposure is considered an important factor alongside the pressure of pathogens and parasites (vanEngelsdorp et al., 2009). Because honey bees are commonly exposed to pesticides, they are important subjects of regulatory insect risk assessments related to agrochemical use and registration in any country (Desneux et al., 2007; Johnson et al., 2010). Strict regulations for pesticide application exist in different countries; however, poisoning incidents of honey bees occur annually. Although a number of studies have focused on determining pesticide residues in bee matrices, few studies have investigated the presence of pesticides or poisoning incidents in honey bee colonies.

Bees are principally exposed to the effects of pesticides by different routes (Krupke et al., 2012); however, the key issue is whether bees are surface-sprayed or whether they deliver contaminated substances to the hive as nutrient sources for the colony. Foragers are older honey bee workers responsible for the delivery of nutritional products (nectar, pollen, and water) to the colony. Only foragers come into direct contact with pesticides that are applied to plants, and these bees can die in the field due to direct pesticide exposure (Marzaro et al., 2011; Henry et al., 2012). Such a mode of severe pesticide exposure leading to rapid death (within hours or a few days) is denoted as an acute toxic effect. A second possible route of pesticide exposure is when foragers are not directly exposed to pesticides in the field but bring contaminated nutrient sources into the colony; thus, the entire colony can be influenced by stored supplies during processing of the contaminated material (Henry et al., 2012; Rumkee et al., 2017; Sponsler and Johnson, 2017; Crenna et al., 2020). These two routes of exposure to members of the colony can be determined as primary or secondary, respectively (Purdy, 2015). Thus, the importance of pesticide acquisition by foragers and the distribution of the pesticides brought by foragers into the colony have been underlined. In addition, while some of foragers acquire dangerous doses of pesticides, the majority of the colony can remain unaffected (Sponsler and Johnson, 2017). However, the oral pesticide uptake for bees in the colony has been shown to be higher than for the foragers bringing the nutrients to the colony (Crenna et al., 2020). Predicting the pesticide distribution in a colony brought by foragers with food source is difficult due to random distribution, but the exposure of larva and in-hive bees is more likely when the food with pesticides grows near them. Indeed, the pesticide contents for the receiver bees and larva depend on particular days that the workers bring food to the colony (Rumkee et al., 2017).

If the in-hive environment is contaminated, the sublethal exposure can, in severe cases, lead to chronic poisoning. In case of honey bees, this situation can be considered worse than acute toxic exposure with dead exposed foragers because the entire bee colony can be poisoned virtually unobserved; in that connection, it is important consider that bees are social insects with tens of thousands of members in a strong colony and, thus, are able to substitute the sudden losses of hundreds of members (i.e., acutely poisoned workers). Wu et al. (2011) highlighted the importance of contaminated brood comb. In such cases, brood development and adult bee longevity, as well as queen survival, can be affected by pesticide exposure (Wu et al., 2011). In addition, the bees developing from high pesticide residues containing brood comb were more susceptible to Nosema ceranae infection (Wu et al., 2012), but this susceptibility can also be true for other pathogens.

The chronic exposure of bees to pesticides has increased importance because it can be connected to additional adverse effects, such as interactions of pesticides with diverse microbial pathogens of bees (Doublet et al., 2015). The effects of sublethal doses of pesticides on the behavior of honey bees have been demonstrated, and most attention has been paid to systemic pesticides, such as neonicotinoids and fipronil (Thompson, 2003; El Hassani et al., 2005; Aliouane et al., 2009; Henry et al., 2012; Schneider et al., 2012). Sublethal pesticide exposure has raised the possibility of pesticide interactions that exhibit different stressors than those of individual pesticide exposure (Blanken et al., 2015; Wegener et al., 2016; Booton et al., 2017; Straub et al., 2019). Thus, interactions among multiple pesticides can have synergistic or additive effects on colony fitness (Pilling and Jepson, 1993; Pilling et al., 1995; Johnson et al., 2009, 2010, 2013; Wu et al., 2011), and bees with potentially altered immune systems due to pesticide exposure can also be more susceptible to parasites, such as the microsporidian N. ceranae (Pettis et al., 2012, 2013; Wu et al., 2012) or other pathogenic stressors, such as bacterial pathogens (Hernandez Lopez et al., 2017) and viruses (DeGrandi-Hoffman et al., 2013; Simon-Delso et al., 2014; Straub et al., 2019). For instance, chronic thiamethoxam exposure was observed to increase chronic bee paralysis virus (CBPV) levels in bees (Coulon et al., 2019). Moreover, it has been indicated that pesticide pressure influences the honey bee immune system through an altered gut microbiome (Kakumanu et al., 2016; Motta et al., 2018; Daisley et al., 2020). Honey bee exposure to sublethal pesticide doses also leads to additional toxic compounds related to pesticides: potential metabolites that are produced in bees and/or plants and represent risks to non-targets similar to the precursor compound (Suchail et al., 2001; Seifrtova et al., 2017; Erban et al., 2019b). An important example of a hazardous pesticide metabolite is imidacloprid-olefin, a substance found to be formed in honey bees (Suchail et al., 2001, 2004a, b), bumblebees (Erban et al., 2019b) and plants (Nauen et al., 1998; Seifrtova et al., 2017; Li et al., 2019). Imidacloprid-olefin was found to be more toxic to honey bees (Suchail et al., 2001) and different insects (Nauen et al., 1998, 1999) than the precursor compound imidacloprid. Moreover, in chronic exposure of bumblebees to imidacloprid in laboratory experiments, imidacloprid-olefin prevailed in bumblebees on the imidacloprid. This result, together with the suppression of the mevalonate pathway and fatty acid synthesis (Erban et al., 2019b), could explain the observed cumulative and delayed effect of imidacloprid exposure (Rondeau et al., 2015), including the nest behavior (Crall et al., 2018). Therefore, the environmental fate, transformations and side effects of pesticides are of importance. Finally, it is important to consider for each pesticide that the formulation in which a pesticide is applied can be a significant factor in bee poisoning (Mullin et al., 2015), although it is not commonly mentioned. However, the effect of formulation can be considered less meaningful for poisoning incidents in the case of high-toxicity pesticides compared to low-toxicity pesticides.

Different studies have aimed to screen pesticide residues in honey bees and bee matrices, such as pollen, honey or wax, to connect exposure data to bee health (e.g. (Chauzat and Faucon, 2007; Chauzat et al., 2009; Mullin et al., 2010; Wu et al., 2011, 2012; Lozowicka, 2013; Pettis et al., 2013; Barganska et al., 2014; Ravoet et al., 2015; Herrera Lopez et al., 2016; Calatayud-Vernich et al., 2018)). Studies have also analyzed samples from colonies showing mortality episodes, losses or death incidents for pesticide residues (Porrini et al., 2003; Kasiotis et al., 2014; Calatayud-Vernich et al., 2016, 2019; Beyer et al., 2018); however, these investigations are troublesome due to reductions in colony size and deaths being multifactorial issues (Kasiotis et al., 2014). The toxicity of pesticides to bees, with a focus on the situation in the USA, was reviewed by Johnson et al. (2010); the presence of pesticides has been shown in wax, pollen, bees and honey (Johnson et al., 2010). Additionally, a review by Blacquiere et al. (2012) focused specifically on neonicotinoid exposure in bees (Blacquiere et al., 2012). However, in this study, we aimed to investigate acute bee poisoning incidents – which were, importantly, officially examined by the authority of the Czech Republic after notification by beekeepers. Although bee poisoning incidents are of high interest and poisoning incidents are legislatively controlled in different countries, relevant scientific studies are rare. Kiljanek et al. (2016a) reviewed the pesticide poisoning of bees. Kiljanek et al. (2016b) also investigated pesticides from 70 suspected bee poisoning incidents and detected an array of 57 pesticides and metabolites in poisoned bee samples (Kiljanek et al., 2016b). The incidents of bee poisoning by pesticides in the UK were summarized in different years (Fletcher and Barnett, 2003; Barnett et al., 2007). Furthermore, there is an available report of bee poisoning in Germany in spring 2008 from the abrasion of active substances on treated seeds during the sowing of maize (Pistorius et al., 2009) and a study of incidents in Canada in 2007–2012 due to neonicotinoid exposure (Cutler et al., 2014) There is also a report on the largest mass incident in Czechia due to the insecticide Regent containing the active substance fipronil (Modra and Svobodova, 2009). Along with these and other publications, additional information should be found from the governmental organizations in different countries that analyze samples collected by local bee inspectors of the suspected poisonings from beekeepers. Overall, it is important to stress that the concentration of pesticides in bee samples can indicate the cause of bee poisoning. However, proving the source or offender of the poisoning requires determining a connection between coincidently/recently applied pesticides on crops and the dead bees; this is valid and most important, especially from the point of view of legislation.

In this study, we investigated suspected bee poisoning incidents in Czechia in 2015 and 2016. For comparison with the incidents, we also included samples collected from colonies/apiaries that did not show poisoning symptoms. The samples of suspected poisoning used in this study were collected in relation to the cases investigated by the State Veterinary Administration of the Czech Republic (SVA CR) in 2015 and 2016. We inspected selected samples with a more detailed approach than the valid regulation protocol in Czechia, which simply requires a comparison of the dead bees and a plant treated with PPPs that is suspected as the source of poisoning. We addressed the investigation of honey bee pesticide poisoning from a novel perspective; that is, we evaluated cases of honey bee poisoning and risks associated with the level and array of pesticides in samples (Fig. 1). We also discuss pesticides currently evaluated to represent high risks to bees and, therefore, are banned in the EU.