Fluvalinate accumulation in the beehive environment and its effects on semen quality and gene expression of drones under field conditions

TL;DR

This study shows that fluvalinate, a miticide used in beekeeping, accumulates in hives and harms drone bees' sperm quality and detox genes.

Contribution

The study reveals fluvalinate's negative effects on drone reproductive health and gene expression under field conditions.

Findings

Fluvalinate residues were found in hive stores and wax, with higher levels in the high-dose group.

High-dose fluvalinate caused increased dead cells in drone semen and reduced detox enzyme gene expression.

Control samples had trace fluvalinate, suggesting honeybee drifting between groups.

Abstract

This study investigated the accumulation of fluvalinate, a common miticide used in beekeeping, in hive compartments and bee products, as well as its potential harmful effects on detoxifying enzyme expression and sperm quality in drones. Twelve colonies were divided into three groups: a control, a high-dose fluvalinate group (nominal concentration 750 µg/kg diet), and a low-dose fluvalinate group (nominal concentration 75 µg/kg diet). Bees were continuously fed fluvalinate-spikedsugar syrup for 22 days, and a synchronized drone brood was introduced on day 3 to these colonies. After the exposure period, samples of wax, hive stores, drone semen, and intestinal tissues were collected to assess fluvalinate residues, semen quality, and gene expression of antioxidant enzymes. In the high-dose group, fluvalinate residues reached 0.12 mg/kg in carbohydrate stores and 0.196 mg/kg in wax. In the low-dose group, levels were lower but still detectable at 0.017 mg/kg in stores and 0.022 mg/kg in wax. Surprisingly, control samples contained trace residues of fluvalinate, suggesting honeybee drifting among the tested groups. Flow cytometry revealed significantly increased number of dead cells in drone semen from the high-dose group, alongside significantly reduced expression of superoxide dismutase 1 and glutathione S-transferase - key detoxifying enzymes, compared to the control group. These findings indicate that fluvalinate accumulates in hive environments and negatively impacts drone reproductive health and gene expression. This could have broader implications for colony health and the viability of future bee generations.

The online version contains supplementary material available at 10.1007/s10646-026-03073-0.

Figures

Scheme of the experiment, including important timepoints

The expression levels of selected genes. a- significant difference to control b- significant difference to 1/10 FLU. * *p* ≤ .05; *p* ≤ .01; * *p* ≤ .001

Full text

Introduction

European honey bees (Apis mellifera L.) are the most well-known and widely managed pollinators globally (Garibaldi et al. 2013). Their pollination services heavily rely on their health status. Variations in queen quality are reported as one of the many factors that can impact the overall health of the honey bee colony (van Engelsdorp and Meixner 2010). The queen plays an irreplaceable role as the only reproductive female in the colony.

(Abou-Shaara et al. 2021). Any stressor that affects her reproductive function can ultimately determine the colony’s destiny (McAfee et al. 2021). The honey bee queen is polyandrous and typically mates shortly after her emergence, usually within a week.

(Abou-Shaara et al. 2021). Virgin queens generally mate with an average of twelve drones (Rhodes 2002; Kraus et al. 2005), but recent studies indicate that queens may mate with as many as 34–77 drones (Withrow and Tarpy 2018).

The queen receives, on average, around six million spermatozoa into her oviducts from each drone (Kerr et al. 1962), but only up to 5% of the total ejaculated spermatozoa from each drone is ultimately stored in the queen’s spermatheca (Schlüns et al. 2005; Ellis et al. 2015). After her final mating flight, the queen begins to store sperm in her spermatheca using both active and passive mechanisms (Laidlaw and Page 1984; Ellis et al. 2015). After this, she remains in the hive to lay eggs unless she becomes involved in a swarming event (Johnson et al. 2010a).

A queen with a functioning reproductive system, which has mated, lays both fertilized and unfertilized eggs (Ratnieks and Keller 1998), producing up to 2000 eggs per day depending on the season, nutrition and colony size (Harbo 1986). The observed change in the chemical profile of the mandibular gland, resulting from poorly inseminated queens, causes lower attractiveness to workers within the colony and a higher frequency of queen losses (Severson and Erickson 1989). These queens tend to lay unfertilized eggs, which develop into haploid drones (Ellis et al. 2015). Furthermore, drone sperm, which is insufficient in both quality (viable sperm) and quantity, is considered a major cause of queen losses and their supersedure (Burley et al. 2008). In addition to reproductive limitations, there are also pathogens that negatively affect the health of bees. Bacteria, viruses and parasites pose danger to the bee colony and also to the queen (Genersch 2010). The mite Varroa destructor is a bee parasite that feeds on developing brood and adult bees and vectors various viruses (Damayo et al. 2023), impairing colony development with the potential to cause the death of the entire colony (Ramsey et al. 2019).

Nowadays, there are many integrated pest management programs against V. destructor (Anderson and Trueman 2000) used worldwide. Veterinary medicinal products containing pyrethroids (tau-fluvalinate and flumethrin) have been used consistently since the introduction of V. destructor to Europe and the USA, despite reports of resistance and/or cross-resistance having emerged long ago (Colin et al. 1997; Thompson et al. 2002). Moreover, in agriculture, several registered plant protection products containing tau-fluvalinate as their active ingredient are in use (Klartan^®^, Apistan^®^, Mavrik^®^). The intensive use of tau-fluvalinate in apiculture and/or agriculture, along with its high fat solubility, leaves residues in relevant bee matrices (honey, beebread, wax, propolis), as documented by numerous scientific studies worldwide (Wallner 1999; Tsigouri et al. 2004; Johnson et al. 2010a; Mullin et al. 2010; Lambert et al. 2013; Böhme et al. 2018; Martinello et al. 2019; Staroň et al. 2024). Importantly, bees (including drones) and their developmental stages are chronically exposed to pyrethroid residues as it has been demonstrated, under laboratory conditions, that tau-fluvalinate is stable for over 8 months in honey samples, stored at 35 °C in the dark (Tsigouri et al. 2001).

Research addressing the effects of insecticides on honey bee drones is comparatively sparse relative to studies on workers and queens, a pattern that likely reflects the historical emphasis on colony productivity and maintenance rather than a lack of biological importance of drones. However, their role in the reproduction of the colony cannot be denied, and the fact that their exposure to chemicals affects their sexual fitness and, consequently, the health of the whole colony has been documented (Kairo et al. 2016; Fisher and Rangel 2018). Research suggests that these haploid drones are even more susceptible to the damaging effects of insecticides (Friedli et al. 2020) and that the pesticides present in the environment affect the morphology of their reproductive organs, potentially impacting their function (Stoyanova et al. 2025).

The quality of spermatozoa is also affected by pesticides. Their viability and morphology can be negatively changed in drones affected by pesticides (Ciereszko et al. 2017; Abdelkader et al. 2021).

This study aims to test in vivo the effect of tau-fluvalinate (tech.) used as a miticide in apiculture on the drone’s spermatozoa exposed during their post-emergent developmental stage until their sexual maturity, and on gene expression of selected detoxification enzymes in drones, under realistic field conditions.

Materials and methods

Test item used in the bioassay

The pyrethroid tau-fluvalinate (tech.) was purchased from Sigma-Aldrich with the details as follows: τ-fluvalinate; Pestanal^®^; purity 95.2%, 46294-100MG and expiration Mar/2021.

Drone rearing

Four colonies per treatment and control group, all with young, fresh, naturally mated queens, which were sisters originating from single mother “CER180312” and managed in Warre-type hives (chamber size 300 × 300 × 210 mm), were reared according to good beekeeper practice, in relation to this work, good beekeeping practice involves prioritizing colony welfare and health through providing ample forage during the whole season and managing parasites with natural treatments to avoid chemical in-hive contamination. Colonies were placed at the apiary of the Institute of Apiculture near the city Liptovský Hrádok (49°02’09.7” N 19°45’22.6” E). Randomly selected nuclei in three tested groups were five meters apart from each other to minimize drifting and twenty meters from other apiary hives not included in this bioassay.

Instead of frames, guides on the bars from which the bees build their combs were used. Equalized nuclei containing 12,800–17,400 bees were selected randomly for two treatment groups (four colonies per treatment group) and one control group (four colonies). Tested maximum nominal concentration of 750 µg tau-fluvalinate/kg diet (1/1 FLU), and its 1/10 in our bioassay (1/10 FLU), was based on the maximum measured concentration in the study by Atienza et al. (1993). The test item was diluted in acetone as a solvent, with its maximum concentration in the final feeding sucrose solution reaching 5%. The stock solution was prepared freshly every day (OECD 245 2017). A solvent control group was not used in this bioassay, as the acetone was expected to evaporate quickly (Cornement et al. 2017) in the open feeder under hive conditions.

Our in vivo bioassay was conducted according to Lückmann and Schmitzer (2019) with some modifications described below. Adult worker and pupal mortality and bee brood development were not recorded in our chronic (multiple) feeding bioassay; we focused on drone semen quality after repeated exposure to tau-fluvalinate.

The scheme of the experiment is depicted in Fig. 1 and was as follows: between June 11 and June 14, 2020, twelve tested drone-free nuclei were kept in a dark, cold place. The nuclei were transferred to the testing apiary on June 14, with free access to natural sources of water, pollen, and nectar. The apiary is close to the Low Tatras National Park (with 1.3 km distance), with no bee-attractive agricultural plants within a radius of 5 km from the apiary (Beekman and Ratnieks 2000). Before hive transportation, a queen excluder was placed between the brood chamber and the bottom board, as well as between the brood chamber and the feeder, to prevent drones from entering the hive from non-tested colonies (migratory or drifting) at the apiary. This was later intended to avoid the loss of grafted drones during the exposure phase of our bioassay. The nuclei were constantly fed with syrup (50:50), and during the period from June 11 to June 22, they received a total of 3.4 L of sucrose syrup per colony from a top feeder (with a volume of 3 L). On June 17, early in the morning, the presence and egg-laying performance of each young queen were inspected and confirmed for the first time, and one tested hive in the 1/10 group was found to be queenless and thus excluded from the bioassay. On this date, all queen-right colonies were fully occupied in one Warre chamber with six combs built from virgin wax, therefore the bees were considered as acclimatized. The Standard methods for toxicology research in apiculture consider 7 days minimum period for acclimatization (Medrzycki et al. 2013). Between June 22 and July 14, all colonies were offered sucrose syrup spiked with 750 µg tau-fluvalinate/kg diet and its 1/10 (Atienza et al. 1993). The control group received sucrose syrup without the test item. In total, they received between 11.1 L and 15.5 L of syrup during the exposure phase of the bioassay. On June 24, three pieces of comb measuring 10 × 10 cm with synchronized drone brood (triplet, each representing a non-sister queen) were grafted to each tested colony in the middle of the nest. The synchronized drones emerged on June 29 and were confined in one Warre chamber until they reached sexual maturity under realistic field conditions. The sexual maturity was determined based on the drone age. Due to the growing colony size, an additional Warre chamber of half size (300 × 300 × 105 mm) was added on July 3, beneath the occupied chamber of every tested colony, still isolated with a queen excluder.

Fig. 1. Scheme of the experiment, including important timepoints

Queen’s egg-laying performance and colony strength were inspected weekly, always early in the morning to avoid drones entering the tested hive from non-tested colonies (migratory or drifting) at the apiary; potential bee robbing and syrup consumption were visually inspected daily by experienced beekeepers to ensure comparable experimental conditions among treatment groups; however, these parameters were assessed qualitatively and were not included in formal statistical analyses. All tested colonies were naturally infected with V. destructor; no mite or disease treatment measures were conducted 4 weeks prior to and/or during this bioassay.

Weather conditions

Data on weather conditions (temperature, atmospheric precipitation, and duration of atmospheric precipitation), purchased from the Slovak Hydrometeorological Institute, were collected from a close meteorological station (2.9 km apart from the testing site). The data were collected between 5 am to 8 pm, when the bees are expected to be active. Raw temperature and precipitation/rainfall data are based on a one-hour basis. A detailed report is given in Online Resource 1.

Sample collection

Drone semen was collected on July 14. Synchronized mature drones were stimulated by pressing their thorax between the two fingers of the operator. The result was the eversion of the drone endophallus and ejaculation. The semen was collected from the tip of the endophallus with a glass capillary. Each glass capillary was closed with plasticine (both ends). The pooled semen (8–10 µL of semen) originated from about 8–10 drones from the same hive was stored at 4˚C until laboratory analysis (the next day).

Drones, determined for gene expression analysis, were collected on the next day (July 15), early in the morning. Other samples (bees, wax, and sucrose stores for residue analysis) were also collected on the next day and immediately stored in -20˚C until their next laboratory analysis. Stores of carbohydrates and wax were collected from the top Warre chamber, preferably from marginal combs.

Residue analysis of fluvalinate (TECH.)

Eurofins Food Testing Slovakia s.r.o. was contracted to perform accredited tests for residue analysis of fluvalinate (tech.) in collected in-hive stores of carbohydrates and waxes. Stores of carbohydrates and wax samples were analyzed using GC-MS with a limit of quantification (LOQ) of 0.002 and 0.01 mg/kg for stores of carbohydrates and waxes, respectively. In addition to the test item tau-fluvalinate, the accredited laboratory also analyzed other selected pesticides/miticides residues including 2,4’-formoxylidid (amitraz metabolite), amitraz (as 2,4-dimethylaniline), BTS 27,271 (amitraz metabolite), amitraz (sum), acetamiprid (neonicotinoid insecticide) boscalid (fungicide), and thiacloprid (neonicotinoid insecticide) in stores of carbohydrates using LC-MS/MS with the LOQ of 0.002 mg/kg. For more details, please see the respective analytical reports, which are supplied as Online Resource 2 and Online Resource 3.

Drone semen analysis

Flow cytometry

The count and the viability of the sperm were determined by flow cytometry on a BD FACSCanto™ cytometer (Becton Dickinson Biosciences, San Diego, USA) equipped with blue (488 nm) and red (633 nm) lasers and six fluorescence detectors. BD FACS Diva^™^ Software was used to analyze the data obtained.

Determination of sperm count

Counting beads, 123 count eBeads™ (eBioscience, ThermoFisher Scientific, Carlsbad, USA), were used for sperm counting according to the manufacturer’s instructions. For the analysis, the semen was diluted 1:100 and 1:1000 in sperm buffer (BTS powder, Jørgen Kruuse A / S, Langeskov, Denmark), and 25 µL of the sample was mixed with 100 µL of well-resuspended counting particles. The position of the sperm was delimited on the FSC-A (forward scatter-area) base dot blot graph, opposite to SSC-A (side scatter-area) (the size of the cells, opposite to their granularity or inner complexity). This delimitation also includes particle counts, which have a high granularity. The particle position was then determined on the FITC-A (fluorescein isothiocyanate) dot blot graph opposite to PE-A (phycoerythrin). The cell count was determined according to the formula:

\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{array}{lll}\:Absolute\:number\:(sperm/mL)=&\frac{sperm\:count\:\times\:used\:particle\:volume}{particle\:count\:\times\:used\:sperm\:volume}\\&\times\:concentration\:of\:particles\:\left(99\times\:{10}^{4}\right)\end{array}$$\end{document}

Sperm viability

Sperm viability was determined by combined staining with propidium iodide (PI; Sigma Aldrich, St. Louis, MO, USA) and carboxyl fluorescein diacetate (cFDA; Sigma Aldrich, St. Louis, MO, USA) according to Ricci et al. (2002). This combination of fluorescent dyes allows us to distinguish between three types of cells: (1) living sperm cells emitting green fluorescence as a result of positive staining with cFDA; (2) dead sperm cells emitting red fluorescence by positive staining with PI; and (3) damaged sperm cells emitting fluorescence of both dyes (Dolník et al. 2019).

Staining procedure: 2.5 µL of 1 mM cFDA and 222.5 µL of PBS (phosphate-buffered saline; MP Biomedicals, Illkirch-Graffenstaden, France) were added to 25 µL of sperm samples, diluted 1:100 in sperm buffer (BTS powder, Jørgen Kruuse A / S, Langeskov, Denmark). The sample was incubated for 15 min at 37 °C. Consequently, 1.5 µL of PI (1 mg/mL) was added and incubated for another 15 min at 37 °C. The sperm position was delimited on the FSC-A dot plot graph opposite to SSC-A, and the viability was determined on the dot plot graph FITC-A opposite to the PE-A.

Rna isolation and cdna synthesis

The RNA for the determination of relative gene expression of antioxidant enzymes was obtained from the intestines (midgut and hindgut) of drones. From each colony, 10 drones were killed by freezing and subsequently dissected.

Each intestine collected was washed in phosphate-buffered saline (Sigma-Aldrich, MA, USA) to remove any remaining intestinal content and was stored in RNAlater (Thermo Fisher, Lithuania) at -20 °C until further processing. Before RNA isolation, the intestines from all drones from the individual group were homogenized using a hand pestle. The RNA was isolated using TRIzol (Invitrogen, USA) reagent according to the manufacturer’s protocol. The concentration and purity of the obtained RNA were assessed using a spectrophotometer, Nanodrop 8000 (Thermo Scientific, USA) at 260/280 nm. For the cDNA synthesis, the Quantitect Reverse Transcription Kit (Qiagen, Germany) was used. The same amount of RNA (1000ng) was transcribed from each sample. The resulting cDNA was used as a template for qPCR.

Gene expression analysis (qPCR)

Relative gene expression of superoxide dismutase (Sod1), glutathione S-transferase (Gst), catalase (Cat), and thiol-specific peroxidase (Tpx3) was quantified using the thermocycler BioRad CFX96 Touch (BioRad, Hercules, CA, USA). PCR was then done in 10 µL reactions, each containing 5 µL Luna^®^ Universal qPCR Master Mix (New England Biolabs, USA), 0.3 µL forward and 0.3 µL reverse primer (c = 10 µM/µL) for selected genes (Table 1), 4 µL of cDNA with a concentration of 10 ng/µL, and 0.4 µL PCR-grade water. The protocol for PCR consisted of initial denaturation at 95 °C for 60 s, followed by 40 cycles of denaturation at 95 °C for 15 s and subsequent extension at 60 °C for 30 s. Melt curve analysis was performed after the PCR to confirm amplification of the specific product. Each reaction was run in duplicate, and each assay contained a negative control without a cDNA template. β-actin (Act) was used as a reference gene. Expression levels were calculated using the 2^−ΔΔCt^ (± SD) method.

Table 1. The sequences of primers used in this studyGeneSequenceReference Sod1 F: AGCAGATGCAAGTGGTGTTGCollins et al. 2004R: GAGCACCAGCATTTCCTGTAG Gst F: AGGAGAGGTGTGGAGAGATAGTGLi et al. 2014R: CGCAAATGGTCGTGTGGATG Cat F: TGGAGCAAGTCCTGATAAAATGCCorona et al. 2005R: TGGGCCAAGACGATGTCTATG Tpx3 F: CTGCACCTGAATTTTCCGGCorona et al. 2005R: CCTTTGTAATCACTTAATTTGATTTCTT Act F: TGCCAACACTGTCCTTTCTGScharlaken et al. 2008R: AGAATTGACCCACCAATCCA

Statistics

Sperm count and viability, and relative gene expression were statistically evaluated using a one-way ANOVA test with a post hoc Tukey test with Prism software (GraphPad Software, version 9.0, San Diego, CA, USA). For all analyses, ten drones per hive were pooled to generate one sample, and each hive was treated as a single biological replicate. Significance level was set at α = 0.05.

Results

No apparent treatment-related differences in queen egg-laying performance or overall colony strength were observed during the experimental period.

Detailed weather conditions are supplied as Online Resource 1. During the period from June 11 to June 22, when tested colonies built their combs, the average daily temperature ranged from 16.1 to 20.5˚C, with a total of five precipitation events ranging from 0.6 to 18.2 mm. The duration of atmospheric precipitation covered 16.3% of their total active flight time during this period.

For our purpose, the period between June 22 and July 14, when they received spiked sucrose solution, is crucial. During this exposure period average daily temperature ranged from 13.7 to 24.1˚C, with a total of nine precipitation/rainfall events ranging from 1.3 to 25.5 mm. The duration of atmospheric precipitation covered 3.4% of their active flight time during this period.

Analysis of residues

Accredited residue analysis of fluvalinate (tech.) in collected stores of carbohydrates and waxes is summarized in Table 2 below. A detailed report is available as Online Resource 2 and Online Resource 3; the numbers of hives in Table 2 are identical to those in the reports.

Table 2. Summary of residue analysis of fluvalinate (tech.) in in-hive stores of carbohydrates and wax samplesAnalyzed measured residues (mg/kg)% of nominal concentrationHive #GroupStores of carbohydratesWaxƩ5 Control 0.0050.010.015n/a60.007< LOQ0.007n/a80.006< LOQ0.006n/a90.002< LOQ0.002n/a1 1/10 FLU

0.075 mg/kg 0.0100.0250.03546.740.0130.0130.02634.7100.0170.0220.03952.012*2 1/1 FLU

0.75 mg/kg 0.1200.1960.31642.130.0610.0860.14719.670.0860.1230.20927.9110.0870.1150.20226.9 ^LOQ of 0.002 mg/kg for analysis of stored carbohydrate samples, LOQ of 0.01 mg/kg for analysis of wax samples. n/a − not applicable, * Hive 12 inspected as queenless on 17.6.2020^

In our in vivo bioassay, all tested colonies were offered sucrose syrup spiked with 750 µg tau-fluvalinate/kg diet and its 1/10 (Atienza et al. 1993). During the first inspection (17.6.2020), in all queenright colonies, which fully occupied one Warre chamber with six combs built from virgin wax, the presence of open and sealed brood was confirmed.

None of the analyzed carbohydrate stores contained the expected nominal administered concentration; all samples had lower measured concentrations of residues. Carbohydrate stores were diluted 6.25 to 12.3 times in the 1/1 FLU group and 4.4 to 12.5 times in the 1/10 FLU group, respectively. Moreover, all carbohydrate stores and one wax sample from the control group contained trace residues of tau-fluvalinate. The highest analyzed residue of fluvalinate (tech.) in carbohydrate stores is 0.12 mg/kg diet in the higher tested group (1/1 FLU). Analysis of the wax sample from the same colony showed a similar trend, with 0.196 mg fluvalinate (tech.)/kg wax. The highest analyzed residue of fluvalinate (tech.) in carbohydrate stores in the 1/10 FLU group is 0.017 mg/kg diet, while analysis of the wax sample from the same colony indicated a similar trend with 0.022 mg fluvalinate (tech.)/kg wax.

All selected pesticides/miticides residues (amitraz, amitraz metabolites, acetamiprid, boscalid, thiacloprid) were below the LOQ of 0.002 mg/kg.

The sum of measured concentrations in stores of carbohydrates and wax ranged from 19.6 to 42.1% in the 1/1 FLU group, 34.7–52.0% in the 1/10 FLU group, respectively.

Results of flow cytometry

The sperm count and quality were evaluated by flow cytometry and are shown in Table 3. No significant difference in sperm count was observed between the individual groups. The range of sperm count for all experimental groups was from 3.07 × 10^9^ per mL to 3.45 × 10^9^ per mL (Table 3) and can be considered within the normal range, although this factor is highly variable in studies. The percentage of live cells was not significantly different among the individual groups and was approximately 98% (Table 3). Significant differences were observed in the numbers of dead sperm. The 1/1 FLU group had a significantly higher concentration of dead cells compared to the control group and the 1/10 FLU group. The 1/10 FLU group did not show a significant increase in dead sperm over the control group.

Table 3. Counts and viability of spermatozoa of dronesCounts/mLlog counts/mLLive (%)Damaged (%)Dead (%)Control3,23 × 10^9^9.49 ± 0.1498.47 ± 0.410.07 ± 0.050 ± 0 1/10 FLU 3,45 × 10^9^9.51 ± 0.1498.20 ± 0.790 ± 00.10 ± 0.07 1/1 FLU 3,07 × 10^9^9.48 ± 0.0898.00 ± 0.220.07 ± 0.050.30 ± 0.08**a, **b ^a− significant difference to control, b− significant difference to 1/10 FLU. ** p ≤ .01^

Results of gene expression analysis

Fig. 2. The expression levels of selected genes. a- significant difference to control b- significant difference to 1/10 FLU. * p ≤ .05; ** p ≤ .01; *** p ≤ .001

The relative gene expression (Fig. 2) of the Sod1 gene was significantly decreased in the group that received the full dose of fluvalinate (1/1 FLU) compared to the control group. In the case of the other two genes of antioxidant molecules (Cat and Tpx3), no meaningful change was observed between the experimental groups and the control. In the case of Gst, relative gene expression was significantly suppressed in the 1/1 FLU group when compared to the control and 1/10 FLU group.

Discussion

Every bee colony consists of a queen, sterile female workers, and haploid drones. The number of workers and drones varies throughout the season, but the fate of the whole colony depends on their health status. A drone’s physiological status is crucial for mating with the queen. Successful mating (apart from disease control and proper nutrition) is important to avoid the queen’s losses and their supersedure (Amiri et al. 2017). To date, little is known about stressors impacting queen and/or drone fertility. Pesticides are considered one of the stressors resulting in drone fertility impairment (Kairo et al. 2016; Ciereszko et al. 2017; Fisher and Rangel 2018).

In this bioassay, maintained emergent drones were chronically exposed to fluvalinate (tech.) until their sexual maturation under field conditions.

Weather conditions during the exposure period were within the expected seasonal range and did not indicate environmental constraints that could have confounded treatment-related effects on colony behavior or drone development.

During the exposure phase, tested colonies received, on average, 14.0 L per colony of offered sucrose syrup spiked with 750 µg tau-fluvalinate/kg diet (1/1 FLU group), and 14.5 L per colony of offered sucrose syrup spiked with 75 µg tau-fluvalinate/kg diet (1/10 FLU group). Bee colonies in the control group received, on average, 13.3 L per colony of sucrose syrup offered without any test item during the exposure phase of the bioassay.

Results showed that the concentration of tau-fluvalinate in hive stores of carbohydrates was diluted 6.25–12.3 times in the 1/1 FLU group (750 µg tau-fluvalinate/kg diet), 4.4–12.5 times in the 1/10 FLU group, respectively (Table 2). Observed dilution is due to their free access to surrounding natural nectar sources under good weather conditions supporting high daily flight intensity. Moreover, the mobility of tau-fluvalinate residues from deponed stores to wax, discussed below, cannot be underestimated. Residue analyses of an accredited laboratory showed that all selected pesticides/miticides residues were below the LOQ of 0.002 mg/kg, confirming that the bioassay was not conducted in an intensively used landscape and tested colonies were not treated with the miticide amitraz.

The presence of tau-fluvalinate residues in the control group (Table 2) may be explained by the forager’s drifting during the exposure phase of the bioassay. Robbing and drifting represent distinct colony interaction behaviors with different biological implications. Robbing typically involves active and often aggressive acquisition of resources from neighboring colonies by multiple workers (Wang et al. 2024), whereas drifting refers to the passive movement of workers between hives, often due to orientation errors (Free 1958). The distinction is important because during robbing, bees can be killed and stressed, affecting the colony status, whereas drifting is not stressful for bees. Each tested group was at five meters from each other, which was confirmed as insufficient according to the new OECD protocol (OECD 332 2021). Visually, bee robbing was not observed throughout our study, which means the bees from the control group collected spiked sucrose solution from neighboring hives (cross-drifting), leading to detected concentrations ranging from 0.002 to 0.007 mg/kg diet in this group (Table 2).

Tau-fluvalinate is a fat-soluble chemical that leaves residues in in-hive matrices (Végh et al. 2023), posing a dangerous impact not only on all developmental stages and adult bees (chronic exposure) (Benito-Murcia et al. 2024) but potentially also on humans as consumers (bee bread, honey, and raw propolis) (Brancato et al. 2018). Several residue studies have detected tau-fluvalinate residues in relevant bee matrices (Wallner 1999; Johnson et al. 2010b; Mullin et al. 2010; Lambert et al. 2013; Böhme et al. 2018; Staroň et al. 2024). In addition to proteins, lipids, and amino acids, carbohydrates represent a fundamental component of bee nutrition (Paoli et al. 2014; Vaudo et al. 2015). Therefore, sublethal but biologically relevant effects may be expected when bees are exposed to contaminated carbohydrate sources, as assessed in our bioassay. Despite this, the mobility of tau-fluvalinate residues within the in-hive environment remains poorly understood. Our results demonstrate that residue levels in wax were 47% ±27% higher than those detected in stored carbohydrates in the 1/1 FLU treatment group, and 54% ±32% higher in the 1/10 FLU group (Table 2). These findings confirm the pronounced affinity and mobility of tau-fluvalinate toward wax matrices within the hive.

Adult drones exposed to several miticides (tau-fluvalinate, coumaphos, fenpyroximate, amitraz, thymol, and oxalic acid) showed no effect on drone sperm viability two to three weeks after their single contact exposure (Johnson et al. 2013; Murray et al. 2025). Contact exposure of drones to coumaphos, even though used in recommended concentration, significantly reduced their sperm viability during six weeks (Burley et al. 2008). Our results show a similar situation, where viability remained in all groups, above 98% (Table 3). However, we also observed that the percentage of dead sperm was significantly higher (0% compared to 0.3%) in the group that consumed 1/1 FLU (750 µg tau-fluvalinate/kg diet) compared to the control (Table 3). The biological significance of such a minor change is not certain despite the statistical significance. Chronic exposure to insecticides present in beeswax affects the viability of drone sperm if the drones develop in such an environment (Fisher and Rangel 2018). In the current study, the drones were not exposed to fluvalinate within the hive structures during their development, indicating that pesticide exposure may have a more significant impact during developmental stages and spermatogenesis than in adulthood. The average half-life of the pesticides in beeswax is around five years (Bogdanov 2004). This suggests that although new wax is used each consecutive year poses a risk of accumulation and exposure of the drones to these chemicals, affecting their sperm viability. Theoretically, if the exposure window in our study was prolonged, we would observe a markedly dose-dependent effect on drone sperm viability. Notably, using the cytometer does not allow for observing any morphological changes of sperm cells, which would affect the fertility of drones, nor does it observe DNA changes, which would translate to the damage of the population (Mitkovska et al. 2025).

The gene expression of detoxification genes showed a mixed trend in our research (Fig. 2). Expression of the Gst, which is responsible for protection against oxidative damage induced by insecticides (Che-Mendoza et al. 2009), had significantly decreased expression in the drone gut, which contrasts with different studies where Gst is strongly induced upon exposure to external oxidative stimuli (Nikolić et al. 2019; Shan et al. 2022). The Sod1 gene, which codes for another detoxification enzyme, also showed a significant decrease in the 1/1 FLU group. The observed downregulation of Gst and Sod1 in the drone gut following flumethrin exposure may reflect a shift in antioxidant defense priorities rather than a simple suppression of detoxification capacity. Pyrethroids such as flumethrin can alter redox homeostasis through multiple pathways. One possibility is metabolic trade-offs: detoxification is energetically costly, and under chronic or sublethal exposure, resources may be reallocated toward tissues critical for reproductive success (e.g., testes, accessory glands), as suggested by Abdelkader et al. (2019), at the expense of gut-based defenses. Another is tissue-specific regulation: antioxidant gene expression can vary markedly between organs, with downregulation in the gut potentially coinciding with compensatory upregulation in neural or reproductive tissues, mediated by differential transcription factor activation or hormonal signaling (Qi et al. 2020). However, the gene expression results in this study capture only one timepoint and therefore should be considered as an indication of the pathways included. Future studies should incorporate multiple timepoints to fully characterize the progression of these responses.

Additionally, gut microbiota-mediated effects may play a role: pyrethroid exposure can disrupt microbial communities, which in turn modulate host oxidative stress signaling and antioxidant gene expression (Hotchkiss et al. 2022). Finally, direct enzyme inhibition by pyrethroids or their metabolites could contribute, as some compounds have been shown to alter SOD and GST activity through direct interaction or oxidative damage to the enzymes themselves (Price and McGraw 2025). This decrease in the expression of genes coding for detoxifying enzymes may be problematic for drones in the long term, as their capacity to detoxify insecticides or other chemicals could potentially be impaired, making them more vulnerable to oxidative stress. In contrast, the expressions of Tpx3 and Cat were not affected, suggesting a different mode of action in the regulation of these genes. Future research should focus on analyzing the response in reproductive organs or the ejaculate.

Conclusion

Modern beekeeping relies heavily on the use of miticides to control the population of Varroa mites (Mitton et al. 2022), a significant bee parasite and vector of various diseases (Rosenkranz et al. 2010). Without treatment, significant losses would be expected in bee colonies (Korpela et al. 1992).

This study demonstrated under field conditions that the commonly used pesticide/miticide fluvalinate residues can be detected not only in the carbohydrate matrices stored within the hive but also at elevated concentrations within the wax structures, indicating chemical persistence and potential bioavailability. The results directly indicate the ability of virgin beeswax to bind active fluvalinate residues. It acts as a natural buffer in the hive environment. This fact needs to be constantly reminded to practicing beekeepers so that they do not forget the importance of the proportion of virgin beeswax in the production of new honeycomb foundations. Of particular significance is the demonstrable adverse effect on drone physiology. Chronic oral exposure to fluvalinate at a nominal concentration of 750 µg tau-fluvalinate/kg diet resulted in a marked downregulation of detoxification enzyme expression, suggesting a diminished capacity of drones to metabolize xenobiotics. The number of dead sperm increased after continuous oral exposure to a nominal concentration of 750 µg tau-fluvalinate/kg diet, suggesting a worse reproductive capacity of these individuals. While drones do not directly contribute to a colony’s productivity, they play a crucial role in establishing new colonies, and any reduction in their reproductive ability could impact future generations.

Further research is needed to evaluate how development in older wax foundations with potentially higher residue levels affects the development of bees and drones, and how the offspring of these drones are affected. Addressing these sublethal, yet biologically consequential effects of chemicals used in beekeeping is important for the sustainable apiculture and the protection of bee health.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Supplementary Material 3