Conceived and designed the experiments: JYW CMA WSS. Performed the experiments: JYW. Analyzed the data: JYW. Contributed reagents/materials/analysis tools: WSS. Wrote the paper: JYW CMA WSS.
The authors have declared that no competing interests exist.
Numerous surveys reveal high levels of pesticide residue contamination in honey bee comb. We conducted studies to examine possible direct and indirect effects of pesticide exposure from contaminated brood comb on developing worker bees and adult worker lifespan.
Worker bees were reared in brood comb containing high levels of known pesticide residues (treatment) or in relatively uncontaminated brood comb (control). Delayed development was observed in bees reared in treatment combs containing high levels of pesticides particularly in the early stages (day 4 and 8) of worker bee development. Adult longevity was reduced by 4 days in bees exposed to pesticide residues in contaminated brood comb during development. Pesticide residue migration from comb containing high pesticide residues caused contamination of control comb after multiple brood cycles and provided insight on how quickly residues move through wax. Higher brood mortality and delayed adult emergence occurred after multiple brood cycles in contaminated control combs. In contrast, survivability increased in bees reared in treatment comb after multiple brood cycles when pesticide residues had been reduced in treatment combs due to residue migration into uncontaminated control combs, supporting comb replacement efforts. Chemical analysis after the experiment confirmed the migration of pesticide residues from treatment combs into previously uncontaminated control comb.
This study is the first to demonstrate sub-lethal effects on worker honey bees from pesticide residue exposure from contaminated brood comb. Sub-lethal effects, including delayed larval development and adult emergence or shortened adult longevity, can have indirect effects on the colony such as premature shifts in hive roles and foraging activity. In addition, longer development time for bees may provide a reproductive advantage for parasitic
Losses associated with colony collapse disorder (CCD) represent a continuation in sudden and often catastrophic population crashes in honey bee (
Honey bee colony health can be affected by many factors including hygienic behavior, innate immunity, pesticide sensitivity, nutrition, adult age, and temperature. As social insects, honey bees have evolved various traits, such as grooming or other hygienic behaviors (including removal of mites and dead or diseased brood) that protects the colony against pests and pathogens. Social immunity provides significant protection for honey bee colonies and it has been suggested that this may explain why, compared to non-social insects, honey bees are relatively immunologically deficient (i.e., express fewer immune response proteins)
In this study we examined the sub-lethal effects of developmental exposure to pesticide residues on worker bees. Worker bees were reared in brood comb containing high levels of known pesticide residues or in brood comb relatively free of pesticide residues. We discuss implications of sub-lethal and indirect effects of pesticide residues in brood comb on colony health and structure.
Frames of treatment brood comb originated from migratory Pacific Northwest beekeeping operations that used miticides and from colonies provided by the USDA- ARS honey bee laboratory, Beltsville, MD that were suspected to have died from Colony Collapse Disorder. Pesticide residue analyses were performed on brood comb samples and thirteen frames of brood combs positive for high levels of pesticide residues were cut into treatment blocks (11×11-cm), each containing roughly 450 cells. Control brood combs were newly drawn out from a single colony or sampled from feral colonies that tested negative for pesticide residue contamination.
Standard Langstroth frames, with the center area (22×11-cm) of the frame removed, were used as frame supports for a pair of comb blocks, i.e., one low pesticide residue control comb placed next to a treatment comb block containing high pesticide residue levels (n = 17). Three colonies of similar strength were used from May through August of 2008 and 2009 to host experimental frames supporting paired comb blocks. Placing control and treatment combs within the same colony during larval development equalized possible effects of colony activity and quality of resources fed to brood. Laying sister queens from each colony were caged for 24 hours over experimental frames, allowing access to both control and treatment comb blocks. Queens were released the following day and excluded to the bottom box for the duration of the experiment. Frames containing a patch of 224 eggs on control and treatment blocks were photographed and frames with insufficient number of eggs were removed from the experiment. Egg patches were monitored for larval mortality on days 4, 8,12, and 19 of development, and photographs taken of larvae developing in control and treatment comb were mapped using Microsoft Paint 2007. On day 19, experimental frames containing pupae reared in control and treatment comb were incubated at 33±1°C. Push-in cages were used to isolate treatment and control blocks. Emergence of adult bees was recorded daily and bees were counted, tagged with Testor’s enamel, and placed in a 3.2 mm mesh metal cage (11×9×5-cm). Bees reared in treatment blocks were placed in the same cage with bees reared in corresponding control blocks from the same frame. Worker bees were fed water, 50% sucrose syrup, and pollen supplement
Brood comb samples were sent to Roger Simonds USDA-AMS-National Science Laboratory, Gastonia, NC to be analyzed using QuEChERS method. Pesticide residue extraction and analysis was accomplished using liquid chromatography combined with tandem mass spectrometry (LC/MS/MS - Agilent 1100 LC equipped with a Thermo Quantum Discovery Max Triple Quadrupole Mass Spectrometer or equivalent), gas chromatography coupled with mass selective detection in electron impact mode (GC/MS-EI - Agilent 6890 GC equipped with a Agilent 5975 Mass Selective Detector in EI mode or equivalent), and gas chromatography coupled with mass selective detection in negative chemical ionization mode (GC/MS-NCI - Agilent 6890 GC equipped with a Agilent 5975 Mass Selective Detector in NCI mode or equivalent). Pesticide residues extracted from comb samples were quantified using matrix matched calibration standards of known concentrations prepared from neat standard reference material. Measurements were reported in nanograms of active ingredient per gram of wax (ng/g) or parts per billion (ppb). Identification of extracted residues was achieved through mass spectral comparison of ion ratios with standards, 171 of the most commonly used pesticides and their metabolites, of known identity. Limits of detection were in the low parts per billion (ppb).
To assess the sub-lethal effects of exposure to pesticide residues, biologically meaningful parameters were measured throughout the main stages of the honey bee life cycle. Egg eclosion, or successful hatching was measured at day 4; larval mortality and development time from egg to pupa were recorded at day 8; pupation was recorded at day 12 and 19; adult emergence rate was recorded on day 20 and continued daily until emergence was no longer observed; and adult longevity was recorded daily until all caged bees were dead. Observations of abnormal larval development and signs of disease or pest infection were also recorded. Taken together, these life cycle parameters enabled assessment of the health effects of exposure to sub-lethal pesticide residues in brood comb.
Pairwise comparisons with repeated measures were performed on larval mortality, adult longevity, and adult emergence rate of worker bees reared in relatively uncontaminated brood comb and brood comb containing high levels of pesticide residues. Comparisons of both treatments were made by sample day (4, 8, 12 and 19) and by the number of brood cycles (Rep 1, 2, 3). Differences in pesticide analyses, specifically the number of pesticide residues and the levels detected in control and treatment comb used multiple times, were compared before and after the experiment. Normality assumptions were accepted for bee mortality on day 4, 9, 12, and 19 in both control and treatment combs (Shapiro-Wilk W = 0.844 and 0.929, respectively). Statistical differences were detected by one-way analysis of variance (ANOVA) followed by paired two-tailed t-tests on control and treatment combs with significance determined at
The number of different pesticide residues detected in treatment combs ranged from 4 to 17, averaging 10. The total number of pesticides detected in all treatments was 39 including 7 fungicides, 2 herbicides, 23 insecticides (miticides included) and 7 metabolites (
Active ingredient | Chemical Family | Purpose of use | Toxicity honey bee | Average (ng/g) | % detected | min | max | LOD |
2,4 Dimethylphenyl formamide (DMPF) | metabolite | 145 | 15 | 142 | 147 | 4 | ||
3-hydroxycarbofuran | metabolite | 23 | 8 | * | 23 | 4 | ||
Aldicarb | Carbamate | INSECT | High | 20 | 8 | * | 20 | 4 |
Azoxystrobin | Strobilurin | FUNG | 19 | 38 | 5 | 29 | 2 | |
Boscalid | Carboxamide | FUNG | 35 | 15 | 35 | 64 | 4 | |
Carbendazim (MBC) | metabolite | 21 | 31 | 4 | 48 | 5 | ||
Carbofuran | Carbamate | INSECT | High | 32 | 8 | * | 32 | 5 |
Chlorothalonil | Chloronitrile | FUNG | 17 | 62 | 4 | 66 | 1 | |
Chlorpyrifos | Ogranophosphate | INSECT | High | 8 | 62 | 3 | 15 | 1 |
Clothianidin | Neonicotinoid | INSECT | High | 35 | 8 | * | 35 | 20 |
Coumaphos | Ogranophosphate | INSECT | Mod | 8079 | 100 | 281 | 22100 | 1 |
Coumaphos oxon | metabolite | 596 | 100 | 10 | 3140 | 1 | ||
Cyfluthrin | Pyrethroid | INSECT | Low | 43 | 17 | 8 | 79 | 2 |
Cypermethrin | Pyrethroid | INSECT | High | 2 | 8 | * | 2 | 2 |
Cyprodinil | Anilinopyrimidine | FUNG | 27 | 31 | 13 | 61 | 16 | |
Diazinon | Ogranophosphate | INSECT | High | 1 | 15 | 1 | 2 | 1 |
Dicofol | Organochlorine | INSECT | Low | 6 | 23 | 4 | 8 | 1 |
Dinotefuran | Neonicotinoid | INSECT | High | 97 | 8 | * | 97 | 30 |
Diphenylamine | Amine | INSECT | 151 | 23 | 20 | 281 | 1 | |
Endosulfan I | Organochlorine | INSECT | Mod | 2 | 54 | 1 | 4 | 1 |
Endosulfan II | Organochlorine | INSECT | Mod | 2 | 38 | 1 | 5 | 1 |
Endosulfan sulfate | metabolite | 1 | 31 | 1 | 2 | 1 | ||
Esfenvalerate | Pyrethroid | INSECT | High | 5 | 46 | 1 | 12 | 1 |
Fenhexamid | Hydroxyanilide | FUNG | 46 | 8 | * | 46 | 6 | |
Fenpropathrin | Pyrethroid | INSECT | High | 7 | 8 | * | 7 | 1 |
Fluvalinate | Pyrethroid | INSECT | High | 6712 | 100 | 164 | 24340 | 1 |
Imidacloprid | Neonicotinoid | INSECT | High | 45 | 8 | * | 45 | 20 |
Iprodione | Dicarboximde | FUNG | 283 | 8 | * | 283 | 20 | |
Malathion oxon | metabolite | 22 | 8 | * | 22 | 4 | ||
Norflurazon | Fluorinated pyridazinone | HERB | 5 | 8 | * | 5 | 6 | |
Oxamyl | Carbamate | INSECT | High | 22 | 8 | * | 22 | 5 |
Oxyfluorfen | Diphenyl ether | HERB | 2 | 23 | 1 | 2 | 1 | |
Permethrin total | Pyrethroid | INSECT | High | 103 | 8 | * | 103 | 10 |
Phosalone | Ogranophosphate | INSECT | Mod | 32 | 8 | * | 32 | 10 |
Pyrethrins | Pyrethroid | INSECT | High | 229 | 8 | * | 229 | 50 |
Thiacloprid | Neonicotinoid | INSECT | Low | 113 | 8 | * | 113 | 8 |
Thiamethoxam | Neonicotinoid | INSECT | High | 38 | 8 | * | 38 | 20 |
THPI | metabolite | 96 | 15 | 93 | 99 | 50 | ||
Vinclozolin | Dicarboximde | FUNG | 1 | 8 | * | 1 | 1 |
Toxicity category for honey bee: High; LD50 ≤2 µg/bee = highly toxic; Mod; LD50 2–11 µg/bee = moderately toxic; minimum and maximum ranges of pesticides detected, LOD; limit of detection.
There was no statistical difference in total larval mortality between bees reared in control and treatment combs (26 and 33%, respectively; p = 0.059) (
Significance denoted with different letters.
(A) Normal larval development of bees reared in relatively uncontaminated control brood comb. (B) Larval development of bees reared in brood comb containing 17 different pesticides, expressing delayed development at day 4 and day 8. (C) Worker brood reared in brood comb containing 17 different pesticides at day 8 of development. Left: delayed growth. Right: normal development.
Brood mortality in bees reared from control comb was significantly greater on day 4 of development than on days 8, 12, and 19 (p = 0.0243; p = 0.0005; p<0.0001, respectively). In contrast, brood mortality in bees reared in treatment combs was not significantly different between days 4 and 8, although mortality was significantly higher on days 4 and 8 than on days 12 and 19 (p≤0.017 and p = 0.0001, respectively). The repeated use of experimental frames over several replicates may have allowed the migration of pesticide residues from treatment to control blocks, reducing the difference in residue levels between treatment and control combs and treatment effect differences (
Significance denoted with different letters.
Frame 1 | Frame 2 | Frame 3 | Frame 4 | Frame 5 | ||||||||||||||||||
Control | Treatment | Control | Treatment | Control | Treatment | Control | Treatment | Control | Treatment | |||||||||||||
Chemical | Class | before | before | before | before | before | before | before | before | before | before | |||||||||||
Boscalid | Carboxamide | F | 35 | |||||||||||||||||||
Chlorothalonil | Chloronitrile | F | 4 | 66 | ||||||||||||||||||
Cyprodinil | Anilinopyrimidine | F | 61 | |||||||||||||||||||
Fenhexamid | Hydroxyanilide | F | 46 | |||||||||||||||||||
Iprodione | Dicarboximide | F | 283 | |||||||||||||||||||
Myclobutanil | Dithiocarbamate | F | ||||||||||||||||||||
THPI | Phthalimide | F | 93 | 99 | ||||||||||||||||||
Oxyfluorfen | Diphenyl ether | H | 1 | 2 | ||||||||||||||||||
Diphenylamine | Amine | I | 281 | |||||||||||||||||||
Aldicarb | Carbamate | I | 20 | |||||||||||||||||||
Carbofuran | Carbamate | I | 32 | |||||||||||||||||||
Oxamyl | Carbamate | I | 22 | |||||||||||||||||||
Clothianidin | Neonicotinoid | I | 35 | |||||||||||||||||||
Dinotefuran | Neonicotinoid | I | 97 | |||||||||||||||||||
Imidacloprid | Neonicotinoid | I | 45 | |||||||||||||||||||
Thiacloprid | Neonicotinoid | I | 113 | |||||||||||||||||||
Thiamethoxam | Neonicotinoid | I | 38 | |||||||||||||||||||
Endosulfan 1 | Organochlorine | I | 1 | 1 | 2 | |||||||||||||||||
Endosulfan II | Organochlorine | I | 2 | |||||||||||||||||||
Endosulfan sulfate | Organochlorine | I | ||||||||||||||||||||
Chlorpyrifos | Organophosphate | I | 8 | 5 | 9 | |||||||||||||||||
Coumaphos | Organophosphate | I | 22100 | 21 | 281 | 3140 | 21 | 8200 | 21 | 7230 | ||||||||||||
Phosalone | Organophosphate | I | 32 | |||||||||||||||||||
Cyfluthrin | Pyrethroid | I | 79 | |||||||||||||||||||
Esfenvalerate | Pyrethroid | I | 12 | |||||||||||||||||||
Fluvalinate | Pyrethroid | I | 164 | 11280 | 24340 | 9850 | 6800 | |||||||||||||||
Permethrin total | Pyrethroid | I | 103 | |||||||||||||||||||
Pyrethrins | Pyrethroid | I | 229 | |||||||||||||||||||
Paradichlorobenzene | Halogenated organic | I | ||||||||||||||||||||
2,4 Dimethylphenyl formamide (DMPF) | Amidine | m | 147 | 142 | ||||||||||||||||||
3-hydroxycarbofuran | Carbamate | m | 23 | |||||||||||||||||||
Chlorferone | Organophosphate | m | ||||||||||||||||||||
Coumaphos oxon | Organophosphate | m | 1850 | 10 | 3140 | 474 | 231 | |||||||||||||||
Malathion oxon | Organophosphate | m | 22 | |||||||||||||||||||
Results reported in ng/g or parts per billion. (F = fungicide; H = herbicide; I = insecticide; m = metabolite).
Comparisons of chemical analyses, performed on five paired control and treatment combs before and after the experiment (n = 10), confirmed pesticide residue transfer and contamination of control combs over a 3-month period. Four additional new pesticide residues were detected in control comb, on average, compared to a reduction of 3 pesticide residues in treatment combs after the experiment (
Worker bees reared in relatively uncontaminated brood comb lived an average of 4 days longer than bees reared in comb containing high levels of pesticide residues (
(A) Percent emergence and survivorship of caged control and treatment bees over 50 days. (B) Average adult longevity of caged control and treatment bees. Adult bees reared in control combs lived an average of 4 days longer than adult bees reared in combs containing high levels of pesticides (p = 0.005).
(A) First brood cycle emergence (Rep 1). (B) Second brood cycle emergence (Rep 2). (C) Third brood cycle emergence (Rep 3). Different capital letters denote significant differences in emergence of control bees on different days; different lower case letters denote significant differences in emergence of treatment bees on different days.
Honey bees of all ages and castes are susceptible to effects from pesticide exposure
In this study, worker bees reared in comb containing high levels of pesticide residues had lower survivorship than bees reared in relatively uncontaminated comb. Comb age may have been a factor as well, given that brood mortality was higher in newly drawn control comb than in older control comb sampled from feral colonies. Newly drawn comb lacks exuviae (molted larval cuticles), which contain brood pheromone cues that indicate brood presence to nurse bees and increase larval survivorship
For economic reasons, beekeepers typically reuse wax foundation, but pesticide residues accumulate in wax and may persist for years
Sub-lethal effects of pesticides on bees, including delayed adult emergence, may seem inconsequential but may provide a reproductive advantage for
Pesticides | Chemical family | Systemic | Toxicity honey bee | (ng/g) ppb | LOD |
3-hydroxy-carbofuran | metabolite | Systemic | 23 | 4 | |
Aldicarb | Carbamate | Systemic | High | 20 | 4 |
Carbofuran | Carbamate | Systemic | High | 32 | 5 |
Chlorothalonil | Fungicide | --- | 4 | 1 | |
Clothianidin | Neonicotinoid | Systemic | High | 35 | 20 |
Coumaphos | Organophosphate | Moderate | 22100 | 1 | |
Coumaphos oxon | metabolite | 1850 | 5 | ||
Cyfluthrin | Pyrethroid | High | 7.9 | 2 | |
Dinotefuran | Neonicotinoid | Systemic | High | 97 | 30 |
Diphenylamine | Amine | --- | 281 | 1 | |
Endosulfan 1 | Organochlorine | Moderate | 1 | 1 | |
Fluvalinate | Pyrethroid | High | 164 | 1 | |
Imidacloprid | Neonicotinoid | Systemic | High | 45 | 20 |
Malathion Oxon | metabolite | 22 | 4 | ||
Oxamyl | Carbamate | Systemic | High | 22 | 5 |
Thiacloprid | Neonicotinoid | Systemic | High | 113 | 8 |
Thiamethoxam | Neonicotinoid | Systemic | High | 38 | 20 |
Toxicity category for honey bee: High; LD50 ≤2 µg/bee = highly toxic; Mod; LD50 2–11 µg/bee = moderately toxic; LOD; limit of detection.
Worker bees reared in treatment comb containing high levels of pesticide residues lived an average of 4 days less than bees reared in relatively uncontaminated control combs in cage trials (
Combined effects from honey bee exposure to pesticide residue in brood comb, such as reduced adult longevity, increased brood mortality, higher fecundity of
Honey bees are biological indicators, picking up chemicals and other pollutants from their environment both external and internal to their hives. Our findings suggest that one of the underlying commonalities in the worldwide reports of a decline in honey bee health and observations of Colony Collapse Disorder (CCD) may be exposure of honey bees and bee products to pesticides. Developmental exposure of honey bees to pesticide contaminated brood comb may appear subtle and indirect, but can lead to sub-lethal effects that actually have serious consequences.
We thank E. Olson and numerous other beekeepers, and J. Pettis (USDA ARS) for providing comb samples, R. Simonds from the USDA-AMS-National Science Laboratory in Gastonia, NC for chemical residues analysis, M. Evans (WSU) for assistance with statistics, and J. Stark (WSU) for comments that improved the manuscript.