Next Article in Journal
Distribution, Site-Specific Water Quality Criteria, and Ecological Risk Assessment of Heavy Metals in Surface Water in Fen River, China
Next Article in Special Issue
Olfactory Learning Behavior and Mortality of the Honey Bee Apis mellifera jemenitica in Response to Pyrethroid Insecticide (Deltamethrin)
Previous Article in Journal
Indoor Environmental Quality (IEQ): A Comparison between TOPSIS- and PROMETHEE-Based Approaches for Indirect Eliciting of Category Weights
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Toxics
Volume 11
Issue 8
10.3390/toxics11080703
toxics-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Availability of Using Honeybees as Bioindicators of Pesticide Exposure in the Vicinity of Agricultural Environments in Taiwan

by
Chien-Che Hung
and
Lih-Ming Yiin
*
Department of Public Health, Tzu Chi University, 701, Sec. 3, Zhongyang Road, Hualien City 970374, Taiwan
*
Author to whom correspondence should be addressed.
Toxics 2023, 11(8), 703; https://doi.org/10.3390/toxics11080703
Submission received: 10 July 2023 / Revised: 10 August 2023 / Accepted: 11 August 2023 / Published: 15 August 2023
(This article belongs to the Special Issue Assessment of Pesticide Exposure and Toxicity: Honeybee as Indicator)
Browse Figures
Versions Notes

Abstract

:
While pollinating, honeybees are subject to exposure to a variety of pesticides; with their characteristics of certain foraging distances, they could serve as bioindicators of pesticide exposure in a neighborhood. We conducted a study to assess availability by collecting and analyzing bee samples from 15 apiaries located in East Taiwan and dust samples from the adjacent environment, and by finding relations between both samples. Seventeen pesticides were selected for the analysis using gas or liquid chromatography coupled with mass spectrometry, and eight (three insecticides, two herbicides, and three fungicides) were more frequently detected from bee or dust samples; the levels of these pesticides were mostly under 1000 ng/g. Significant correlation results (r ≅ 0.8) between residue concentrations in bees and in dust suggest that honeybees could be a good bioindicator for exposure to herbicides and fungicides within certain ranges. The pesticide contents of sick/dead bees were much higher than those of healthy counterparts regarding any pesticide type, with the mean total concentrations of 635 ng/g and 176 ng/g, respectively. We conclude that honeybees could be used as bioindicators of pesticide exposure; sick/dead bees could serve as a warning sign of the severity of pesticide pollution.

1. Introduction

Pesticides are a large group of chemicals or biological agents that control pests for agricultural, gardening, and environmental purposes. Because of the impact of climate change on insect pest breeding, doses and varieties of pesticides are increasingly applied in the present time [1,2], and thus the overuse of pesticides becomes a crucial problem for human and environmental health [3,4]. Many recent studies have confirmed that pesticides applied in agricultural areas may drift over to nearby residential areas [5,6,7,8] and even cause a health concern [9]. In those studies, measuring pesticide residues in dust collected from residential homes adjacent to the agricultural areas was commonly practiced to provide exposure information at the receptor sites; the assessment of potential pesticide exposure from the sources (i.e., agricultural areas with pesticide application), however, needs a sampling approach that gathers information more accurately than house-dust sampling does.
A bioindicator is an organism carrying certain indicative information that is helpful to evaluate the health of an environment or ecosystem. It can be any species, including algae, fungi, plants, animals (sentinel species), and insects, whose population, status, or function (physiological or biological) represents the qualitative status of the environment [10,11]. For the assessment of exposure to pesticides originating from agricultural activities, honeybees (Apis mellifera) are considered as an available bioindicator, because they possess various sensitivities to pesticides and the characteristics of limited foraging ranges and little behavioral change due to seasonal changes [12]. There is a history in which honeybees have been used for the biomonitoring of environmental pollutants, including pesticides, heavy metals, radionuclides, and others [13]. An example showing the sensitivity of honeybees to pesticides is Colony Collapse Disorder (CCD) that has been occurring for more than a decade. CCD caused a majority of worker bees to disappear [14], and was later confirmed to be partially attributed to exposure to neonicotinoids, a relatively new type of insecticides [15,16,17]. With the foraging nature that likely translates to contact with pesticides, honeybees used as bioindicators could provide useful information on pesticide exposure caused with agricultural activities. In addition, richness for sampling and convenience for hive mobilization, as desired, make farm-raised honeybees an excellent choice over other wild pollinators for the role of bioindicator.
There have been a number of studies investigating the relationships between bees and pesticides. It was found that insecticides are a lethal threat, whereas herbicides and fungicides could cause a variety of health effects on bees; synergistic effects are even observed when mixtures of various pesticides are applied [18,19,20,21,22,23,24]. Other studies demonstrate significant pesticide exposure for bees, indicating the prevalent use of pesticides, even though several bees’ main activity regions are not intended for pesticide use [25,26,27]. Another study using a geographic information system indicates a strong positive correlation between glyphosate residue concentrations in honey and agricultural land use, suggesting that bee products could be used as indicators for pesticide exposure [28]. Since honey is produced by honeybees, the pesticide residue in honey supposedly originates from bees, especially worker bees; thus, honeybees should work as bioindicators of pesticide exposure as functionally as honey does.
Pesticide use in Taiwan has been one of the highest in the world for recent years [29]. Most people in Taiwan are subject to pesticide exposure because of no significant boundaries between residential and agricultural areas; therefore, an indicator of pesticide exposure around the living places is very much needed. Although honeybees have been used for biomonitoring for years, little is carried out for verifying whether they can be used for exposure assessment. This study is to evaluate the availability of using honeybees as bioindicators of pesticide exposure in the neighborhood of agricultural areas in Taiwan. We expected this approach to derive information from honeybees to help construct the exposure model for communities around the agricultural areas.

2. Materials and Methods

2.1. Participating Apiaries and Sample Collection

This study was conducted in Hualien County, a rural area located in East Taiwan, from October 2020 to October 2022. Because of the sub-tropic and tropic climate in the region, the apiaries operated all-year-round regardless of seasons. There were 15 apiary teams within the area, which were willing to collaborate with this study and allowed the study team to take necessary samples during the honey collecting periods. These apiaries were adjacent to fruit gardens and/or betel nut groves; 1 of the 15 apiaries had a site in an experimental forestry station to the south of Hualien, which was used as a blank because of no pesticide applications (Figure 1).
At each apiary, we collected worker bee and honey samples from half of the hive boxes. Healthy and sick/dead bees inside and outside the hive boxes were collected in 50 mL centrifuge tubes using a pair of Teflon® tweezers, respectively. Approximately 15~20 bees of each type (i.e., healthy and sick/dead) were placed in a centrifuge tube and analyzed as a composite sample. Honey was collected and stored in a 50 mL centrifuge tube using a Teflon® spoon or a disposable dropper. Bee samples were taken as planned, whereas honey was collected upon availability. On average, 3~5 composite samples of bees (healthy or sick/dead) were collected from each apiary. Environmental dust was collected with a changeable-bag vacuum system (Makita DVC261ZX18, Makita Co., Aichi, Japan) up to 10~15 g from public places in the neighborhood around an apiary within a radius of 300, 1000, or 5000 m. Prior to the analysis, dust samples were transferred into brown glass bottles for storage at room temperature with relative humidity controlled below 60%, while bee and honey samples were temporarily stored in a 4 °C portable refrigerator on site and transferred to an ultra-low temperature freezer at −80 °C in a laboratory.

2.2. Sample Treatment and Analysis

We selected a range of pesticides that were known to be commonly sold in Taiwan to be the analytes of interest. They were acetamiprid, carbaryl, carbofuran, chlorpyrifos, cypermethrin, fipronil, imidacloprid, permethrin, prallethrin, and tetramethrin for insecticides, glufosinate, glyphosate, and paraquat for herbicides, and benomyl, chlorothalonil, mancozeb, and propineb for fungicides. We followed and modified analytic methods that were previously published for sample treatment and an analysis using gas-chromatography–mass-spectrometry (GC-MS, Agilent 6890/5973, Agilent Technologies, Santa Clara, CA, USA) or liquid-chromatography–mass-spectrometry (LC-MS, Agilent 1200/6460A, Agilent Technologies, Santa Clara, CA, USA) [30,31]. The information of target pesticides of this study is listed in Table 1.
Bee samples were first weighed, homogenized, and added to 50 mL of ethyl acetate for 30 min ultrasonic extraction; the mixture was then centrifuged at 5000 rpm for 10 min. The supernatant was concentrated, reconstituted with 500 μL of n-hexane, filtered, and transferred to an insert vial for the GC-MS or LC-MS analysis. For dust sample treatment, three grams of dust from each sample was first weighed, and the remaining procedures were similar to the treatment of bee samples, which were processes of extraction, concentration, reconstitution with n-hexane, filtration, and the analysis using GC-MS or LC-MS. Honey samples were diluted with deionized water with the ratio of 1:0.3~1 (honey/water), depending on the viscosity. The diluted sample solutions were introduced to C18 cartridges (Strata C18 SPE cartridge, Phenomenex, Torrance, CA, USA) with pressure for solid-phase extraction. The cartridges were washed with deionized water and dried with flushing nitrogen; the eluate was then eluted with n-hexane, and processed for the chromatographic analysis after concentration and solvent reconstitution.
The GC-MS system used in this study was the same as reported previously [5], with different settings regarding the flow rate, split mode, and temperature profile. Ultra-purity helium (99.9995%) was used as a carrier gas with the constant flow rate set at 1.0 mL/min; the inlet condition was set at 280 °C with a split model (30:1). The oven temperature began at 60 °C, and was increased to 160 °C at a rate of 20 °C/min, to 300 °C at 10 °C/min, and held at 300 °C for 8 min (total runtime: 27 min). The LC-MS system was equipped with a C18 column (Agilent Poroshell 120 EC-C18, 3.0 × 100 mm, 2.7 μm, Agilent Technologies, Santa Clara, CA, USA), and deionized water and methanol, with both containing 10 mM of ammonium acetate, were used for gradient separation. Twenty percent methanol was used for elution in the beginning, and the percentage increased linearly to 50% at 10 min, to 70% at 13.5 min, to 71% at 20 min, and to 100% at 29 min; the elution continued with 100% methanol until 35 min. The limits of detection (LODs) for all analytes using GC-MS or LC-MS were determined to be 1 ng/g or lower. The recovery rates were between 94% and 99% by spiking samples with standards; the coefficients of variance of the overall analysis were lower than 5%.

2.3. Data Management and Statistical Analysis

The correlation analysis was used to examine the association between pesticide concentrations in bee samples and those in environmental dust samples. Because not every pesticide in a bee or dust sample was detectable, we summarized the detected concentrations by pesticide type (i.e., insecticide, herbicide, and fungicide) for an individual sample, and used them for the correlation analysis. The concentration of each sample that was undetectable was replaced by half the LOD. A general linear model was employed for the two-way analysis of variance (ANOVA) to examine the difference in pesticide residue in healthy and sick/dead bees, and whether the distribution patterns of insecticides, herbicides, and fungicides were different between these two types of bees. All statistical analyses were performed using the SPSS statistical software package version 23.0 (SPSS Inc., Chicago, IL, USA, 2015).

3. Results

We collected a total of 112 composite samples of bees (healthy, 52; sick/dead, 60), 28 honey samples, and 180 environmental dust samples. Among the 17 target pesticides, 8 were frequently detected from bee or dust samples, whereas nearly nothing was detected from honey (Table 2). Bee or dust samples collected from the blank site were all under LODs, supporting the quality control of the sample analysis. The detection rates of insecticides (>40%) or fungicides (>20%) for bees and dust were similar, except those of herbicides were low for bee samples (<33%) and high for dust samples (>52%). Among the types of pesticides, data from bee and dust samples indicate that fungicides were less abundant than the others in this study (i.e., lower detection rates and median concentrations). For the comparison of pesticide concentrations, it appears that herbicides and fungicides were richer in dust than in bees, whereas insecticides did not quite follow this pattern. Most of the pesticide levels were under 1000 ng/g, except the maximum values of certain pesticides in dust.
To evaluate the availability of using honeybees as bioindicators, we conducted a correlation analysis between pesticide concentrations in bees and those in environmental dust collected from various distances surrounding the apiaries (Table 3). As mentioned previously, the sums of pesticide concentrations of the same types were used in the analysis to avoid skewed results due to non-detects. All except correlation for insecticides in the near or middle range were significant (p < 0.05), indicating the associations between pesticide residue concentrations in honeybees and in dust of the nearby environment, especially those for herbicides and fungicides within the near and middle ranges (r ≅ 0.8). All three types of pesticides yielded significant correlation results in the far range, albeit merely fair, suggesting that the foraging distances of honeybees could reach up to 5000 m. The correlation data further confirm that using honeybees as bioindicators is feasible.
We further analyzed the data of pesticides in bees with the health status, and found significant differences in the number of total detected pesticides per bee and total pesticide concentration between healthy and sick/dead bees (0.79 pesticides/bee and 176 ng/g, and 1.74 pesticides/bee and 635 ng/g, respectively; p < 0.001) (Figure 2). The average number of detected pesticides of each type per bee is at least two times higher for sick/dead bees than healthy counterparts, except that of fungicide is 1.9 times higher (0.45/0.24). Similarly, the mean concentration of each pesticide type is higher in sick/dead bees than healthy ones, but the differences among the types are not quite similar. The levels of insecticides and herbicides of sick/dead bees are at least 3.6 times higher than those of healthy ones, whereas the concentration difference between healthy and sick/dead bees for fungicides is merely 2.2-fold. Given the abovementioned data, it appears that sick/dead bees might have been exposed to much more insecticides and herbicides than fungicides. The ANOVA model further confirmed the difference between pesticide distribution patterns of healthy and sick/dead bees by showing a significance level of 0.001.

4. Discussion

Compared with previous studies [25,26], the detection rates and concentrations of pesticides in bees in this study were relatively high; the mean detection rate was roughly 31%, and the median concentrations of insecticides and herbicides were above 180 ng/g (Table 2). One study that investigated pesticide exposure for wild bees and butterflies in the cultivated agricultural areas located in mid-northern Missouri, USA found that most of the total pesticide concentrations in bees were under 100 ng/g, with the average detection rate being around 25% [25]. Another study attempted to find out whether border plantings could be pesticide sources for bees, and they detected several common pesticides from over 25% of honeybee samples with the median concentration of each pesticide type being around 100 ng/g [26]. The reason that this study yielded relatively high pesticide levels could be that the inclusion of sick/dead bees elevated the overall means (Figure 2). For those two previous studies, bee samples were collected on site with certain nets or traps, and thus those bees were presumably healthy. As for our healthy bees’ data (Figure 2), the levels were relatively low and seemingly of no substantial difference from those of other studies, despite exposure to different types of insecticides, herbicides, and fungicides. Thus, in terms of healthy bee sampling, the pesticide exposure found in this study was not particularly higher than that reported by others; with the inclusion of sick/dead bees, however, the exposure profile became different. Although the quantity of sick/dead bees might be relatively small, they represented the high end of pesticide exposure, which was valuable and necessary for a complete assessment of pesticide exposure. If a significant number of dead/sick bees is observed with hive boxes, that may indicate ongoing pesticide pollution, to which attention should be paid.
Sick/dead bees resulted in not only an increase in pesticide concentration, but also an enhancement in detection rates, which was beneficial to the quantitative analysis. It is of no surprise to see higher insecticide levels in sick/dead bees than in healthy bees (~6 times), because insecticides are lethal and definitely a cause of death. The levels of herbicides and fungicides are also elevated and could be the causes of sickness, considering that both types of pesticides are less toxic than insecticides. Given the fact that the mean concentration of herbicides was much higher than that of fungicides in sick/dead bees (286 ng/g vs. 155 ng/g, Figure 2), the sick bees could be mostly attributed to exposure to herbicides. It is also likely that the death and sickness of bees could have been caused by the synergistic effect of mixtures of a variety of pesticides, with 1.74 pesticides detected from a sick/dead bee on average [23,24].
The correlations between herbicides/fungicides in bees and in environmental dust were significantly strong (r ≅ 0.8) within the near and middle ranges (<1000 m), suggesting that bees could be used as bioindicators of exposure to herbicides or fungicides in a vicinity as a result of pesticide drift. Of a certain surprise, no significant correlation was, however, found for insecticides within the same ranges. One of the reasons may have been the toxicity of pesticides of different types. As mentioned in the previous paragraph, insecticides are lethal and probably acute, and herbicides and fungicides are less toxic and chronic; thus, there must have been a certain number of honeybees dying outside the apiary after exposure to insecticides, which would not be accounted for in samples. Honeybees that were exposed to herbicides and/or fungicides could return to an apiary and serve as samples. Consequently, the strong correlations for herbicides and fungicides may have resulted from sufficient bee samples, while the poor correlations for insecticides may have resulted from insufficient bee samples.
The detection rates for honey samples were relatively low, which were 9 detects out of 224 (8 pesticides × 28 samples); in addition, the maximum values were far lower than those of bee or dust samples. Similarly, low levels of pesticides in honey were reported in previous studies [32,33], which analyzed bee products for fungicides and always found lower residue concentrations in honey than in pollen. The reason for low residue concentrations in honey is likely due to detoxification with enzymes. As honey is produced with a series of a worker bee’s tasks, including flower nectar suction and regurgitation among bees, detoxifying enzymes inside the bodies (e.g., 450 s), albeit relatively few [34,35], could metabolize pesticides in the nectar (later turned into honey) to a certain extent. A similar metabolism of pesticides may be driven by microbes in the bee gut. The majority of the honey samples under detection in this study are a good sign in the perspective of food safety, but few detects may still be of concern for approaching or exceeding the maximum residue levels.
It appears that herbicides are prevalently used in modern agriculture, as shown with the bee and dust data. Paraquat, one of the detected herbicides, has been banned for use in Taiwan since February 2020, but is still detected from bee and dust samples. The degradation mechanisms of paraquat are considered to be slow [36], but the water solubility is high [37], suggesting that paraquat would not degrade soon but could be easily washed out with rain. Therefore, paraquat detected in this study should more likely originate from recent applications after the ban than from remaining residues of paraquat use in the past. Source management for paraquat is recommended for an effective control.
There are limitations in the study. First of all, we observed and used dead/sick bees as samples, but had difficulty knowing about their percentage of the whole bee population. Had we had this information, the weighting of healthy bees’ and sick/dead bees’ data with that should have helped construct a fit model of pesticide exposure for bees. This is the information that we can strive to collect in future work. Secondly, we may have lost honeybee samples to any possible ongoing CCD, of which we were never aware. Given the fact that imidacloprid is the major CCD affecting insecticides [38] and our result of the detection of imidacloprid from samples is rare, this limitation is considered minimal. Thirdly, dust sampling was subject to location accessibility (e.g., private lands), and the sample collection could not be as complete as bee sample collection; thus, such a reduction in sample size could lead to low statistical significance. Fortunately, most of the correlation results are significant, meeting the goal of the study. Speaking of limited accessibility to certain areas, honeybees, once confirmed to serve in the role of bioindicator, could be useful to overcome problems of limited access to certain sites.

5. Conclusions

We conclude that honeybees could be used as bioindicators of exposure to pesticides in an adjacent environment, especially herbicides and fungicides. Honey, as a bee product, was not suitable for use as a bioindicator, because of the low detection rates. Sick/dead bees contained more contents of insecticides, herbicides, and fungicides than healthy bees did, suggesting that an observation of a significant number of sick/dead bees could be a warning sign of the severity of pesticide pollution.

Author Contributions

Conceptualization, C.-C.H. and L.-M.Y.; methodology, C.-C.H.; software, C.-C.H.; validation, C.-C.H. and L.-M.Y.; formal analysis, C.-C.H.; investigation, L.-M.Y.; resources, L.-M.Y.; data curation, C.-C.H.; writing—original draft preparation, C.-C.H.; writing—review and editing, L.-M.Y.; visualization, C.-C.H.; supervision, L.-M.Y.; project administration, C.-C.H.; funding acquisition, L.-M.Y. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Science and Technology Council of Taiwan, grant number: MOST109-2314-B-320-002-MY2. The APC was funded by a supplement grant of Tzu Chi University (610400184-11).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

Acknowledgments

We want to thank all participating apiaries for collaborating with the research work.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Chen, C.-C.; McCarl, B.A. An investigation of the relationship between pesticide usage and climate change. Clim. Chang. 2001, 50, 475–487. [Google Scholar] [CrossRef]
  2. Delcour, I.; Spanoghe, P.; Uyttendaele, M. Literature review: Impact of climate change on pesticide use. Food Res. Int. 2015, 68, 7–15. [Google Scholar] [CrossRef]
  3. Jurewicz, J.; Hanke, W. Prenatal and childhood exposure to pesticides and neurobehavioral development: Review of epidemiological studies. Int. J. Occup. Med. Environ. Health 2008, 21, 121–132. [Google Scholar] [CrossRef]
  4. Spanoghe, P.; Maes, A.; Steurbaut, W. Limitation of point source pesticide pollution: Results of bioremediation system. Commun. Agric. Appl. Biol. Sci. 2004, 69, 719–732. [Google Scholar]
  5. Simaremare, S.R.S.; Hung, C.-C.; Yu, T.-H.; Hsieh, C.-J.; Yiin, L.-M. Association between Pesticides in House Dust and Residential Proximity to Farmland in a Rural Region of Taiwan. Toxics 2021, 9, 11. [Google Scholar] [CrossRef] [PubMed]
  6. Lu, C.; Fenske, R.A.; Simcox, N.J.; Kalman, D. Pesticide exposure of children in an agricultural community: Evidence of household proximity to farmland and take home exposure pathways. Environ. Res. 2000, 84, 290–302. [Google Scholar] [PubMed]
  7. Gunier, R.; Riggs, P.; Ward, M.; Reynolds, P.; Rull, R.; Hertz, A.; Colt, J.; Metayer, C.; Nishioka, M.; Buffler, P. Proximity to agricultural pesticide use and concentrations in carpet dust. Epidemiology 2008, 19, S269–S270. [Google Scholar]
  8. Béranger, R.; Billoir, E.; Nuckols, J.R.; Blain, J.; Millet, M.; Bayle, M.-L.; Combourieu, B.; Philip, T.; Schüz, J.; Fervers, B. Agricultural and domestic pesticides in house dust from different agricultural areas in France. Environ. Sci. Pollut. Res. 2019, 26, 19632–19645. [Google Scholar] [CrossRef] [PubMed]
  9. Lee, S.J.; Mehler, L.; Beckman, J.; Diebolt-Brown, B.; Prado, J.; Lackovic, M.; Waltz, J.; Mulay, P.; Schwartz, A.; Mitchell, Y.; et al. Acute Pesticide Illnesses Associated with Off-Target Pesticide Drift from Agricultural Applications: 11 States, 1998–2006. Environ. Health Perspect. 2011, 119, 1162–1169. [Google Scholar]
  10. Siddig, A.A.H.; Ellison, A.M.; Ochs, A.; Villar-Leeman, C.; Lau, M.K. How do ecologists select and use indicator species to monitor ecological change? Insights from 14 years of publication in Ecological Indicators. Ecol. Indic. 2016, 60, 223–230. [Google Scholar] [CrossRef]
  11. Burger, J. Bioindicators: A review of their use in the environmental literature 1970–2005. Environ. Bioindic. 2006, 1, 136–144. [Google Scholar] [CrossRef]
  12. Porrini, C.; Sabatini, A.G.; Girotti, S.; Fini, F.; Monaco, L.; Celli, G.; Bortolotti, L.; Ghini, S. The death of honey bees and environmental pollution by pesticides: The honey bees as biological indicators. Bull. Insectol. 2003, 56, 147–152. [Google Scholar]
  13. Girotti, S.; Ghini, S.; Ferri, E.; Bolelli, L.; Colombo, R.; Serra, G.; Porrini, C.; Sangiorgi, S. Bioindicators and biomonitoring: Honeybees and hive products as pollution impact assessment tools for the Mediterranean area. Euro-Mediterr. J. Environ. Integr. 2020, 5, 62. [Google Scholar] [CrossRef]
  14. Stokstad, E. The case of the empty hives. Science 2007, 316, 970–972. [Google Scholar] [CrossRef] [PubMed]
  15. Brandt, A.; Gorenflo, A.; Siede, R.; Meixner, M.; Büchler, R. The neonicotinoids thiacloprid, imidacloprid, and clothianidin affect the immunocompetence of honey bees (Apis mellifera L.). J. Insect. Physiol. 2016, 86, 40–47. [Google Scholar] [CrossRef] [PubMed]
  16. Lu, C.; Hung, Y.-T.; Cheng, Q. A review of sub-lethal neonicotinoid insecticides exposure and effects on pollinators. Curr. Pollut. Rep. 2020, 6, 137–151. [Google Scholar] [CrossRef]
  17. Furlan, L.; Pozzebon, A.; Duso, C.; Simon-Delso, N.; Sánchez-Bayo, F.; Marchand, P.A.; Codato, F.; Bijleveld van Lexmond, M.; Bonmatin, J.M. An update of the Worldwide Integrated Assessment (WIA) on systemic insecticides. Part 3: Alternatives to systemic insecticides. Environ. Sci. Pollut. Res. 2021, 28, 11798–11820. [Google Scholar]
  18. Sánchez-Bayo, F.; Goulson, D.; Pennacchio, F.; Nazzi, F.; Goka, K.; Desneux, N. Are bee diseases linked to pesticides?—A brief review. Environ. Int. 2016, 89, 7–11. [Google Scholar] [CrossRef] [PubMed]
  19. Almasri, H.; Tavares, D.A.; Pioz, M.; Sené, D.; Tchamitchian, S.; Cousin, M.; Brunet, J.-L.; Belzunces, L.P. Mixtures of an insecticide, a fungicide and a herbicide induce high toxicities and systemic physiological disturbances in winter Apis mellifera honey bees. Ecotoxicol. Environ. Saf. 2020, 203, 111013. [Google Scholar] [CrossRef]
  20. Ulziibayar, D.; Jung, C. Comparison of acute toxicity of different groups of pesticides to honey bee workers (Apis mellifera L.). J. Apic. 2019, 34, 305–313. [Google Scholar] [CrossRef]
  21. Belsky, J.; Joshi, N.K. Effects of fungicide and herbicide chemical exposure on Apis and non-Apis bees in agricultural landscape. Front. Environ. Sci. 2020, 8, 81. [Google Scholar] [CrossRef]
  22. Farruggia, F.T.; Garber, K.; Hartless, C.; Jones, K.; Kyle, L.; Mastrota, N.; Milone, J.P.; Sankula, S.; Sappington, K.; Stebbins, K. A retrospective analysis of honey bee (Apis mellifera) pesticide toxicity data. PLoS ONE 2022, 17, e0265962. [Google Scholar] [CrossRef] [PubMed]
  23. Migdal, P.; Murawska, A.; Berbeć, E.; Plotnik, M.; Skorus, A.; Latarowski, K. Selected Biochemical Markers Change after Oral Administration of Pesticide Mixtures in Honey Bees. Toxics 2022, 10, 590. [Google Scholar] [CrossRef]
  24. Zawislak, J.; Adamczyk, J.; Johnson, D.R.; Lorenz, G.; Black, J.; Hornsby, Q.; Stewart, S.D.; Joshi, N. Comprehensive survey of area-wide agricultural pesticide use in southern United States row crops and potential impact on honey bee colonies. Insects 2019, 10, 280. [Google Scholar] [CrossRef]
  25. Main, A.R.; Hladik, M.L.; Webb, E.B.; Goyne, K.W.; Mengel, D. Beyond neonicotinoids–Wild pollinators are exposed to a range of pesticides while foraging in agroecosystems. Sci. Total Environ. 2020, 742, 140436. [Google Scholar] [CrossRef] [PubMed]
  26. Ward, L.T.; Hladik, M.L.; Guzman, A.; Winsemius, S.; Bautista, A.; Kremen, C.; Mills, N.J. Pesticide exposure of wild bees and honey bees foraging from field border flowers in intensively managed agriculture areas. Sci. Total Environ. 2022, 831, 154697. [Google Scholar] [CrossRef]
  27. Ostiguy, N.; Drummond, F.A.; Aronstein, K.; Eitzer, B.; Ellis, J.D.; Spivak, M.; Sheppard, W.S. Honey bee exposure to pesticides: A four-year nationwide study. Insects 2019, 10, 13. [Google Scholar] [CrossRef] [PubMed]
  28. Berg, C.J.; King, H.P.; Delenstarr, G.; Kumar, R.; Rubio, F.; Glaze, T. Glyphosate residue concentrations in honey attributed through geospatial analysis to proximity of large-scale agriculture and transfer off-site by bees. PLoS ONE 2018, 13, e0198876. [Google Scholar] [CrossRef] [PubMed]
  29. Hung, C.-C.; Huang, F.-J.; Yang, Y.-Q.; Hsieh, C.-J.; Tseng, C.-C.; Yiin, L.-M. Pesticides in indoor and outdoor residential dust: A pilot study in a rural county of Taiwan. Environ. Sci. Pollut. Res. 2018, 25, 23349–23356. [Google Scholar] [CrossRef]
  30. Żaneta Bargańska, M.S.; Namieśnik, J. Determination of pesticide residues in honeybees using modified QUEChERS sample work-up and liquid chromatography-tandem mass spectrometry. Molecules 2014, 19, 14. [Google Scholar] [CrossRef]
  31. Fernández, M.; Picó, Y.; Manes, J. Analytical methods for pesticide residue determination in bee products. J. Food Prot. 2002, 65, 1502–1511. [Google Scholar] [CrossRef] [PubMed]
  32. Kubik, M.; Nowacki, J.; Pidek, A.; Warakomska, Z.; Michalczuk, L.; Goszczyñski, W. Pesticide residues in bee products collected from cherry trees protected during blooming period with contact and systemic fungicides. Apidologie 1999, 30, 521–532. [Google Scholar] [CrossRef]
  33. Kubik, M.; Nowacki, J.; Pidek, A.; Warakomska, Z.; Michalczuk, L.; Goszczyñski, W.; Dwużpnik, B. Residues of captan (contact) and difenoconazole (systemic) fungicides in bee products from an apple orchard. Apidologie 2000, 31, 531–541. [Google Scholar] [CrossRef]
  34. Claudianos, C.; Ranson, H.; Johnson, R.; Biswas, S.; Schuler, M.; Berenbaum, M.; Feyereisen, R.; Oakeshott, J.G. A deficit of detoxification enzymes: Pesticide sensitivity and environmental response in the honeybee. Insect Mol. Biol. 2006, 15, 615–636. [Google Scholar] [CrossRef]
  35. Gong, Y.; Diao, Q. Current knowledge of detoxification mechanisms of xenobiotic in honey bees. Ecotoxicology 2017, 26, 1–12. [Google Scholar] [CrossRef]
  36. Sartori, F.; Vidrio, E. Environmental fate and ecotoxicology of paraquat: A California perspective. Toxicol. Environ. Chem. 2018, 100, 479–517. [Google Scholar] [CrossRef]
  37. Lock, E.A.; Wilks, M.F. Paraquat. In Hayes’ Handbook of Pesticide Toxicology, 3rd ed.; Krieger, R., Ed.; Elsevier: Amsterdam, The Netherlands, 2010; Volume 1, pp. 1771–1827. [Google Scholar]
  38. Halm, M.-P.; Rortais, A.; Arnold, G.; Taséi, J.; Rault, S. New risk assessment approach for systemic insecticides: The case of honey bees and imidacloprid (Gaucho). Environ. Sci. Technol. 2006, 40, 2448–2454. [Google Scholar] [CrossRef]
Toxics 11 00703 g001
Figure 1. Location of apiaries in East Taiwan.
Figure 1. Location of apiaries in East Taiwan.
Toxics 11 00703 g001
Toxics 11 00703 g002
Figure 2. Pesticide detection and levels of healthy and sick/dead bees. (The number above each box denotes the mean number of detected pesticides per bee. The line and “X” in the box depict median and mean values, respectively).
Figure 2. Pesticide detection and levels of healthy and sick/dead bees. (The number above each box denotes the mean number of detected pesticides per bee. The line and “X” in the box depict median and mean values, respectively).
Toxics 11 00703 g002
Table 1. General information of target pesticides.
Table 1. General information of target pesticides.
Type and NameAnalytical MethodRetention Time (min)Main Use for Agriculture, Environmental Sanitation, or Both
Insecticide
 AcetamipridLC-MS2.31Agriculture
 CarbarylGC-MS27.21Agriculture
 CarbofuranGC-MS23.92Agriculture
 ChlorpyrifosGC-MS8.62Both
 CypermethrinGC-MS22.71Agriculture
 FipronilGC-MS6.31Both
 ImidaclopridLC-MS1.93Both
 PermethrinGC-MS21.92Both
 PrallethrinGC-MS21.43Environmental sanitation
 TetramethrinGC-MS20.51Environmental sanitation
Herbicide
 GlufosinateLC-MS1.39Agriculture
 GlyphosateLC-MS1.13Agriculture
 ParaquatLC-MS3.29Agriculture
Fungicide
 BenomylLC-MS5.41Agriculture
 ChlorothalonilLC-MS7.43Agriculture
 MancozebGC-MS11.76Agriculture
 PropinebGC-MS12.44Agriculture
Table 2. Pesticide concentrations by type in different sample media.
Table 2. Pesticide concentrations by type in different sample media.
Sample TypePesticideDF (%)50th Percentile
(ng/g)		75th Percentile
(ng/g)		Maximum
(ng/g)
Bee
(n = 112)		Insecticide
Carbaryl50.30114.18126.25726.58
Carbofuran44.68207.63254.31823.22
Chlorpyrifos41.03366.21403.64331.35
Herbicide
Glyphosate32.07231.26247.11523.26
Paraquat16.11184.32246.31677.85
Fungicide
Benomyl21.5871.1973.39763.09
Mancozeb22.19126.76215.33449.11
Propineb25.0878.86105.22892.26
Honey
(n = 28)		Insecticide
Carbaryl3.57LODLOD50.23
CarbofuranNDLODLODLOD
Chlorpyrifos7.14LODLOD79.22
Herbicide
GlyphosateNDLODLODLOD
ParaquatNDLODLODLOD
Fungicide
Benomyl3.57LODLOD43.37
Mancozeb7.14LODLOD126.25
Propineb10.71LODLOD79.18
Dust
(n = 180)		Insecticide
Carbaryl47.78337.26368.542314.35
Carbofuran59.44206.54245.332033.54
Chlorpyrifos41.11191.25197.652778.52
Herbicide
Glyphosate62.22306.54377.522788.5
Paraquat52.22317.85404.171926.3
Fungicide
Benomyl22.78207.15236.74885.39
Mancozeb36.11155.23283.311943.25
Propineb23.89289.65303.977336.47
DF: detection frequency; LOD: limit of detection; ND: not detected.
Table 3. Correlation between pesticide concentrations in bees and dust collected at distances.
Table 3. Correlation between pesticide concentrations in bees and dust collected at distances.
Pesticide TypeDistance from Apiary to Dust Collecting Location
Near (0–1000 m)
n = 84		Middle (300–1000 m)
n = 53		Far (1000–5000 m)
n = 43
rprprp
Insecticide−0.3860.1550.0750.7910.6740.006
Herbicide0.838<0.0010.811<0.0010.4670.08
Fungicide0.783<0.0010.834<0.0010.752<0.001
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Hung, C.-C.; Yiin, L.-M. Availability of Using Honeybees as Bioindicators of Pesticide Exposure in the Vicinity of Agricultural Environments in Taiwan. Toxics 2023, 11, 703. https://doi.org/10.3390/toxics11080703

AMA Style

Hung C-C, Yiin L-M. Availability of Using Honeybees as Bioindicators of Pesticide Exposure in the Vicinity of Agricultural Environments in Taiwan. Toxics. 2023; 11(8):703. https://doi.org/10.3390/toxics11080703

Chicago/Turabian Style

Hung, Chien-Che, and Lih-Ming Yiin. 2023. "Availability of Using Honeybees as Bioindicators of Pesticide Exposure in the Vicinity of Agricultural Environments in Taiwan" Toxics 11, no. 8: 703. https://doi.org/10.3390/toxics11080703

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop