A field-based quantitative analysis of sublethal effects of air pollution on pollinators
Edited by Gene E. Robinson, University of Illinois at Urbana Champaign, Urbana, IL, and approved July 13, 2020 (received for review May 12, 2020)
Letter
October 13, 2020
Significance
India is the world’s largest fruit producer and second most populous country. Pollinators are therefore important for India’s food security. India also contains 9 of the world’s 10 most polluted cities, but the impact of air pollution on plant and animal systems is largely unknown. We performed a multiyear study in the megacity of Bangalore to correlate the mechanistic effects of air pollution on a major Indian pollinator, the Giant Asian honey bee, Apis dorsata. Wild honey bees and lab-reared Drosophila exposed to air pollution exhibited differences in survival, behavior, heart rate, blood cell count, and/or the expression of genes related to stress, immunity, and metabolism. Our study indicates the urgency for more studies on wild systems to better inform international air quality guidelines.
Abstract
While the impact of air pollution on human health is well studied, mechanistic impacts of air pollution on wild systems, including those providing essential ecosystem services, are largely unknown, but directly impact our health and well-being. India is the world’s largest fruit producer, second most populous country, and contains 9 of the world’s 10 most polluted cities. Here, we sampled Giant Asian honey bees, Apis dorsata, at locations with varying air pollution levels in Bangalore, India. We observed significant correlations between increased respirable suspended particulate matter (RSPM) deposition and changes in bee survival, flower visitation, heart rate, hemocyte levels, and expression of genes related to lipid metabolism, stress, and immunity. Lab-reared Drosophila melanogaster exposed to these same sites also exhibited similar molecular and physiological differences. Our study offers a quantitative analysis on the current impacts of air pollution on insects, and indicates the urgency for more nonhuman studies to accurately assess the effects of pollution on our natural world.
Sign up for PNAS alerts.
Get alerts for new articles, or get an alert when an article is cited.
Rapid urbanization, intense agriculture, and poor emission control and enforcement coupled with a dependence on ‘old and poorly maintained vehicles’ (1) over the past century have led to immense increases in air pollution in the developing world. According to the World Health Organization (WHO) recent estimates, India now contains 9 of the world’s 10 most polluted cities (2). India also serves as the world’s largest fruit and second-largest vegetable producer (3). Animal pollination is responsible for 35% of all crop productivity (4) and contributes significantly to the productivity of at least 75% of our global crop species (4, 5). Insects, and particularly bees, provide the majority of this pollination (4). Recent studies have documented the cataclysmic decline of insect populations across the world (6–8), and have suggested that reductions in habitat quality, including pollution, contribute to this loss (6, 8). One in six species of bees are also reported locally extinct in various regions throughout the world (8). However, few studies have considered the molecular and physiological impacts of air pollution on organisms other than humans that are equally exposed to airborne toxins such as respirable suspended particulate matter [RSPM; a complex mixture of particles less than 10 or 2.5 µM in diameter respirable by humans (9)] ground-level ozone (O3), nitrogen dioxide (NO2), sulfur dioxide (SO2), and volatile organic compounds (10).
A number of studies over the past 40 y have acknowledged the impact of air pollution on insect populations and olfactory behavior. Effects of air pollution on insects include both direct toxicity and indirect decline due to disequilibria with higher or lower trophic levels (11). In some cases, air pollutants such as sulfur and nitrogen dioxide may increase the prevalence of insect pests such as aphids, perhaps due to a loss of natural enemies (12, 13). Pollutants can inhibit the search behavior of natural enemies through direct impacts on sensory physiology and behavior (14). For insects, air pollutants are likely to impact olfactory-guided behavior strongly, as chemosensory information is used in almost every aspect of insect biology from locating food, mates, and host sites, to avoiding predators (15). For example, the transport of chemical signals in the air used by pollinators to locate flowers and other organisms may be hindered or even destroyed by chemical reactions with pollutants (16). Confounding airborne particles, including smoke, have been shown to decrease bee alarm pheromone detection (17). Interestingly, a recent study implies that despite the potential exposure to higher air pollution levels, bumblebees in the UK can exploit resources in the city more effectively than in agricultural areas (18). The authors suggest this could be due to the reduced pesticide use in comparison to agricultural areas, and that urban areas may provide a refuge for pollinator populations.
Nevertheless, few studies have examined the underlying molecular and physiological correlation of air pollution on wild systems. For humans, air pollution has been associated with respiratory issues such as asthma (19) and lung cancer (20), circulatory impacts including ventricular hypertrophy (20) and heart disease, (21) sensory and neurological disorders such as retinopathy (22), Alzheimer’s and Parkinson’s diseases (21), autism (23), cognitive performance (24), and general effects associated with stress, toxicity, homeostasis, or immunity including obesity (21), fetal growth impairment (25), low birth weight (26), deaths, disease burden, and life expectancy (27), among many others. As a first principle, one could initially hypothesize that insects might also exhibit similar respiratory, circulatory, neurological, stress, homeostatic and/or immune system impacts upon exposure to air pollution, in addition to the behavioral and population level impacts already observed in previous insect studies.
Here, we have developed a field-based analysis to correlate the impact of RSPM on wild pollinator behavior, respiratory and circulatory physiology, immunology, and expression of genes related to stress, toxicity, homeostasis, and immunity as suggested by current insect behavior and human health studies. To this end, we examined the impact of air pollution on insect pollinators by sampling Apis dorsata, the Giant Asian honey bee, at distinct locations in the megacity of Bangalore, India [Fig. 1A; current population: 13.1 million (28)]. Bangalore has witnessed a 47% population increase in just 10 years. The measured RSPM values from 2015 to 2016 have exceeded the national permissible limit [60.0 µg/m3, which is defined by the Ministry of Environment and Forest (MoEF), Government of India] (29) in 15 locations of Bangalore (30). A. dorsata (Fig. 1B) produces more than 80% of the honey in India and has been found to pollinate more than 687 plant species (31). A. dorsata is a rock bee often found nesting in cities. This species has been shown to migrate its colonies repeatedly over many kilometers throughout the year among urban, rural, and forest habitats (31, 32). As a consequence, any long-term physiological or behavioral impacts of urban air pollution will also impact the pollination services these insects provide to agricultural and forest areas where they migrate. A. dorsata is a rarely managed species (33), and there are no current beekeeping efforts in Bangalore. Coupled with its agricultural and ecosystems relevance, its common occurrence in cities makes it an excellent species to evaluate the impact of urban air pollution on wild systems and ecosystem services. Here, we correlate air pollution levels at various sites in Bangalore with differences in A. dorsata survival, floral visitation behavior, and cellular and molecular physiology including circulatory, immune, and gene expression aspects as suggested by human studies and known insect toxins such as pesticides. To control for potential impacts of diet and colony variation, and to determine if these responses could be generalized to other insect species, we also exposed laboratory-reared and age-matched Drosophila melanogaster to the same sites and performed the same physiological and molecular experiments.
Fig. 1.
Results and Discussion
Site Selection, Floral Visitation, and Survival.
Collection and observation sites were chosen as representative of varying pollution levels within a maximum linear distance of approximately 21 km between the two most distant sites. Each location was within 8 km of the next closest site (Fig. 1A). The RSPM (measured directly here as PM10, particles < 10 µM) averaged 28.32 ± 10.51 (mean ± SD, n = 78 measurement days) at rural (R), 33.73 ± 11.03 at low (L, n = 115), 45.95 ± 20.52 at moderate (M, n = 103), and 98.59 ± 55.43 µg/m3 at the highly polluted site (H, n = 118) between January 2017 to April 2019 (Fig. 1C and SI Appendix, Fig. S1A; Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test). The specific sampling locations were chosen to reduce the effects of other potential variables such as pesticide use, immediate floral abundance within 100 m of the collection site, traffic, shade, presence of human-made structures, and presence of Giant Asian honey bee colonies (SI Appendix, Table S1). As a proxy for general floral abundance encompassing the maximum foraging range of A. dorsata at each site (up to 3.5-km radius) (32, 35), we also analyzed variation in the Normalized Difference Vegetative Index (NDVI) concurrent with experimental dates (36–38 and SI Appendix, Fig. S1). At each location, we identified the [nonnative (39)] flowering tree, Tecoma stans, a very common ornamental plant present across India which blooms throughout the year and is pollinated by A. dorsata. We therefore choose this flowering plant as our species of interest for A. dorsata visitation behavior, and made all observations and collections during morning bee visitation times for this species (40). At each site, we observed multiple colonies (SI Appendix, Table S1), but could not assign collected bees to specific colonies. Given that A. dorsata frequently migrate their colonies (31, 32), it is highly unlikely that the same colonies were present at each site across the entire 3-y study (SI Appendix, Table S2). Therefore, to reduce the potential for pseudoreplication from specific colonies as well as effects of diet and colony variation within and between sites, we randomly collected bees while they were foraging at T. stans flowers at multiple sites over several time points throughout the study (SI Appendix, Table S2). Multiple sites were visited within the same time period to mitigate weather or seasonal concerns. In total, we have sampled 1,820 Giant Asian honey bees over 3 y (n = 455 bees/site).
The number of bees per inflorescence (5 min observation of 20 inflorescences per 20 d; n = 1,600, 400 inflorescences/site) was significantly different between sites with varying pollution levels (Fig. 1D; Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test) even though the number of flowers in the least polluted rural site was less (SI Appendix, Fig. S1D; n = 4,000, 1,000/site, Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test). This is in agreement with previous studies correlating population levels or pollinator behavior with RSPM levels (16, 17). Temperature, humidity, and wind speed did not differ between sites over the time periods where bees were observed or collected, and NDVI was significantly higher only in the rural site, while the other three sites did not differ in vegetative index over the same period (SI Appendix, Fig. S1 A–C; n = 200 samples per site for temperature, humidity, and windspeed, n = 16 per site for NDVI, Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test). Giant Asian honey bees collected from moderate and highly polluted sites exhibited significantly lower survival rates over the total observation period (Fig. 1E; n = 200, 50 bees/site; χ2 test), despite providing 10% honey water ad libitum and removing bees that did not feed and died within the first 4 h. We note that this difference is not likely due to extreme age differences as no bees from any site exhibited wing damage generally observed during senescence (41) (SI Appendix, Fig. S1L). Collectively, the field studies suggest a correlation between RSPM levels and A. dorsata behavior and survival. We thus continued to examine underlying physiological and molecular correlates to these pollution levels.
Morphology and Toxicology.
We next examined if there was evidence of air pollution on the exoskeleton and internal tissues of field-collected Giant Asian honey bees from each site. Previous studies have shown increased deposition of particulate matter (RSPM) on bee epicuticular waxy layers corresponding to higher air pollution levels (42). First, we observed no difference in the total weight (SI Appendix, Fig. S1K; n = 120, 30 bees/site; Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test) and morphometric measurements of hindlegs, antennae, wings, and intertegular spans (43) for bees collected from the four sites (SI Appendix, Fig. S1 F–I; n = 80, 20 bees/site, Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test). We then quantified the deposition of RSPM by measuring the percentage area covered by particulate matter on the outermost layer [0.2 to 0.4 μm thick epicuticular waxes (44)] for hindlegs, antennae, and wings with scanning electron microscopy (SEM; n = 480, 120 sections/site), after allowing bees to groom (45, 46). We observed RSPM present on the edges and upper surface of the wings (Fig. 2 A, B, and G). Antennae displayed RSPM deposition near the tip when compared to the rest of the segments (Fig. 2 C, D, and H), and the inner and outer cuticular layer of hindlegs also showed congestion of particles (Fig. 2 E, F, and I).
Fig. 2.
Overall, the percent area covered with RSPM for all three selected body sections (Fig. 2 J–L) was significantly higher for bees collected from the highly polluted site (n = 480 total sections, 40 samples/section/site, 20 bees/site; Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test). To determine the elemental composition of this RSPM, we coupled SEM with X-ray spectroscopy (SEM-EDX) analysis (n = 80 bees, 20/site), which revealed the presence of several heavy metals including arsenic (As), lead (Pb), tungsten (W), aluminum (Al), and several other metals, with the highest number of metallic elements present on bees collected from the highly polluted site (Fig. 2M), as in ref. 42. Several of these metals have been shown to be toxic for insects, invertebrates, and humans in a number of studies (47, 48). This provided our first direct evidence of increased exposure to pollutants and known toxins correlating with increased RSPM levels in our field sites.
Respiratory and circulatory physiology and immunology.
As suggested by human health studies, we then examined other aspects of bee physiology, including respiratory and circulatory physiology, and immunology. Respiration rates of Giant Asian honey bees collected from all four sites showed few significant differences between sites [Fig. 3A, n = 200, 40 bees/site including unexposed bees (3- to 5-d-old bees maintained in the incubator at 33 °C), Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test]. Phenoloxidase activity, which has been shown to initiate the humoral and cellular immune response against pathogens (49), also did not differ in bees between sites (Fig. 3B; n = 80, 20/site, Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test). However, while the mean heart rate itself did not differ between bees from different sites (SI Appendix, Fig. S1J; n = 15 bees/site), the mean SD of interbeat interval of heart rate (IBI), an alternate method and an indicator of arrhythmicity (50), was significantly different for bees collected from the highly polluted site (Fig. 3C, SI Appendix, Fig. S4D, and Movie S1; n = 180, 450 sections/site with 30 sections/bee; Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test). Total hemocyte count (THC) was also significantly different for bees from the low and highly polluted sites (Fig. 3D; n = 100, 20 bees/site including unexposed bees, Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test). Together, these results suggest a correlation between increased RSPM and impacts on insect circulatory physiology.
Movie S1.
Sample heartbeat videos of Giant Asian honey bees collected from low and highly polluted sites.
Fig. 3.
Gene expression.
Given the potential impact of air pollution on olfactory communication (51), and the correlation with circulatory physiology shown in this study, we also assessed whether the observed differences in behavior, survival, and physiology also correlated to changes in gene expression in the antennae and heart of Giant Asian honey bees collected from the different sites (n = 80, 20 bees/site for antenna and heart, separately). Here, we concentrated on genes responsible for stress, metabolism, nutrition and defense, homeostasis and innate immunity, and cellular regulation as suggested from human health impacts. We performed comprehensive RNA sequencing (RNA-seq) analysis to assess the gene expression profiles from the antennae of all four sites. A total of 46, 29, and 39 differentially expressed genes (DEGs) were found expressed in bee antennae collected from low, moderate, and highly polluted sites when compared with rural site bees (Fig. 3E and SI Appendix, Fig. S2 A–C) using P value < 0.05, and 1.5 log2FoldChange. We used UniProt (52) for gene ontology (GO) enrichment to identify the function and the Kyoto Encyclopedia of Genes and Genomes (KEGG) (53) to define the pathways of the DEGs (Fig. 3F and SI Appendix, Table S3). A biomolecular interaction network was constructed using Cytoscape 3.6.1 (54) to visualize the relationships among genes. Gene ontology enrichment (53) of antennae-expressed genes revealed that all of the DEGs responsible for stress and lipid metabolism were up-regulated in bees collected in the highly polluted site when compared to rural bees (Fig. 3F and SI Appendix, Figs. S3A and S4 A–C). From gene ontology and pathway analysis of the heart transcriptome, we found all 14 identified genes responsible for stress and 60% of lipid metabolism-related genes were differentially expressed in bees collected from the highly polluted site when compared to bees from the low polluted site (Fig. 3F and SI Appendix, Figs. S3B and S4E).
We thus performed qRT-PCR using unexposed Giant Asian honey bees and bees collected from low and highly polluted sites (n = 80, 20 bees/site) with three biological replicates using nine differentially expressed genes (SI Appendix, Fig. S3), responsible for stress (vitellogenin-like, sterol-o-acyltransferase, and laccase-5-like), lipid metabolism (fatty acid synthase), nutrition and defense (major royal jelly protein), homeostasis and innate immunity (hymenoptaecin, cytochrome P450, and tyrosine 3-monooxygenase), and transcriptional regulation (histone H3). The KEGG pathway and gene ontology enrichment analysis showed that DEGs from antennal and heart tissue collected from A. dorsata at the different sites were involved in 10 different pathways (SI Appendix, Fig. S4). We selected specific genes from these pathways based on their biological significance for honey bees (such as known relationships with stress, metabolism, nutrition and defense, homeostasis, innate immunity, and cellular regulation) and common parameters like primer length, melting temperature (Tm), annealing temperature, Mg2+ concentration, and other factors. The gene-specific primers used are listed in SI Appendix, Table S4.
Vitellogenin is a phospholipoglycoprotein (55) released from the fat body that circulates in the hemolymph (56). It protects cells from oxidative damage (57) and has incidentally been suggested to serve as a potential biomarker for neonicotinoid pesticide exposure in bees (58). Variation in worker behavior and lifespan is also correlated with vitellogenin depletion (59). Likewise, we observed low survival rates for Giant Asian honey bees from the highly polluted site (Fig. 1E) accompanied by low expression of vitellogenin in the antenna (Fig. 3G, two‐tailed t test). The sterol-o-acyltransferase gene is an evolutionarily conserved metabolic gene (60) that plays an important role in cellular cholesterol metabolism (61). In our study, the sterol-o-acyltransferase gene exhibited significantly increased expression in bees from the highly polluted site (Fig. 3G, two‐tailed t test). Laccases are important in cuticle sclerotization and toxin oxidation in insects (62), and cytochrome P450 monooxygenase is an important enzyme involved in detoxification and has also been implicated in bees’ response to toxins such as pesticides (63, 64). These two genes correspondingly displayed increased levels of expression for antennal and heart tissue, respectively, for bees collected from the highly polluted site (Fig. 3 G and H, two‐tailed t test). Finally, the expression of tyrosine 3-monooxygenase, which is responsible for homeostasis and innate immunity, was significantly increased in heart tissue from bees collected from the highly polluted (Fig. 3H, two‐tailed t test). Tyrosine 3-monooxygenase plays a major role in cuticle tanning, pigmentation, wound healing, and melanization of microbes and parasites during immune responses (49, 65). Interestingly, we did not see a correlation between the observed increase in tyrosine 3-monooxygenase expression and differences in phenoloxidase activity in bees from different sites, which also leads to melanin production for immune response (Fig. 3B). Perhaps this is because the change in tyrosine 3-monooxygenase activity is due to its other roles, or our physiological assay was not sensitive enough to detect the changes observed.
Exposure of laboratory-reared and age-matched Drosophila.
Our results indicate that Giant Asian honey bees exposed to RSPM exhibit significant differences in flower visitation behavior, heart rate, hemocyte levels, and ultimately survival. These differences are reflected by significant differences in the expression of genes related to stress, lipid metabolism, homeostasis, and innate immunity, which are also impacted in human health studies on air pollution. While the behavioral and physiological differences observed did not generally correlate with other potential factors such as date of collection, humidity, temperature, or wind speed, we did note a significant correlation of several of the measured parameters with the NDVI (Normalized Difference Vegetative Index; SI Appendix, Table S6, Pearson’s correlation analysis). As a consequence, while we collected bees while foraging at nectar-rich T. stans at each site, it is possible that the bees were impacted by other differences in floral abundance, age, diet, source colony, or physiological condition between sites that we could not control in this field-based study. We note, however, that the NDVI index was only significantly different at the rural site (SI Appendix, Fig. S1C), implying that pollution, rather than vegetative availability, is a more probable factor driving the observed behavioral, physiological, and molecular differences in bees collected at the different sites.
Furthermore, the observed increases in RSPM and several toxic metals on the exoskeleton and internal tissues of field-collected bees using SEM-EDX suggest a direct relationship between exposure to RSPM and impacts on insect health and survival. Thus, to control for age, diet, source, and physiological differences between insects, we evaluated the survival, RSPM deposition, hemocyte levels, heart rate, and gene expression of laboratory-reared and age matched Canton Special D. melanogaster to laboratory conditions (C = control) as well as our low (L) and highly (H) polluted sites (66). Newly emerged flies (n = 21,000, 7,000 flies/site, 1,000 flies per trial) were exposed for 10 d under similar shade and wind-protected locations and were provided with fresh food and water on alternate days. Under these conditions, flies exposed to the low and highly polluted sites showed significantly lower survival rates after 10 d of exposure (Fig. 4A; n = 21 cages, 7 cages per site, 1,000 flies per cage, Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test). We then correlated the deposition of RSPM on the outermost layer for hindlegs, antennae, and wings with scanning electron microscopy (SEM) as performed for the Giant Asian honey bee (Fig. 4 B–E; n = 450, 150/site), and after allowing flies to groom (67). We observed a similar pattern to that found for bees with significantly increased deposition of RSPM present on the edges and upper surface of the wings, the antennal tip, and on the hindlegs of flies exposed to the highly polluted site (Fig. 4 C–E; n = 450 sections, 50 samples/section/site, Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test). We used SEM-EDX to determine the elements present on flies (n = 150, 50 flies/site), which showed the presence of iron (Fe), manganese (Mn), silicon (Si), aluminum (Al), zinc (Zn), and copper (Cu) only on flies exposed to the highly polluted sites. However, the mean SD of interbeat interval in fly heart rate (Fig. 4F; n = 60, 20 flies/site, Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test) and total fly hemocyte count (THC; Fig. 4G; n = 2,400 flies, 20 trials/site, 40 flies/trial, Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test) were significantly different for laboratory-maintained flies, while the bee results also showed intersite differences. This discrepancy could be due to increased sensitivity in flies to pollution eliciting changes even at low pollution levels, the difference in length of exposure time to air pollution [10 d for flies vs. at least 22+ d for foraging bees (68)], or that the intersite effects in bees resulted from additional factors beyond air pollution.
Fig. 4.
We also performed qRT-PCR to compare the results obtained from bee experiments to laboratory-maintained flies and those exposed to our high and low polluted sites. We quantified the same genes in Drosophila we selected for bees for stress (yolk protein-1, sterol-o-acyltransferase, and laccase-2), lipid metabolism (fatty acid synthase), defense (yellow protein), homeostasis and innate immunity (cytochrome P450 and tyrosine 3-monooxygenase), and transcriptional regulation (histone H3-3a; Fig. 4H; n = 180, 60 flies/site). Similar to our bee qRT-PCR, we found increased expression of cytochrome P450, sterol-o-acyltransferase, and tyrosine 3-monooxygenase in flies exposed to the highly polluted site (Fig. 4H, two‐tailed t test). However, fly yolk protein displayed a significantly increased level of expression in flies from the highly polluted condition, unlike the analogous vitellogenin gene in bees. Interestingly, a previous study (69) showed that increased expression levels of Drosophila yolk protein family genes were negatively correlated with life span, opposite to vitellogenin for bees (59). Thus, these results do correspond to species-specific differences between stress-related genes and longevity between bees and flies.
Taken together, our results for both field-collected pollinators and laboratory-reared fruit flies indicate that RSPM is the most parsimonious cause for our observed molecular, cellular, physiological, morphological, and behavioral differences. In the case of the Giant Asian honey bee, A. dorsata, these effects could be either from direct or indirect exposure through contaminated food, water (70), or other substrates. These observed effects should now be subjected to controlled laboratory exposure studies to determine the specific RSPM components and levels at which these effects are observed. Surprisingly, increased RSPM levels did not correlate with quantitative changes in respiration rate or innate immunity, two physiological correlates most implicated for impacts of pollution on human health. Thus, increased air pollution may have unique and unexpected impacts on animal systems not anticipated from human studies alone. Nevertheless, our study offers an initial quantitative examination of the impacts of air pollution on insects, and shows that these molecular and physiological effects can be generalized to multiple insect orders (in this case, Hymenoptera and Diptera). Furthermore, our indication of molecular and physiological impacts on a migratory ecosystem service provider, the Giant Asian honey bee, suggest that the effects of urban air pollution could persist far from their point of origin.
Finally, the associated impacts of RSPM were observed at a PM10 of roughly 50 µg/m3 (our moderately polluted site). This level is equivalent to the “Interim Target-2” levels recommended by the World Health Organization for policy adoption (71). At these mean levels, we observed over 80% reduction in pollinator survival as well as significant molecular and physiological changes. We thus emphasize the urgency for more studies on wild plant and animal systems to better inform international air quality guidelines. Such studies are imperative to reveal the full impact of air pollution not only on human and environmental ecosystem health but, as in the case of this study, economic loss to pollinator-dependent crops and food security in highly polluted and vulnerable regions such as India.
Materials and Methods
This study was performed to correlate the effect of air pollution on pollinator survival, floral visitation behavior, circulatory physiology, and gene expression. We used several approaches to assess Giant Asian honey bees, A. dorsata, from four sites selected with similar conditions except for varying air pollution levels: 1) field behavior to assess floral visitation and survival, 2) SEM-EDX to assess the particulate matter and its composition on the Giant Asian honey bee body, 3) respirometry, cardiac physiology, and biochemistry to assess impacts on respiratory, circulatory, and immune systems, respectively, 4) antennae and heart transcriptomics and qRT-PCR analysis to assess impacts on gene expression. We separately used these same approaches to measure survival, circulatory physiology, and gene expression in laboratory-reared and age-matched D. melanogaster exposed to laboratory conditions and our low and highly polluted sites as described in ref. 66 for 10 d with food and water. Sample numbers, data assembly, measurements, and statistical analyses are described in SI Appendix, Materials and Methods.
Data Availability
All data used in this study are made available within this article and SI Appendix.
Acknowledgments
This work was supported by the National Centre for Biological Sciences, Tata Institute of Fundamental Research under the Department of Atomic Energy, Government of India, project no. 12-R&D-TFR-5.04-0800 and 12-R&D-TFR-5.04-0900. G.G.T. was supported by the Science and Engineering Research Board (SERB) National Postdoctoral Fellowship (PDF/2017/000366), S.B.O. by a Tata Education Trust grant and a SERB Ramanujan Fellowship, and P.S.D. with the Wellcome Trust-India Alliance Intermediate Fellowship (IA/I/16/1/502367). We thank the NCBS SEM facility for providing access to SEM-EDX, NCBS fly facility for providing flies, Frédéric Marion-Poll for initial insights into this study, the INK Fellows Program for support, R. Manaswi for support in selecting field sites, Pavan Kumar Kaushik for assistance with initial data processing of heart rate analysis, and Prathap Nagraj for his support with NDVI analysis. We are grateful to NASA for providing datasets for NDVI analysis. We thank the managements of all the study sites for the permission to use their facilities for this research work.
Supporting Information
Appendix (PDF)
- Download
- 3.52 MB
Movie S1.
Sample heartbeat videos of Giant Asian honey bees collected from low and highly polluted sites.
- Download
- 33.28 MB
References
1
S. Ahmed, Air pollution and its impact on agricultural crops in developing countries. J. Anim. Plant Sci. 25, 297–302 (2015).
2
WHO, Data from “Ambient (outdoor) air quality database, by country and city.” WHO Global Ambient Air Quality Database. Available at https://www.who.int/airpollution/data/cities/en/. Accessed 4 July 2019.
3
FAO, Data from “Countries by commodity.” FAOstat. Available at http://fenix.fao.org/faostat/internal/en/#rankings/countries_by_commodity. Accessed 4 July 2019.
4
A. M. Klein et al., Importance of pollinators in changing landscapes for world crops. Proc. Biol. Sci. 274, 303–313 (2007).
5
T. D. Breeze et al., Agricultural policies exacerbate honeybee pollination service supply-demand mismatches across Europe. PLoS One 9, e82996–e82998 (2014).
6
B. C. Lister, A. Garcia, Climate-driven declines in arthropod abundance restructure a rainforest food web. Proc. Natl. Acad. Sci. U.S.A. 115, E10397–E10406 (2018).
7
F. Sánchez-Bayo, K. A. G. Wyckhuys, Wyckhuys, worldwide decline of the entomofauna: A review of its drivers. Biol. Conserv. 232, 8–27 (2019).
8
C. A. Hallmann et al., More than 75 percent decline over 27 years in total flying insect biomass in protected areas. PLoS One 12, e0185809 (2017).
9
J. Liang, “Particulate matter” in Chemical Modeling for Air Resources, (Elsevier Science, 2013), pp. 189–219.
10
A. T. Archibald, G. Folberth, D. C. Wade, D. Scott, A world avoided: impacts of changes in anthropogenic emissions on the burden and effects of air pollutants in Europe and North America. Farad. Discuss. 200, 475–500 (2017).
11
D. N. Alstad, G. F. J. Edmunds, L. H. Weinstein, Effects of air pollutants on insect populations. Annu. Rev. Entomol. 27, 369–384 (1982).
12
M. Brändle, U. Amarell, H. Auge, S. Klotz, R. Brandl, Plant and insect diversity along a pollution gradient: Understanding species richness across trophic levels. Biodivers. Conserv. 10, 1497–1511 (2001).
13
J. N. B. Bell, F. M. Marshall, “Field studies on impacts of air pollution on agricultural crops” in Environmental Pollution Plant Responses, (CRC Press, 2000), pp. 99–110.
14
I. M. Gate, S. McNeill, M. R. Ashmore, Effects of air pollution on the searching behaviour of an insect parasitoid. Water Air Soil Pollut. 85, 1425–1430 (1995).
15
V. G. Dethier, Chemical Insect Attractants and Repellents, (Blakiston, Philadelphia, 1947), p. 289.
16
Q. S. McFrederick, J. D. Fuentes, T. Roulston, J. C. Kathilankal, M. Lerdau, Effects of air pollution on biogenic volatiles and ecological interactions. Oecologia 160, 411–420 (2009).
17
P. K. Visscher, R. S. Vetter, G. E. Robinson, Alarm pheromone perception in honeybees is decreased by smoke (Hymenoptera: Apidae). J. Insect Behav. 8, 11–18 (1995).
18
A. E. Samuelson, R. J. Gill, M. J. F. Brown, E. Leadbeater, Lower bumblebee colony reproductive success in agricultural compared with urban environments. Proc. Biol. Sci. 285, 20180807 (2018).
19
M. A. Riedl, The effect of air pollution on asthma and allergy. Curr. Allergy Asthma Rep. 8, 139–146 (2008).
20
B. Brunekreef et al., Effects of long-term exposure to traffic-related air pollution on respiratory and cardiovascular mortality in the Netherlands: The NLCS-AIR study. Res. Rep. Health Eff. Inst. 139, 5–71, discussion 73–89 (2009).
21
L. Calderón-Garcidueñas et al., Long-term air pollution exposure is associated with neuroinflammation, an altered innate immune response, disruption of the blood-brain barrier, ultrafine particulate deposition, and accumulation of amyloid beta-42 and alpha-synuclein in children and young adults. Toxicol. Pathol. 36, 289–310 (2008).
22
S. K. West et al., Is household air pollution a risk factor for eye disease? Int. J. Environ. Res. Public Health 10, 5378–5398 (2013).
23
A. L. Roberts et al., Perinatal air pollutant exposures and autism spectrum disorder in the children of Nurses’ Health Study II participants. Environ. Health Perspect. 121, 978–984 (2013).
24
X. Zhang, X. Chen, X. Zhang, The impact of exposure to air pollution on cognitive performance. Proc. Natl. Acad. Sci. U.S.A. 115, 9193–9197 (2018).
25
S. Liu, D. Krewski, Y. Shi, Y. Chen, R. T. Burnett, Association between maternal exposure to ambient air pollutants during pregnancy and fetal growth restriction. J. Expo. Sci. Environ. Epidemiol. 17, 426–432 (2007).
26
D. M. Stieb, L. Chen, M. Eshoul, S. Judek, Ambient air pollution, birth weight and preterm birth: A systematic review and meta-analysis. Environ. Res. 117, 100–111 (2012).
27
India State-Level Disease Burden Initiative Air Pollution Collaborators, The Impact of Air Pollution on Deaths, Disease Burden, and Life Expectancy across the States of India: The Global Burden of Disease Study 2017, (Lancet Planetary Health, 2018).
28
Directorate of Census Operations, District Census Handbook, (Ministry of Home Affairs, Government of India–Press Information Bureau, Karnataka, 2011).
29
The Ministry of Environment and Forest, Revised national ambient air quality standards. https://www.ksndmc.org/Uploads/Pollution.pdf. Accessed 25 July 2020.
30
Central Pollution Control Board (CPCB), Annual report 2015-16: National Air Quality Monitoring Programme, Ministry of Environment, Forest & Climate Change. https://cpcb.nic.in/openpdffile.php?id=UmVwb3J0RmlsZXMvNjE0XzE1MjYzMDE0ODhfbWVkaWFwaG90bzEyMzg3LnBkZg==. Accessed 25 July 2020.
31
S. Basavarajappa, K. S. Raghunandan, Colony status of Asian giant honeybee, Apis dorsata fabricius in Southern Karnataka, India. Afr. J. Agric. Res. 8, 680–689 (2013).
32
F. C. Dyer, T. D. Seeley, Dance dialects and foraging range in three Asian honey bee species. Behav. Ecol. Sociobiol. 28, 227–233 (1991).
33
N. Q. Tan, P. H. Chinh, P. H. Thai, V. Mulder, Rafter beekeeping with Apis dorsata: Some factors affecting the occupation of rafters by bees. J. Apic. Res. 36, 49–54 (1997).
34
Sinergise Ltd., Data from “Sentinel-2.” EO Browser. https://apps.sentinel-hub.com/eo-browser/?zoom=12&lat=12.99113&lng=77.55318&themeId=DEFAULT-THEME. Accessed 25 July 2020.
35
P. L. Kohl et al., Adaptive evolution of honeybee dance dialects. Proc. Biol. Sci. 287, 20200190 (2020).
36
J. W. Rouse, R. H. Haas, J. A. Schell, D.W. Deering, “Monitoring vegetation systems in the great plains with ERTS” in Third NASA ERTS Symposium, (SP-351 I, NASA, Washington, DC, 1974), pp. 309–317.
37
A. R. Formaggio, I. D. Sanches, Sensoriamento Remoto Em Agricultura, (Oficina de Textos, São Paulo, SP, 2017).
38
F. Ponzoni, Y. Shimabukuro, T. Kuplich, Sensoriamento remoto da vegetação. 2a Edição Atualizada e Ampliada, (Oficina de Textos, São Paulo, SP, 2012).
39
C. S. Reddy, G. Bagyanarayana, K. N. Reddy, S. V. Raju, Invasive alien Flora of India, Published by, National Biological Information Infrastructure, (US Geological Survey, USA, 2008).
40
K. Henry Jonathan, A. J. Solomon Raju, K. S. Branham, D. Sunanda, Interactions between insect pollinators and the ornamental tree, Tecoma stans (L.). J. Threat. Taxa 1, 126–127 (2009).
41
A. D. Higginson, F. Gilbert, Paying for nectar with wingbeats: A new model of honeybee foraging. Proc. Biol. Sci. 271, 2595–2603 (2004).
42
I. Negri, C. Mavris, G. Di Prisco, E. Caprio, M. Pellecchia, Honey Bees (Apis mellifera, L.) as active samplers of airborne particulate matter. PLoS One 10, e0132491 (2015).
43
H. James, Estimation of bee size using intertegular span (Apoidea). J. Kans. Entomol. Soc. 60, 145–147 (1987).
44
K. H. Lockey, The thickness of some insect epicuticular wax layers. J. Exp. Biol. 37, 316–329 (1960).
45
G. J. Amador et al., Honey bee hairs and pollenkitt are essential for pollen capture and removal. Bioinspir. Biomim. 12, 026015 (2017).
46
O. Boecking, W. Ritter, Grooming and removal behaviour of Apis mellifera intermissa in Tunisia against Varroa jacobsoni. J. Apic. Res. 32, 127–134 (1993).
47
E. K. Taha, A. M. Al-Jabr, S. N. Al-Kahtani, Honey Bees, Bee-collected pollen and honey as monitors of environmental pollution at an industrial cement area in Saudi Arabia. J. Kans. Entomol. Soc. 90, 1–10 (2017).
48
C. Cruz-Landim, J. E. Serrão, Ultrastructure and histochemistry of the mineral concretions in the midgut of bees (Hymenoptera: Apidae). Neth. J. Zool. 47, 21–29 (1997).
49
V. J. Marmaras, N. D. Charalambidis, C. G. Zervas, Immune response in insects: The role of phenoloxidase in defense reactions in relation to melanization and sclerotization. Arch. Insect Biochem. Physiol. 31, 119–133 (1996).
50
J. R. Keene, R. B. Clayton, C. K. Berke, T. Loof, P. D. Bolls, On the use of beats-per-minute and interbeat interval in the analysis of cardiac responses to mediated messages. Commun. Res. Rep. 34, 265–274 (2017).
51
J. Andreas, B. Mascha, Changing odour landscapes: The effect of anthropogenic volatile pollutants on plant–pollinator olfactory communication. Funct. Ecol. 31, 56–64 (2017).
52
Uniprot Consortium, Reorganizing the protein space at the Universal Protein Resource (UniProt). Nucleic Acids Res. 40, D71–D75 (2012).
53
M. Kanehisa, S. Goto, KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 28, 27–30 (2000).
54
P. Shannon et al., Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res. 13, 2498–2504 (2003).
55
D. E. Wheeler, J. K. Kawooya, Purification and characterization of honey bee vitellogenin. Arch. Insect Biochem. Physiol. 14, 253–267 (1990).
56
G. V. Amdam, K. Norberg, R. E. Page Jr., J. Erber, R. Scheiner, Downregulation of vitellogenin gene activity increases the gustatory responsiveness of honey bee workers (Apis mellifera). Behav. Brain Res. 169, 201–205 (2006).
57
S. C. Seehuus, K. Norberg, T. Krekling, K. Fondrk, G. V. Amdam, Immunogold localization of vitellogenin in the ovaries, hypopharyngeal glands and head fat bodies of honeybee workers, Apis mellifera. J. Insect Sci. 7, 1–14 (2007).
58
V. Christen, M. S. Vogel, T. Hettich, K. Fent, A vitellogenin antibody in honey bees (Apis mellifera): Characterization and application as potential biomarker for insecticide exposure. Environ. Toxicol. Chem. 38, 1074–1083 (2019).
59
G. V. Amdam, S. W. Omholt, The hive bee to forager transition in honeybee colonies: The double repressor hypothesis. J. Theor. Biol. 223, 451–464 (2003).
60
D. E. Dove et al., ACAT1 deficiency disrupts cholesterol efflux and alters cellular morphology in macrophages. Arterioscler. Thromb. Vasc. Biol. 25, 128–134 (2005).
61
T.-Y. Chang, B. L. Li, C. C. Chang, Y. Urano, Acyl-coenzyme A:cholesterol acyltransferases. Am. J. Physiol. Endocrinol. Metab. 297, E1–E9 (2009).
62
Shraddha, R. Shekher, S. Sehgal, M. Kamthania, A. Kumar, Laccase: Microbial sources, production, purification, and potential biotechnological applications. Enzyme Res. 2011, 217861 (2011).
63
M. Gottardi, A. Kretschmann, N. Cedergreen, Measuring cytochrome P450 activity in aquatic invertebrates: A critical evaluation of in vitro and in vivo methods. Ecotoxicology 25, 419–430 (2016).
64
R. L. Vannette, A. Mohamed, B. R. Johnson, Forager bees (Apis mellifera) highly express immune and detoxification genes in tissues associated with nectar processing. Sci. Rep. 5, 16224 (2015).
65
M. J. Gorman, C. An, M. R. Kanost, Characterization of tyrosine hydroxylase from Manduca sexta. Insect Biochem. Mol. Biol. 37, 1327–1337 (2007).
66
S. L. de Santana et al., Drosophila melanogaster as model organism for monitoring and analyzing genotoxicity associated with city air pollution. Environ. Sci. Pollut. Res.Int. 25, 32409–32417 (2018).
67
A. L. Szebenyi, Cleaning behaviour in Drosophila melanogaster. Anim. Behav. 17, 641–651 (1969).
68
S. Yadav, Y. Kumar, B. L. Jat, “Honeybee: Diversity, castes and life cycle” in Industrial Entomology, Omkar, Ed. (Springer, 2017), pp. 5–34.
69
Y. Ren, K. A. Hughes, Vitellogenin family gene expression does not increase Drosophila lifespan or fecundity. F1000 Res. 3, 125 (2014).
70
P. W. Lau, C. N. James, Salt preferences of honeybee water foragers. J. Exp. Biol. 219, 790–796 (2016).
71
WHO, WHO Air quality guidelines for particulate matter, ozone, nitrogen dioxide and sulfur dioxide. https://apps.who.int/iris/bitstream/handle/10665/69477/WHO_SDE_PHE_OEH_06.02_eng.pdf?sequence=1. Accessed 25 July 2020.
Information & Authors
Information
Published in
Classifications
Copyright
Copyright © 2020 the Author(s). Published by PNAS. This open access article is distributed under Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND).
Data Availability
All data used in this study are made available within this article and SI Appendix.
Submission history
Published online: August 10, 2020
Published in issue: August 25, 2020
Keywords
Acknowledgments
This work was supported by the National Centre for Biological Sciences, Tata Institute of Fundamental Research under the Department of Atomic Energy, Government of India, project no. 12-R&D-TFR-5.04-0800 and 12-R&D-TFR-5.04-0900. G.G.T. was supported by the Science and Engineering Research Board (SERB) National Postdoctoral Fellowship (PDF/2017/000366), S.B.O. by a Tata Education Trust grant and a SERB Ramanujan Fellowship, and P.S.D. with the Wellcome Trust-India Alliance Intermediate Fellowship (IA/I/16/1/502367). We thank the NCBS SEM facility for providing access to SEM-EDX, NCBS fly facility for providing flies, Frédéric Marion-Poll for initial insights into this study, the INK Fellows Program for support, R. Manaswi for support in selecting field sites, Pavan Kumar Kaushik for assistance with initial data processing of heart rate analysis, and Prathap Nagraj for his support with NDVI analysis. We are grateful to NASA for providing datasets for NDVI analysis. We thank the managements of all the study sites for the permission to use their facilities for this research work.
Notes
This article is a PNAS Direct Submission.
Authors
Competing Interests
The authors declare no competing interest.
Metrics & Citations
Metrics
Altmetrics
Citations
Cite this article
A field-based quantitative analysis of sublethal effects of air pollution on pollinators, Proc. Natl. Acad. Sci. U.S.A.
117 (34) 20653-20661,
https://doi.org/10.1073/pnas.2009074117
(2020).
Copied!
Copying failed.
Export the article citation data by selecting a format from the list below and clicking Export.
Cited by
Loading...
View Options
View options
PDF format
Download this article as a PDF file
DOWNLOAD PDFLogin options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginRecommend to a librarian
Recommend PNAS to a LibrarianPurchase options
Purchase this article to access the full text.