Potential of Quercetin to Reduce Herbivory without Disrupting Natural Enemies and Pollinators

: Quercetin is one of the most abundant ﬂavonoids in terrestrial plants and pollen. In living plants, quercetin can function as a secondary metabolite to discourage insect herbivory. Literature on insect-quercetin interactions was searched and data synthesized to test the hypothesis that quercetin can become an effective biocide to reduce herbivory without disrupting natural enemies and pollinators. The USDA, National Agricultural Library, DigiTop Navigator platform was used to search the literature for harmful versus nonharmful effects of quercetin on insect behavior, physiology, and life history parameters. Quercetin effects were evaluated on herbivores in ﬁve insect orders, natural enemies in two orders, and pollinators in one order. Quercetin was signiﬁcantly more harmful to Hemiptera, Diptera, and Lepidoptera but signiﬁcantly more nonharmful to Coleoptera. Harmful and nonharmful effects to Orthoptera were indistinguishable. Quercetin had signiﬁcantly more harmful (than nonharmful) effects on herbivores when data from the ﬁve insect orders were combined. Quercetin concentration (mg/mL) did not signiﬁcantly affect these results. Quercetin was signiﬁcantly more nonharmful to natural enemies (Coleoptera and Hymenoptera, combined) and pollinators (Hymenoptera). This study suggests that quercetin could prevent herbivory without disrupting natural enemies and pollinators, but ﬁeld experiments are necessary to substantiate these results.

The literature was searched and data synthesized to test the hypothesis that quercetin can be utilized as an effective biocide to reduce herbivory without disrupting natural enemies and pollinators. The objectives of this review are to (1) compile the available evidence of harmful versus nonharmful effects of quercetin on herbivores, (2) estimate the influence of quercetin concentration on the observed effects on herbivores, and (3) determine the effects of quercetin on natural enemies and pollinators.
The United States Department of Agriculture (USDA), National Agricultural Library, online digital catalog system (DigiTop) Navigator platform, which includes research databases (such as AGRICOLA, BIOSIS Previews, CAB Abstracts, Scopus, Web of Science, Zoological Record, etc.) was used to retrieve abstracts and then the full text of manuscripts. The key words "quercetin and Hemiptera, Coleoptera, Diptera, Orthoptera, Lepidoptera, or Hymenoptera" were used to search for the effects of quercetin on insects. Only published research that tested pure quercetin was included in this study. Literature searches were restricted to these taxa (insect orders) to encompass herbivorous crop pests, natural enemies of crop pests, or pollinators of crop plants. To synthesize the available data, harmful and nonharmful effects of quercetin on insect species were tabulated. Harmful effects had negative consequences on behavior, physiology, and life history parameters, whereas non-harmful effects had positive or neutral consequences. Statistical analysis included a z-test for proportions to compare quercetin effects (harmful vs non-harmful) on herbivores (in five orders), natural enemies (in two orders), and pollinators (in one order). Second, a z-test compared quercetin effects on all herbivores, all taxa combined. A Mann-Whitney Rank Sum Test, with U statistic, compared the influence of chemical concentration (mg/mL), when available, on the observed effects (harmful vs nonharmful) on Ingestion of quercetin can induce the production of detoxification enzymes in nonadapted herbivores [9,10]. Ingestion of quercetin may increase the sensitivity of nonadapted herbivores to pesticides, which may or may not increase their ability to develop resistance to pesticides in the field. Quercetin ingestion may upregulate detoxification enzymes in pollinators (A. mellifera) with little or no harmful effects [23,24]. Quercetin ingestion can lessen the harmful effects of pesticide exposure on A. mellifera adult workers [23].
The literature was searched and data synthesized to test the hypothesis that quercetin can be utilized as an effective biocide to reduce herbivory without disrupting natural enemies and pollinators. The objectives of this review are to (1) compile the available evidence of harmful versus nonharmful effects of quercetin on herbivores, (2) estimate the influence of quercetin concentration on the observed effects on herbivores, and (3) determine the effects of quercetin on natural enemies and pollinators.
The United States Department of Agriculture (USDA), National Agricultural Library, online digital catalog system (DigiTop) Navigator platform, which includes research databases (such as AGRICOLA, BIOSIS Previews, CAB Abstracts, Scopus, Web of Science, Zoological Record, etc.) was used to retrieve abstracts and then the full text of manuscripts. The key words "quercetin and Hemiptera, Coleoptera, Diptera, Orthoptera, Lepidoptera, or Hymenoptera" were used to search for the effects of quercetin on insects. Only published research that tested pure quercetin was included in this study. Literature searches were restricted to these taxa (insect orders) to encompass herbivorous crop pests, natural enemies of crop pests, or pollinators of crop plants. To synthesize the available data, harmful and nonharmful effects of quercetin on insect species were tabulated. Harmful effects had negative consequences on behavior, physiology, and life history parameters, whereas non-harmful effects had positive or neutral consequences. Statistical analysis included a z-test for proportions to compare quercetin effects (harmful vs non-harmful) on herbivores (in five orders), natural enemies (in two orders), and pollinators (in one order). Second, a z-test compared quercetin effects on all herbivores, all taxa combined. A Mann-Whitney Rank Sum Test, with U statistic, compared the influence of chemical concentration (mg/mL), when available, on the observed effects (harmful vs nonharmful) on herbivores, five orders combined. A z-test was also used to compare quercetin effects on natural enemies (two orders combined) and pollinators (one order). Significant differences were indicated when p < 0.05. Statistical software, SigmaStat, interfaced within SigmaPlot 12.0, and JMP 14 were used for data analysis.
In summary, quercetin caused 0.304 and 0.696 proportional harmful and nonharmful effects, respectively, on herbivorous coleopterans (z = 2.652, p = 0.008, n = 23); nonharmful effects were predominant. There was variability in quercetin concentration and positive feeding responses between and within coleopteran species. For example, a concentration of 0.30 mg/mL stimulated feeding by P. japonica adults in one study; but a higher concentration of 30.22 mg/mL stimulated feeding of the same species in another study ( Table 1). Quercetin had harmful effects on oviposition by C. chinensis in one study but not in another; quercetin concentration was at least five times greater in the bioassay indicating reduced oviposition than in the one indicating neutral effects. At 1-10 mg/mL, quercetin had nonharmful (neutral) effects on oviposition and feeding behavior by A. grandis in one study, but nonharmful (positive) effects on feeding behavior at a lower concentration, 0.5 mg/mL, in another study.
Quercetin had harmful effects on development or body weight, i.e., growth, of noctuid larvae in most studies (Table 1). Effects on feeding behavior were variable, with nonharmful (positive) effects on S. frugiperda at low concentration (0.01 µg/cm 2 ) on treated foliage as well as nonharmful (neutral) effects on H. virescens and H. zea at low concentration (0.10%, w/w) in an artificial diet. Quercetin also had nonharmful (positive) effects on feeding behavior of the arctiid C. rudis at 0.005 mg/mL on treated foliage. Quercetin had harmful effects on development, body weight, or survival of L. dispar, B. mori, and O. nubilalis at a concentration ranging from 0.1-2% (w/w) in an artificial diet (Table 1).
In summary, quercetin caused 0.792 and 0.208 proportional harmful and nonharmful effects on lepidopterans, respectively. A statistical analysis indicated a significant difference between the two effects (z = 4.046, p < 0.001, n = 24); harmful effects were predominant.

Diptera (True Flies)
Dipteran species subjected to quercetin in bioassays included two tephritids Rhagoletis pomonella (Walsh) and Bactrocera cucurbitae (Coquillett), one drosophilid Drosophila melanogaster Meigen and one sciarid Lycoriella pleuroti Yang & Zhang. Records indicated harmful (negative) effects of quercetin on B. cucurbitae, R. pomonella, and L. pleuroti after direct physical bodily contact with the compound in test arenas, in an artificial diet or artificial culture media at variable quercetin concentrations. For example, quercetin at 0.05-3.1 mg/mL, in bioassays involving B. cucurbitae, reduced egg hatch rate, pupation, adult emergence, oviposition, and survival rate (Table 1). In two studies, quercetin had nonharmful (positive) effects on development time and fecundity of D. melanogaster larvae and adult females, respectively. Quercetin concentration ranged from 1.7% to 5.0% across these two studies. In summary of this section, quercetin caused 0.75 and 0.25 proportional harmful and nonharmful effects on dipterans, respectively. The statistical analysis indicated a significant difference between the two effects (z = 2.00, p = 0.046, n = 8); harmful effects were more prevalent.

Orthoptera (Grasshoppers)
Three acridid species were tested against quercetin in field cage and laboratory bioassays. These species included Calliptamus abbreviatus Ikonn, Oedaleus asiaticus Bey-Bienko, and Melanoplus sanguinipes (F.) ( Table 1). Quercetin had harmful (negative) effects on development and survival of C. abbreviatus nymphs at a concentration of 0.10 mg/mL. Quercetin concentrations ranging from 0.10-10 mg/mL significantly reduced growth/development and survival of O. asiaticus nymphs. In contrast, body weight and survival rate of M. sanguinipes nymphs were unaffected by quercetin at a concentration of ranging from 0.125-4.0% (w/w). In summary, quercetin caused 0.67 and 0.33 proportional harmful and nonharmful effects on orthopterans, respectively. A statistical analysis did not indicate a significant difference between the two effects (z = 1.155, p = 0.248, n = 6).

Summary of Herbivores
The sections above indicated that quercetin caused more harmful effects to Hemiptera, Lepidoptera, and Diptera but more nonharmful effects to Coleoptera. In concluding the herbivore section, quercetin caused 0.618 and 0.382 proportional harmful and nonharmful effects on herbivores, respectively, across the five insect orders combined. The two effects were significantly different (z = 2.744, p = 0.006, n = 68); harmful effects were predominant. Quercetin concentration (mg/mL) did not significantly influence the observed harmful and nonharmful effects on herbivores, based on pooling of data, when available, across the five insect orders (U = 105.50; p = 0.583; n = 20 for harmful effects; n = 12 for nonharmful effects). Median values with 25% and 75% confidence intervals were 0.56 mg/mL (0.10, 2.76) for harmful effects and 1.00 mg/mL (0.09, 3.02) for nonharmful effects. Specific harmful effects of quercetin on herbivores, of five orders combined, are illustrated in Figure 2. Quercetin frequently affected survival rate and development/growth. The sections above indicated that quercetin caused more harmful effects to Hemiptera, Lepidoptera, and Diptera but more nonharmful effects to Coleoptera. In concluding the herbivore section, quercetin caused 0.618 and 0.382 proportional harmful and nonharmful effects on herbivores, respectively, across the five insect orders combined. The two effects were significantly different (z = 2.744, p = 0.006, n = 68); harmful effects were predominant. Quercetin concentration (mg/mL) did not significantly influence the observed harmful and nonharmful effects on herbivores, based on pooling of data, when available, across the five insect orders (U = 105.50; p = 0.583; n = 20 for harmful effects; n = 12 for nonharmful effects). Median values with 25% and 75% confidence intervals were 0.56 mg/mL (0.10, 2.76) for harmful effects and 1.00 mg/mL (0.09, 3.02) for nonharmful effects. Specific harmful effects of quercetin on herbivores, of five orders combined, are illustrated in Figure 2. Quercetin frequently affected survival rate and development/growth.

Predators and Parasitoids
Limited research has been published on the effects of quercetin on natural enemies, i.e., predators and parasitoids. Quercetin had nonharmful (positive) effects on the coc-

Natural Enemies Predators and Parasitoids
Limited research has been published on the effects of quercetin on natural enemies, i.e., predators and parasitoids. Quercetin had nonharmful (positive) effects on the coccinellid Coleomegilla maculata DeGeer, an important predator of aphids and other soft-bodied herbivores in agroecosystems ( Table 2). In laboratory bioassays, females were attracted to quercetin (1 mg, 98% pure powder) and stimulated to lay more egg clutches in test cages than control cages. Quercetin had nonharmful (positive) effects on a trichogrammatid Trichogramma chilonis Ishii, a parasitoid of lepidopteran eggs. At low quercetin concentration (0.01-0.03 mg), T. chilonis adults were attracted to treated substrates and females were stimulated to oviposit into host eggs in laboratory and semi-field bioassays. 1 Quercetin had harmful effects (negative (−−)) or non-harmful effects (positive (++) or neutral (oo)) on insects in comparison to control. 2 Effective concentration (concn) was the minimum concentration that caused a significant effect on insect behavior, physiology, or a life history parameter.
In a summary of the natural enemy section, quercetin caused 0.00 and 1.0 proportional harmful and nonharmful effects, respectively, on natural enemies (Coleoptera and Hymenoptera taxa combined). The harmful and nonharmful effects were significantly different (z = 2.828, p = 0.005; n = 4); nonharmful effects were more prevalent. Specific nonharmful (all positive) effects of quercetin on two natural enemies are illustrated in Figure 3. Oviposition behavior was affected most frequently. nonharmful (all positive) effects of quercetin on two natural enemies are illustrated in Figure 3. Oviposition behavior was affected most frequently. 1 Quercetin had harmful effects (negative (−−)) or non-harmful effects (positive (++) or neutral (oo)) on insects in comparison to control. 2 Effective concentration (concn) was the minimum concentration that caused a significant effect on insect behavior, physiology, or a life history parameter.

Domesticated Honeybee (Hymenoptera)
Honeybee, A. mellifera, workers can contact pesticide residues while collecting and consuming pollen. Since quercetin is one of the main constituents in pollen, research has

Pollinators Domesticated Honeybee (Hymenoptera)
Honeybee, A. mellifera, workers can contact pesticide residues while collecting and consuming pollen. Since quercetin is one of the main constituents in pollen, research has addressed the interaction of quercetin and pesticides on the health of A. mellifera (Table 3). Ten studies were identified in the published literature; two studies indicated harmful effects of quercetin. For example, feeding A. mellifera adults a diet containing quercetin and an acaricide (for varroa mites) reduced survival rate. Feeding workers a diet containing quercetin and a fungicide reduced the ability of workers to produce energy, i.e., ATPs. All other studies indicated nonharmful effects of quercetin with or without incorporating pesticides in experimental designs (Table 3). Quercetin ameliorated harmful effects of pesticide exposure and increased A. mellifera survival in laboratory bioassays. Ingestion of fungicide residues with pollen or nectar reduced flight performance, i.e., wing beat frequency, of A. mellifera in an indoor flight mill experiment. Incorporating quercetin in the diet, restored wing beat frequency. Quercetin functioned as an attractant and feeding stimulant. Moreover, quercetin increased maturation of ovaries in A. mellifera workers.

Synthesis
The potential of using quercetin as an effective biocide to discourage insect herbivory without disrupting natural enemies and pollinators has been evaluated in this study. Quercetin had significantly more harmful (negative) than nonharmful (positive or neutral) effects on behavior and life history parameters of herbivorous species in three of five insect orders, i.e., Hemiptera, Diptera, and Lepidoptera. Quercetin had significantly more nonharmful effects on herbivores in the Coleoptera, which suggests that this flavonol has potential as a feeding stimulant or an attractant for herbivorous beetles. Quercetin had no significant effects (neither harmful nor nonharmful) on Orthoptera, perhaps due to a small sample size of published data. When the herbivore data were combined, quercetin had significantly more harmful than nonharmful effects. The most frequent specific harmful effects on herbivores were decreased survival and altered growth and development.
Chemical concentration (mg/mL) did not influence the outcome of the analysis of herbivore data. Note that not all concentration data were reported in mg/mL values, or was convertible to mg/mL, as indicated in Tables 1-3. Consequently, the lack of a significant relationship between concentration for data pooled across all herbivores and quercetin effects (harmful versus nonharmful) indicates that concentration, alone, cannot be used to predict an outcome. Other factors such as herbivore species, life stage examined, bioassay methods, and life history parameters tested, likely contributed to outcomes. Less data were available on the effects of quercetin on natural enemies and pollinators (see Table 2 and Table 3). Therefore, influence of chemical concentration on outcomes (harmful versus nonharmful effects) was not analyzed statistically for beneficials.
A comparison of harmful versus nonharmful effects of quercetin on natural enemies indicated nonharmful effects only. However, the data set was small; just two species had been subjected to quercetin in experiments in the published literature. Nevertheless, these

Synthesis
The potential of using quercetin as an effective biocide to discourage insect herbivory without disrupting natural enemies and pollinators has been evaluated in this study. Quercetin had significantly more harmful (negative) than nonharmful (positive or neutral) effects on behavior and life history parameters of herbivorous species in three of five insect orders, i.e., Hemiptera, Diptera, and Lepidoptera. Quercetin had significantly more nonharmful effects on herbivores in the Coleoptera, which suggests that this flavonol has potential as a feeding stimulant or an attractant for herbivorous beetles. Quercetin had no significant effects (neither harmful nor nonharmful) on Orthoptera, perhaps due to a small sample size of published data. When the herbivore data were combined, quercetin had significantly more harmful than nonharmful effects. The most frequent specific harmful effects on herbivores were decreased survival and altered growth and development.
Chemical concentration (mg/mL) did not influence the outcome of the analysis of herbivore data. Note that not all concentration data were reported in mg/mL values, or was convertible to mg/mL, as indicated in Tables 1-3. Consequently, the lack of a significant relationship between concentration for data pooled across all herbivores and quercetin effects (harmful versus nonharmful) indicates that concentration, alone, cannot be used to predict an outcome. Other factors such as herbivore species, life stage examined, bioassay methods, and life history parameters tested, likely contributed to outcomes. Less data were available on the effects of quercetin on natural enemies and pollinators (see Tables 2 and 3). Therefore, influence of chemical concentration on outcomes (harmful versus nonharmful effects) was not analyzed statistically for beneficials.
A comparison of harmful versus nonharmful effects of quercetin on natural enemies indicated nonharmful effects only. However, the data set was small; just two species had been subjected to quercetin in experiments in the published literature. Nevertheless, these results are encouraging because they provide evidence that quercetin could be used to manipulate the behavior of natural enemies. Stimulating oviposition behavior and manipulating oviposition site selection in predators and parasitoids has relevance to mass production and augmentative biological control.
The results of the analysis of the effects of quercetin on pollinators, i.e., A. mellifera was also encouraging. Although A. mellifera was the only species challenged with quercetin, based on reliable published data, the observation of significantly more nonharmful (than harmful) effects on workers suggests that quercetin, when applied against pests, is not expected to harm foraging bees.
Quercetin is a compound with relatively low volatility, high molecular weight, and very low vapor pressure [19]. Thus, detection of quercetin molecules must primarily involve physical contact with gustatory receptors and/or mechanoreceptors, rather than olfactory receptors, on the mouthparts or antennae of the insect. This fact could limit the applicability of a push-pull strategy to push pest herbivores away from crop plants and pull beneficial insects toward crop plants [66]. Insects must physically contact quercetin molecules on the plant surface before a change in behavior would ensue.

Conclusions
This study has highlighted evidence that quercetin affects the behavior and life history of herbivores, natural enemies, and pollinators. This study suggests that quercetin has utility in a modified push-pull strategy to deter pest herbivores, e.g., aphids, from crop plants and arrest natural enemies (ladybird beetles or aphid parasitoids) and pollinators (honeybees) on crop plants. Future research using more species and a multitrophic approach (the interaction of crop plant, pest, natural enemy, and pollinator) would be more informative than examining quercetin effects on a single trophic level. Additionally, testing the functionality of quercetin under semi-field (greenhouse, high tunnel, nursery) or open field conditions is necessary. Applying quercetin directly onto the leaf surface of non-engineered plants could boost the density of natural enemies (for aphid control) and pollinators (for pollination services) for a net benefit to the crop plant. Alternatively, crop plants could be metabolically engineered to produce and release greater quantities of quercetin. Research on flavonoid mass production via metabolic engineering of plants and microbes is underway [67][68][69]. Perhaps other common flavonoids, e.g., kaempferol [70], could be utilized as an alternative compound to alleviate insect resistance development arising from the overuse of quercetin. The utility of quercetin as a less-toxic natural product to manage herbivorous pests without disrupting the activity of natural enemies and pollinators on greenhouse, high tunnel, or nursery grown crops could become a reality.