Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Oral administration of phytochemicals protects honey bees against cognitive and motor impairments induced by the insecticide fipronil

  • Lina M. García ,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Visualization, Writing – original draft, Writing – review & editing (AJR); (LMG)

    Affiliation Departamento de Biología, Facultad de Ciencias Naturales, Universidad del Rosario, Bogotá, Colombia

  • Valentina Caicedo-Garzón,

    Roles Data curation, Formal analysis, Investigation, Methodology, Writing – review & editing

    Current address: Department of Molecular, Cellular and Developmental Biology, University of Michigan, Ann Arbor, Michigan, United States of America

    Affiliation Departamento de Biología, Facultad de Ciencias Naturales, Universidad del Rosario, Bogotá, Colombia

  • Andre J. Riveros

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Project administration, Resources, Supervision, Validation, Writing – original draft, Writing – review & editing (AJR); (LMG)

    Affiliations Departamento de Biología, Facultad de Ciencias Naturales, Universidad del Rosario, Bogotá, Colombia, Department of Neuroscience, College of Science, University of Arizona, Tucson, Arizona, United States of America


Pollution produced by exposure to pesticides is a major concern for food security because the negative impacts on pollinators. Fipronil, an insecticide broadly used around the globe has been associated with the ongoing decline of bees. With a characteristic neuroactive toxicodynamic, fipronil leads to cognitive and motor impairments at sublethal dosages. Despite of regional bans, multilevel strategies are necessary for the protection of pollinators. Recent evidence suggests that specific nutrients in the diets of bees may induce protection against insecticides. Here, we evaluated whether the administration of three phytochemicals, namely rutin, kaempferol and p-coumaric acid provide protection to the Africanized honey bee Apis mellifera against oral administration of realistic dosages of fipronil. We tested the potential impairment produced by fipronil and the protection induced by the phytochemicals in learning, 24h memory, sucrose sensitivity and motor control. We found that the administration of fipronil induced a concentration-dependent impairment in learning and motor control, but not 24h memory or sucrose sensitivity across a 24h window. We also found that the administration of rutin, p-coumaric acid, kaempferol and the mixture was innocuous and generally offered protection against the impairments induced by fipronil. Overall, our results indicate that bees can be prophylactically protected against insecticides via nutrition, providing an alternative to the ongoing conflict between the use of insecticides and the decline of pollinators. As the studied phytochemicals are broadly present in nectar and pollen, our results suggest that the nutritional composition, and not only its production, should be considered when implementing strategies of conservation via gardens and co-cropping.


Pesticide pollution is a major concern due to its impact on beneficial species and public health. For example, bees are central to crop pollination and food security [1], yet endure individual and colony challenges due to sublethal exposures to pesticides used for pest control [2]. Among sublethal effects, impairment of cognitive functions is particularly critical because of its cryptic yet deteriorating impact [3]. Following robust evidence, certain neuroactive pesticides, including neonicotinoids and fipronil (a phenylpyrazole) were first partially and then permanently banned in the European Union [4]. However, the accelerated decline of managed and wild populations and the timelines for the development and implementation of regulatory policies demand urgent complementary strategies.

In this context, the idea of reducing the impact of pesticides and protecting bees through dietary supplementation with phytochemicals has emerged as an alternative [5]. For example, p-coumaric acid extends the longevity of honey bees [68], and gallic acid, abscisic acid, and quercetin improve pathogen tolerance, immunocompetence, and antioxidant activity [6, 9]. Also, some phytochemicals present in pollen improve the tolerance of bees to pesticides [10] and upregulate gene expression for cytochrome P450, enhancing detoxification [11]. Similarly, quercetin and p-coumaric acid increase the survival of honey bees exposed to pesticides [7, 10]. Moreover, the prophylactic administration of phytochemicals such as rutin, kaempferol, and p-coumaric acid are investigated as alternative treatments for human neurological disorders [12, 13].

Thus, nutrition with phytochemicals that are readily available to bees in pollen and nectar seems an important component positively influencing the physiology of bees that are exposed to pesticides. However, supplementation with phytochemicals in bees requires more evidence in the context of protection against cognitive impairments. We have found that the flavonoids rutin and quercetin are associated with the protection of the Africanized honey bees against oral administration of imidacloprid, a first-generation neonicotinoid (LM. Garcia, JJ. Sutachan, CF. Morantes-Ariza, V. Caicedo-Garzón, SL. Albarracin, AJ. Riveros, unpublished). Recently, we also found that the administration of rutin protects the performance of bumble bees against impairment by imidacloprid and fipronil [14]. These results open new questions on the scope of protection against pesticides with different toxicodynamic and the potential of other plant-derived secondary metabolites as neuroprotectors of bees. Here, we tackle these questions by testing the protection offered by three phytochemicals, namely rutin, p-coumaric acid, and kaempferol, against the impairments induced by the oral administration of fipronil, a broadly used insecticide.

Our goal was twofold. First, we aimed to evaluate the potential impairment of a sublethal exposure to a commercial form of fipronil on learning and memory, latency of a learned response, sucrose sensitivity, and motor activity. We focused on fipronil because sublethal doses affect the use of energy [15] and the reproductive [16], motor [17], and cognitive performances [18]. Pharmacologically, fipronil acts by inhibiting the GABAergic control (an inhibitory system itself), leading to excitation of neuronal activity [19]. Also, fipronil negatively impacts the mushroom bodies, major regions of multimodal integration within the central brain of insects [20], and the central complex, a region recognized for its role in navigation and motor control [21, 22]. Hence, exposure to fipronil ultimately impacts foraging by the individual and the colony. Lastly, by selecting a commercial form of fipronil, we aimed to test a realistic scenario for impairment and, eventually, protection.

Second, we aimed to test whether the administration of rutin, kaempferol and p-coumaric acid, separately and in combination, prevents cognitive and motor impairments induced by fipronil. Importantly, these phytochemicals found in pollen and nectar can activate signaling pathways for neuronal survival, synaptic plasticity, and detoxification [2325]. However, their protective effect on cognitive abilities has not been tested in honey bees, separately or combined, (but see [26, 27]). Importantly, since bees may be regularly exposed to these molecules through the pollen and nectar collected, our results highlight the potential to use plant-derived molecules as nutraceutical agents for pollinators, and the importance of high diversity of plants. Moreover, our results may help improving conservation strategies through identification of floral resources rich in specific metabolites.


Determination of dosage of fipronil

We found that the administration of fipronil impaired the performance of bees (Kruskal-Wallis’ test: χ24 = 68.82, P < 0.0001; Fig 1A). Bees in all the groups exposed to fipronil exhibited learning scores significantly lower than Control bees (Steel method: Fip-5: z = -3.51, P = 0.0017; Fip-11: z = -5.13, P = 0.0001; Fip-57: z = -5.87, P = 0.0001; Fip-114: z = -6.84, P = 0.0001). Also, the latency of cPER was significantly impacted by the administration of fipronil and differed between bees across treatments (Kruskal-Wallis’ test: χ24 = 20.20, P = 0.0005). Bees in the Fip-5, Fip-57 and Fip-114 groups exhibited significantly longer latencies than Control bees (Dunnett’s method: Fip-5: LSD = 0.04, P = 0.040; Fip-57: LSD = 0.27, P = 0.002; Fip-114: LSD = 0.42, P < 0.0001). In contrast, bees in the Fip-11 group exhibited latencies that were not significantly different from Control (Dunnett’s method: Fip-14: LSD = -0.12, P = 0.152).

Fig 1. Effect of concentration of fipronil on acquisition and retention of an olfactory stimulus.

(A) Learning score as the summatory of cPER across training trials. (B) Retention as percentage of cPER after 24h. Statistical comparisons were conducted between performance at the last training trial (as indicated by symbol) and performance at 24h (as indicated by the bars). Percentages of cPER at the last training trial are presented for comparison. Sample sizes are indicated within each bar. Learning score bars display standard error of the mean. Statistically significant planned comparisons with respect to the control group are indicated (*). Doses used for this experiment were as follow: Control (1M sucrose water), Fip: 0.04ng/bee (Fip-5). 0.1ng/bee (Fip-11), 0.5ng/bee (Fip-57) and 1ng/bee (Fip-114).

After 24 h, the percentage of cPER relative to the last training trial was not significantly different within each treatment (Fischer’s exact test for the comparison between last training trial and 24h test: Control: P = 0.84, Fip-5: P = 0.99, Fip-11: P = 0.98, Fip-57: P = 0.99, Fip-114: P = 0.98; Fig 1B), indicating that memory retention was not affected. Nevertheless, the differences across treatments observed during acquisition were maintained after 24h (Likelihood ratio: χ2 = 29.40, P < 0.0001; Fig 1B) with a dose-dependent decrease in percentage of bees that exhibited a conditioned PER: Control (89.1%), Fip-5 (58.3%), Fip-11 (50%), Fip-57 (37.5%), Fip-114 (37.1%). In all groups receiving fipronil, bees exhibited a probability of a conditioned PER that was significantly lower than the observed in Control bees (Adjusted Wald test for two proportions relative to Control group: Fip-5: 0.31, P = 0.0012; Fip-11: 0.39, P = 0.0002; Fip-57 = 0.51, P = 0.0002; Fip-114 = 0.52, P = 0.0002). Similarly, the odds of remembering were estimated to be always higher for Control bees and dose dependent: 5.8 times relative to Fip-5 bees (95% CI: 1.72–20.18), 8.2 times relative to Fip-11 bees (95% CI: 2.33–29.10), 13.7 times relative to Fip-57 bees (95% CI: 3.89–48.5), and 13.9 times relative to Fip-114 bees (95% CI: 4.02–48.42).

Evaluation of the protective effect of the prophylactic administration of phytochemicals

Olfactory conditioning of the proboscis extension response.

Subsequent experiments were conducted using a dose of 1 ng/bee of fipronil as it significantly impaired learning acquisition and maintained the impairment after 24 h without entirely abolishing the performance (Fig 2).

Fig 2. Effect of the administration of p-coumaric acid (p-CA), kaempferol (Kaemp), rutin (Rut) and their combination (Mix), and fipronil (Fip) on acquisition and retention of an olfactory stimulus.

(A) Learning score as the summatory of cPER across training trials (B) Retention as percentage of cPER after 24h. Statistical comparisons were conducted between performance at the last training trial (as indicated by symbol) and performance at 24h (as indicated by the bars). Percentages of cPER at the last training trial are presented for comparison. Sample sizes are indicated within each bar. Learning score bars display standard error of the mean. Statistically significant planned comparisons with respect to the respective control group are indicated (*). Statistically significant planned comparisons with respect to the fipronil group are indicated (#). Doses used for this experiment were as follow: Control (1M sucrose water), p-CA (328ng/bee), Kaemp (374.9ng/bee), Rut (6ng/bee), Mix (328ng/bee of p-CA + 374.9 ng/bee of Kaemp + 6ng/bee of Rut), Fip: 1ng/bee.

We found that the learning score significantly varied between bees exposed to different feeding treatments (mean score ± s.e.m: Control: 6.0±0.26, Rut = 7.1±0.21, Kaemp = 6.1±0.21, p-CA = 7.4±0.20, Mix = 6.5±0.20, Fip = 4±0.12, Rut+Fip = 5.8±0.26, Kaemp+Fip = 4.1±0.3, p-CA+Fip = 6.2±0.2, Mix+Fip = 5.8±0.2. Kruskal-Wallis’ test: χ29 = 144.23, P < 0.0001; Fig 2A). The performance of bees fed with all the phytochemicals did not significantly differ from Control bees (Steel method test: Kaemp: z = -0.070, P = 1.0; Mix: z = -1.30, P = 0.80) or was significantly enhanced (Steel method test: p-CA: z = 3.64, P = 0.003; Rut: z = 2.93, P = 0.037). In contrast, the administration of 1 ng/bee of fipronil impaired learning and the bees in the Fip group exhibited significantly lower learning scores than bees in the Control group (Steel method test for Fip vs. Control: Fip: z = -5.37, P = 0.0003).

Most importantly, prophylactic administration of most of the molecules or the mix offered full or partial protective effect. Generally, bees fed with a molecule and then with the insecticide exhibited significantly higher scores than bees fed only with fipronil (Steel method test: Rut+Fip vs. Fip: z = 4.92, P = 0.0003; p-CA+Fip vs. Fip: z = 6.83, P = 0.0003; Mix+Fip vs. Fip: z = 6.12, P = 0.0003), indicating successful protection. An exception to this was observed in the bees fed with kaempferol (Kaemp+Fip vs. Fip: z = 1.44, P = 0.67). However, bees fed with a molecule and then with the insecticide generally exhibited learning scores that were significantly lower than bees fed only with their respective control molecule (Steel method test: Rut+Fip vs. Rut: z = -3.66, P = 0.003; Kaemp+Fip vs. Kaemp: z = -4.25, P = 0.0005; p-CA+Fip vs. p-CA: z = -3.90, P = 0.0017; Fig 2). An exception occurred in bees fed with the mix and then with the pesticide, which exhibited learning scores that did not significantly differ from bees fed only with the mix (Steel method test: Mix+Fip vs. Mix: z = -2.52, P = 0.103). Thus, the administration of rutin and p-coumaric acid offered partial protection whereas the administration of the mix offered full protection.

Also, we found that the administration of the treatment significantly affected the latency of response (Kruskal-Wallis’ test: χ29 = 28.63, P = 0.001). In most cases, bees across groups exhibited latencies that did not significantly differ from their respective control bees (Steel method: Fip vs. Control: z = 0.58, P = 1.0; Rut vs. Control: z = -0.25, P = 1.0; Kaemp vs. Control: z = 2.21, P = 0.88; p-CA vs. Control: z = 1.44, P = 1.0; Mix vs. Control: z = 1.07, P = 1.0; Rut+Fip vs. Rut: z = 0.42, P = 1.0; p-CA+Fip vs. p-CA: z = 0.0, P = 1.0; Mix+Fip vs. Mix: z = 1.98, P = 0.88; Kaemp+Fip vs. Kaemp: z = 1.53, P = 1.0). Also, the latency of response did not differ significantly between the bees prophylactically administered with the compositions and the bees receiving fipronil (Steel method: Rut+Fip vs. Fip: z = -0.36, P = 1.0; p-CA+Fip vs. Fip: z = 0.70, P = 1.0; Mix+Fip vs. Fip: z = 2.04, P = 0.88; Kaemp+Fip vs. Fip: z = 2.57, P = 0.80).

After 24 h, we did not observe a significant difference in percentage of cPERs within each treatment relative to the last training trial (Fischer’s exact test: Control: P = 0.47; Fip: P = 0.76; p-CA: P = 0.98; Rut: P = 0.89; Kaemp: P = 0.99, Mix: P = 0.24; Rut+Fip: P = 0.99; p-CA+Fip: P = 0.98; Kaemp+Fip: P = 1.0; Mix+Fip: P = 0.96). Nevertheless, the probability of exhibiting a cPER was significantly different across treatment (Likelihood ratio: χ2 = 37.59, P < 0.0001; Fig 2B). The bees fed with phytochemicals exhibited a percentage of cPER that did not significantly differ from the bees in the Control group (Adjusted Wald test for two proportions: Rut vs. Control: 0.045, P = 0.55; p-CA vs. Control: 0.066, P = 0.44; Kaemp vs. Control: 0.055, P = 0.55; Mix vs. Control: 0.00, P = 1.0). In contrast, the administration of fipronil decreased the probability of exhibiting a cPER relative to Control bees but the difference was barely statistically significant (Adjusted Wald test for two proportions-Control vs. Fip: -0.181, P = 0.05). Importantly, the bees that received the prophylactic administration of most of the phytochemicals had a probability of exhibiting a cPER that did not significantly differ from their respective control bees (Adjusted Wald test for two proportions: Rut+Fip vs. Rut: 0.08, P = 0.36; Mix+Fip vs. Mix: 0.001, P = 1.0; p-CA+Fip vs. p-CA: -0.01, P = 0.86; Kaemp+Fip vs. Kaemp: 0.14, P = 0.113) but in one case was significantly higher than the probability of a cPER in the bees exposed to fipronil (Adjusted Wald test for two proportions: p-CA+Fip vs. Fip: 0.26, P = 0.0013). Also resembling the results observed during acquisition, the bees in some groups had a probability of exhibiting a cPER that did not significantly differ from Fip bees (Adjusted Wald test for two proportions: Kaemp+Fip vs. Fip: 0.02, P = 0.55; Mix+Fip vs. Fip: 0.17, P = 0.05; Rut+Fip vs. Fip: 0.14, P = 0.11).

Climbing assay.

We found that the ability to climb was not significantly affected by the feeding treatments (Likelihood ratio: χ2: 15.54, P = 0.08; Fig 3). Nevertheless, we pursued the planned comparisons motivated by the visual differences indicated by plotting the data (Fig 3). Interestingly, we found that the bees receiving fipronil tended to climb above the threshold significantly more often than bees in the Control group (Adjusted Wald test for two proportions-Control vs. Fip: 0.21 P = 0.02, Fig 3A). Also, the average speed of climbing of the bees that received fipronil was higher yet did not significantly differ from the speed of bees in the Control group (Steel method: mean speed±s.e.m. = Fipronil: 0.98±0.10 cm s-1, Control: 0.56±0.05 cm s-1; Fip vs. Control: z = 2.92, P = 0.067; Fig 3B).

Fig 3. Effect of the administration of phytochemicals and fipronil on motor performance during a climbing assay.

(A) Percentage of bees that crossed the 10 cm threshold. (B) Time (displayed as mean±s.e.m) to reach the threshold. A and B include the same individuals. Statistically significant planned comparisons are indicated with respect to bees in the Control group (*) and relative to bees in the Fip group (#). Only significant comparisons with *P<0.05 or # P<0.05 and indicated for better visualization. Doses used for this experiment were as follow: Control (1M sucrose water), p-CA (328ng/bee), Kaemp (374.9ng/bee), Rut (6ng/bee), Mix (328ng/bee of p-CA + 374.9 ng/bee of Kaemp + 6ng/bee of Rut), Fip: 1ng/bee.

In contrast, the bees fed with any of the phytochemicals did not exhibit significant differences in the number of bees reaching the threshold relative to individuals in the Control group (Adjusted Wald test for two proportions: Control vs. Kaemp: 0.027, P = 0.76; Control vs. Rut: 0.053, P = 0.55; Control vs. p-CA: 0.123, P = 0.19; Control vs. Mix: 0.053, P = 0.55; Fig 3) or in their speed relative to bees in the Control group (Kaemp: mean speed±s.e.m. = 0.54±0.05 cm s-1, z = -0.46, P = 0.99; Rut: mean speed±s.e.m. = 0.54±0.04 cm s-1, z = -0.58, P = 0.99; p-CA: mean speed±s.e.m. = 0.64±0.05 cm s-1, z = 0.90, P = 0.99; Mix: mean speed±s.e.m. = 0.52±0.05 cm s-1, z = -0.92, P = 0.99; Fig 3).

Importantly, the prophylactic administration of phytochemicals generally conferred a protective effect against the impairment induced by fipronil. The percentage of bees reaching the threshold did not significantly differ between bees receiving a phytochemical and the bees that received it and were subsequently exposed to fipronil (Adjusted Wald test for two proportions: Rut vs. Rut+Fip: 0.06, P = 0.55; Kaemp vs. Kaemp+Fip: 0.02, P = 0.76; p-CA vs. p-CA+Fip: 0.12, P = 0.19 and Mix vs. Mix + Fip: 0.1, P = 0.31; Fig 3A). This protective effect was further confirmed by the significant decrease to normality in the percentage of bees reaching the threshold compared to the bees that received only Fip (Adjusted Wald test for two proportions: Rut+Fip: 0.22, P = 0.02; Kaemp+Fip: 0.20, P = 0.02; p-CA+Fip: 0.21, P = 0.02 and Mix+Fip: 0.252, P = 0.011).

Lastly, the effective protection by phytochemicals was observed when measuring climbing speed. Relative to bees in the Fip group, prophylactically protected bees exhibited a significant reduction in speed (Kaemp+Fip: mean speed±s.e.m. = 0.47±0.04 cm s-1, z = -3.93, P = 0.003; Rut+Fip: mean speed±s.e.m. = 0.47±0.04 cm s-1, z = -3.86, P = 0.003; p-CA+Fip: mean speed±s.e.m. = 0.58±0.06 cm s-1, z = -3.63, P = 0.007 and Mix+Fip: mean speed±s.e.m. = 0.49±0.05 cm s-1, z = -4.15, P = 0.003); yet, the speed of protected bees did significantly differ from bees in their respective control groups (see speed for each group above; Rut vs. Rut+Fip: z = -0.56, P = 0.99; Kaemp vs. Kaemp+Fip: z = -0.69, P = 0.99; p-CA vs. p-CA+Fip: z = -1.85, P = 0.706; Mix vs Mix+Fip: z = -0.59, P = 0.99).

Sucrose sensitivity assay. Overall, we did not find any effect of the feeding treatments on the total score of bees (mean score±s.e.m.: Control = 20.6±1.4; Rut = 17.9±1.7; Kaemp = 19.3±1.4; p-CA = 18.1±1.7, Mix = 22.1±1.2; Fip = 20.3±1.3; Rut+Fip = 17.3±1.8; Kaemp + Fip = 18.1±1.6; p-CA+Fip = 19.8±1.5; Mix+Fip = 19.6±1.4; Kruskal-Wallis test: χ29 = 6.02, P = 0.74). Our more detailed comparisons relative to sensitivity exhibited by the bees in the Control group indicated that there were not significant deviations in the sensitivity to sucrose during the 3h, 5h or 27h (P > 0.05 in all comparisons). Nevertheless, we observed a slight increase in sensitivity in bees receiving the Mix of the phytochemicals, yet it rendered non-significant from the Control bees.


The pollution produced by insecticides is a growing concern around the globe. Paradoxically, insecticides are central tools for crop production, yet their impact on pollinators and other beneficial species may compromise productivity. While extant regulations may contribute to ameliorate the problem, multilevel approaches need to be implemented. Here, we aimed to evaluate the potential of prophylactic administration of phytochemicals for protecting Africanized honey bees against impairments produced by oral administration of fipronil, an insecticide commonly used in many countries. Our approach produced three key results. First, the oral administration of fipronil led to a dose-dependent impairment of learning and motor control, but not of memory retention or sucrose sensitivity. Second, the administration of the phytochemicals was innocuous when administered separate or as a mix. Third, the prophylactic administration of the phytochemicals generally, but not always, led to partial or full amelioration of the impairments produced by the fipronil.

The impairment in learning acquisition and motor control observed in our bees administered with fipronil may be explained by its toxicodynamic [28]. Fipronil acts as an antagonist of the GABAAR [29] and the glutamate-gated chloride channels [19, 30, 31], consequently blocking inward chloride currents that are key for inhibitory synapses [32]. Moreover, the exposure to fipronil may lead to neurodegenerative processes [33]. Hence, it is not surprising that fipronil impaired learning acquisition, and motor control, given the abundance of GABAergic pathways in brain centers involved in such processes (e.g., antennal lobes, the mushroom bodies, and the central complex [32, 34]). Thus, lower olfactory acquisition may derive from impairment of the olfactory coding [35, 36]. Interestingly, the impairment observed here differed from previous accounts in two directions. On the one hand, we did not observe an impact on memory retention, inferred from the maintained level of conditioned responses between the last training trial and the 24h test (Fig 2B). Whereas previous accounts (e.g., [18, 36]) have reported impairment of memory, the comparisons have been made relative to the retention of bees in control groups, which does not consider the variation from the last training trial and the testing event within a group. We argue that the low percentage of conditioned PER recorded after 24h derived from poor acquisition and not necessarily from impaired storage capacity (see also [14]). Similarly, we observed a dose-dependent impairment: the administration of the increasing dosages from 0.04ng/bee to 0.1ng/bee, 0.5ng/bee and 1ng/bee led to decreases of 35%, 43.9%, 57.9% and 58.4%, respectively, relative to bees in the Control group (see also Fig 1B). Previous accounts have shown that higher dosages of fipronil (including the one used here) may lead to lower impacts [35, 36]. These differences might derive from interspecific and geographic variations in sensitivity to pesticides (Africanized vs. European honey bee; e.g., [37, 38]), the administration of commercial vs. pure fipronil, and the time of exposure to the pesticide. Also, our results are in agreement with previous observations that the oral administration of fipronil does not impair the sensitivity to sucrose [18], suggesting that the effects observed here in learning acquisition derived from impacts on higher centers within the brain.

Importantly, a commonly reported effect of fipronil intoxication is an impaired motor control. The administration of fipronil caused an increase in walking speed. This result likely derives from the hyperexcitability and overstimulation of the nervous system byproduct of the blockage of the inhibitory pathways [30]. We recently observed a similar impact in the fruit fly Drosophila melanogaster after inducing a so-called parkinsonian phenotype following the oral administration of fipronil [39]. In this case, besides the mushroom bodies, the behavior observed may reflect the impact in the central complex, a GABAAR-rich brain region long recognized for its role in motor control [40]. Moreover, effective locomotion also requires visual information, which was likely impaired in the fruit fly through dopaminergic signaling [39]. Whether fipronil also impacted visual processing in our bees was not evaluated but requires further study because in a real scenario it might impair flight control.

More importantly, the phenolic acids were innocuous and generally protected the performance of the bees during the cognitive and motor tasks. On the one hand, the innocuity of the individual molecules and the mix likely derived from their presence in pollen and nectar and the low dosages used [6, 10, 41]. On the other hand, given the effects of fipronil, the cognitive protection conferred by the administration of the phytochemicals may derive from structural and/or physiological effects. Our behavioral level of analysis does not allow us to define a mechanistic explanatory framework. However, below we introduce four potential scenarios based on our current knowledge of the effect of some of the phytochemicals used. Testing these scenarios requires methods beyond behavioral analyses but could enrich our understanding of the mechanisms of protection.

First, phytochemicals may support protection via enhancement and maintenance of structural integrity. For instance, rutin modulates neuronal plasticity (e.g., growth of dendritic branches) and cell viability in vitro [42, 43]. Whether this regulation occurs within the mushroom bodies and in the central complex of bees would require analyses of the expression of the molecules and cascades known to be involved (e.g., protein kinase 1 ERK1/2, MAPK the mitogen-activated protein kinase (MAPK) cascade [42] the expression of the nuclear factor CREB [44]).

Second, phytochemicals may protect via a reduction of neurodegeneration. As other neuroactive pesticides, the impact of fipronil extends beyond the membrane activity, destabilizing mitochondrial function and eventually triggering apoptosis [28]). Mitochondrial protection may include the regulation of the oxidative stress induced by fipronil [45]. Phenolic acids contribute to the reduction of oxidative stress (reviewed by [46, 47]). In this direction, the phytochemicals may increase the activity level of intracellular antioxidant enzymes to stop the production of free radicals and ROS. Moreover, phenolic acids such as kaempferol, rutin, and p-coumaric acid inhibit the Nf-Kb signaling pathway, preventing neurotoxicity [48]. Finally, in rats, prophylactic administration of kaempferol decreases hippocampal neurodegeneration induced by the insecticide chlorpyrifos [49]. Similarly, p-coumaric acid reduces dopaminergic neurodegeneration and mitigates motor deficits in mice exposed to the insecticide rotenone [50].

Third, phytochemicals may interfere with the action of fipronil or favor its metabolization. On the one hand, we have shown that, in fruit flies, the administration of kaempferol protects against motor impairments produced by fipronil [39], and bioinformatic analyses suggest an allosteric regulation of the GABAAR. On the other hand, kaempferol, rutin, and p-coumaric acid upregulate CYP450 enzymes, favoring insecticide metabolization [51, 52]. Interestingly, hydroxyphenylacetic acid (4-HPAA; [5355], a metabolite of kaempferol and endogenous dopamine, activates enzymes, such as CYP450, involved in degrading xenobiotic molecules [56].

Interestingly, we observed a positive synergy between phytochemicals as predicted for the bees in the Mix treatment. This suggests that the protection offered by one molecule does not interfere with protection offered by the other molecules nor produces negative effects. We hypothesize a more effective protection derived from a convergent regulation of GABAAR, probably in multiple binding sites [39, 5759]. As we observed full protection following administration of the Mix, we suggest that binding sites for our molecules in the GABAAR might be different and regulate together the receptor.

We conclude that rutin, kaempferol and p-coumaric acid are most generally effective separated and as a mix to protect against impairments produced by oral administration of fipronil in Africanized honey bees. Our findings add evidence to the enormous potential of phenolic acids and other secondary metabolites derived from plants as an option to protect bees in environments polluted with insecticides. Whereas this approach may prove more effective for managed bees, the broad and extended variation of phenolic acids in pollen and nectar suggest that urban gardens may be planned with better precision and that increase biodiversity in agroecosystems may be beneficial for bees and other pollinators.

Materials and methods

Collection and maintenance of experimental subjects

We collected workers of the Africanized honey bee Apis mellifera leaving one of the hives at the Apiary of the Universidad del Rosario (Bogota, Colombia) between 0800h and 1000h. The hive had 10 brood combs and the queen bee was 3 to 4 weeks old age. Once in the lab, the bees were cold anesthetized by keeping them on a plastic container on ice and restrained in plastic tubes such that their heads were held in place and the bees could extend their proboscis. The bees were kept under the natural conditions of the laboratory (RH: 62%; T: 19°C) for the duration of the experiment. Bees were fed and maintained according to the assigned treatment.

Determination of dosage of fipronil

We collected and maintained 172 bees. We excluded two bees (one for Control and one for fip-57) after screening for outliers. Thus, we conducted our final analyses using 170 bees distributed across five treatments (Fig 1): Control (N = 37), Fip-5 (N = 36), Fip-11 (N = 30), Fip-57 (N = 32) and Fip-114 (N = 35).

Protective effect of the prophylactic administration of phytochemicals

Olfactory conditioning of the proboscis extension response.

We collected and maintained 858 bees. We excluded 143 bees after an outlier screening (Control = 14 bees, Rut = 14, Kaemp = 9, p-CA = 17, Mix = 10, Fip = 36, Rut+Fip = 11, p-CA+Fip = 12, Mix+Fip = 20). Thus, for the final analyses we included 715 bees across 10 treatments (Fig 2): Control (N = 79), Rut (N = 75), Kaemp (N = 77), p-CA (N = 70), Mix (N = 79), Fip (N = 54), Rut+Fip (N = 68), Kaemp+Fip (N = 76), p-CA+Fip (N = 72), Mix+Fip (N = 65).

Climbing assay.

We collected and maintained 792 bees. We excluded five bees after an outlier screening (p-CA = 2, Fip = 3). Thus, for the final analyses we included 787 bees across 10 treatments (Fig 3): Control (N = 84), Rut (N = 93), Kaemp (N = 81), p-CA (N = 80), Mix (N = 75), Fip (N = 69), Rut+Fip (N = 70), Kaemp+Fip (N = 79), p-CA+Fip (N = 79), Mix+Fip (N = 77).

Sucrose sensitivity assay.

We collected and prepared 377 bees distributed across 10 treatments. We excluded 11 bees that died during the experiment. We also excluded 20 bees after an outlier screen analysis. Thus, we conducted our final analyses using 346 bees distributed across ten treatments: Control (N = 35), Rut (N = 36), Kaemp (N = 35), p-CA (N = 34), Mix (N = 34), Fip (N = 37), Rut+Fip (N = 32), Kaemp+Fip (N = 34), p-CA+Fip (N = 34), Mix+Fip (N = 35).

The bees were kept under laboratory conditions (RH: 62% relative humidity; T: 19°C) for the duration of the experiment. Bees were fed and maintained according to the assigned treatment.

Experimental treatments

Determination of dosage of fipronil.

First, we determined a concentration of fipronil that would impact performance in learning and memory. We relied on reported realistic doses and times of effect of the pesticide [18, 60, 61] and defined four doses below the reported value: 0.04 ng/bee, 0.1 ng/bee, 0.5 ng/bee, and 1 ng/bee. Each dose corresponded to a treatment. Three hours before olfactory conditioning, bees were randomly allocated to one of these treatments and received a dose of 20 μL of fipronil (Astuto, Invesa S.A.) dissolved using sucrose-water (final concentration for sucrose: 1 M) to the final concentration of the assigned treatment. Thus, bees received one of five treatments: sucrose solution (Control), fipronil 5 nM (Fip-5: 0.04 ng/bee), fipronil 11.43 nM (Fip-11: 0.1 ng/bee), fipronil 57.18 nM (Fip-57: 0.5 ng/bee), fipronil 114.33 nM (Fip-114: 1 ng/bee). To deliver the doses, we stimulated the antennae with 1M sucrose solution to induce a reflexive extension of the proboscis and then fed the pesticide. Thus, the antennae were not contaminated with the pesticide during the administration.

Evaluation of the protective effect of the prophylactic administration of phytochemicals.

For testing of cognitive performance, we collected and restrained bees as described above, and randomly allocated them to receive one of five feeding treatments, all dissolved to a final sucrose concentration of 1M: sucrose solution (Control, Invitrogen: 15503022), rutin 1 μM (Rut; Sigma-Aldrich: R5143, concentrations according to [14, 42]; kaempferol 131 μM (Kaemp; Sigma-Aldrich: K0133, concentrations adapted from [6], p-coumaric acid 200 μM (p-CA; Sigma-Aldrich: C9008 concentrations following [8], rutin 1 μM + kaempferol 131 μM + p-coumaric acid 200 μM (Mix). In all cases, bees were fed six doses of 10 μL of the composition (two doses a day for three days). On the fourth day, bees were randomly assigned to one of two treatments three hours before training: 10 μL of 1M sucrose solution or 10 μL of fipronil 228.7 nM (Fip; 1ng/bee). Thus, bees belonged to one of ten treatments according to the composition administered: Control, Rut, Kaemp, p-CA, Mix, Fip, Rut+Fip, Kaemp+Fip, p-CA+Fip, Mix+Fip. These treatments were used for the evaluation of protection of learning and memory, motor activity, and sucrose sensitivity.

Behavioral assays

Olfactory conditioning of the proboscis extension response.

Three hours after the administration of treatment, bees were transferred to a rotatory training apparatus holding 12 bees in separate chambers (Fig 4A; after [62]. Bees were allowed to adapt for five minutes before starting a session. To evaluate learning and memory, we relied on standard protocols for the appetitive olfactory conditioning of the proboscis extension response (PER; [63]). Briefly, the bees were trained to learn the association between an olfactory stimulus (the conditioned stimulus, CS) and a reward (the unconditioned stimulus, US). During training, a clean stream of air generated by a pump (air flow: 1.3 L/min) was directed at the bee for 15 s (Fig 4C). Then, the olfactory stimulus was injected for 10 s into the air current. For injection of the stimulus, a second stream of air originating from the same pump was redirected to flow through a tube containing a piece of filter paper with 1 μl of 1-Hexanol (Sigma-Aldrich: 471402). Thus, the air pressure received by the bee was maintained at the same level. Six seconds after the onset of odor delivery, the antennae were stimulated with a solution of 1.5 M sucrose solution. Following the induced PER, the bee was allowed to drink for 4 s. The odor and reward delivery finished simultaneously. This sequence (hereafter a training trial; Fig 4B) was repeated every 10 minutes for eleven times (hereafter a training session) and the conditioned PER (cPER; a PER that follows the stimulation with the odor without antennal stimulation with the sucrose) and the latency of response (time between the onset of odor delivery and the onset of a response that led to a full extension of the proboscis) were recorded. Timing during a training session was monitored with a 2 Hz sound signal from a metronome (Metronome Beats App, Stonekick Limited 2021, London, UK). Once the training session was finished, bees were transferred to a plastic box, fed 10 μL of a 1 M sucrose solution for maintenance until the memory retention test.

Fig 4. General experimental protocol for administration of feeding treatments, olfactory conditioning, and test of motor control.

(A) Timeline of the overall protocol. (B) Sequence of training trials presentations within a training session and memory retention test. (C) Top view of the arrangement for conditioning of a single bee within an individual chamber. (D) Vertical arena for the climbing assay. A threshold line was drawn 10cm above the starting point.

Memory retention was tested 24 h after the last conditioning trial by presenting only the olfactory stimulus without reward during a single test. If a bee did not exhibit a conditioned PER, we stimulated the antennae with a drop of a 1 M sucrose solution to test for motivation. Training and testing were conducted by an experimenter blind to treatments. Bees that did not exhibit a PER to the stimulation with the sucrose solution during training or memory test were removed from the experiment.

Climbing assay.

Three hours after administration of a treatment, we evaluated motor activity. For the test, we used an arena (H: 20 cm, L: 20 cm, W: 0.5 cm; modified from [17]) with 15 lanes (width lane: 0.5 cm) and a visual mark that was used as a threshold (10 cm from the bottom; Fig 4D). One side of the arena was glass-covered, thus allowing the recording of the bees while moving in the lane. For the test, a single bee was gently introduced at the bottom of each lane (at least three bees per treatment could be recorded at the same time using the arena). For each treatment, we recorded the percentage of bees that crossed the threshold in 15 s and the average time for climbing the 10 cm. Thus, we could calculate a speed for each bee. We recorded up to a maximum of 120 s (iPhone 6s, Apple Inc., California). The tests were performed with the arena in a vertical position, thus stimulating locomotion by negative geotaxis.

Sucrose sensitivity assay.

We assessed responsiveness to sucrose by stimulating the antennae of the bees with ascending concentrations of sucrose [0.1% (w/w), 0.3%, 1%, 3%, 10%, and 30%; [64]]. Bees were collected, restrained in plastic tubes, and fed with 10 μL of a 1 M sucrose solution twice a day for three days. On the fourth day, we measured four times (0 h, 1.5 h, 3 h, and 5 h) the response to sucrose by exposing the bees to the ascending concentrations. Bees were stimulated with pure water before each presentation with the sucrose solution as a control for possible sensitization or habituation. The inter-trial time between the presentation of each concentration of the sucrose solution was on average 4 minutes. A positive response was recorded if the bee exhibited a PER after antennal stimulation with the solution. Thus, for each bee, we obtained a sensitivity score (0–7) that reflected the sensitivity to sucrose at each time interval. Importantly, the first measure (0 h) was conducted prior to the administration of fipronil and was used to normalize the responses across all times. Following the 0 h measure, we randomly assigned the bees to one of the ten treatments described above. After the last measurement (5h), 10 μL of 1 M sucrose solution was fed to the bees for maintenance until the next day for the final measure (27h after pesticide administration). Importantly, the measurements at 3h, 5h and 27h coincide with the period of olfactory conditioning and memory evaluation.

Statistical analyses

Learning was assessed from the total number of conditioned responses across the ten trials. Thus, each bee received a score between 0–10 (bees should not exhibit a learned response at the first trial). Learning scores were compared across treatments using a Kruskal-Wallis test as data were not normally distributed (evaluated with a Shapiro-Wilk W test). Multiple planned comparisons were conducted comparing learning scores relative to Control bees and fipronil bees using a non-parametric one or two-tailed Steel method depending upon predictions. Based on previous evidence, our predictions were: i. A decreased performance in bees fed with fipronil relative to bees in the Control group, ii. An enhancement of performance in bees prophylactically fed with phytochemicals and then exposed to fipronil relative to bees exposed only to fipronil, and iii. A performance of bees exposed to phytochemicals that would not differ from the performance of bees in the Control group. The following comparisons were of interest because they allow us to test: i. the innocuity of administering phytochemicals and the potential impairment produced by fipronil (Control vs all other groups), ii. the protection induced by the prophylactic treatment ([phytochemicals]+fipronil groups vs. fipronil group), iii. the full or partial protection ([phytochemicals]+fipronil groups vs. [phytochemical] groups). We excluded outliers by using a Mahalanobis distance test [65]. Latency of response was calculated using single or multiple cPERs. We Log transformed the latencies to normalize the distribution (Shapiro-Wilk W test after log transformation: P > 0.05 for all treatments for the null hypothesis that latencies data are from a normal distribution). We run multiple planned comparisons between all treatments and the respective control group (e.g., Rut+Fip vs. Rut, Rut vs. Control) using the Steel method. After 24 hours, the response to the retention test was recorded as a binary response to the sole presentation the conditioned stimulus. Memory was compared across groups using an overall Chi-Square and an Adjusted Wald test for the independent planned comparisons. For motor activity, the percentage of bees that crossed the 10 cm threshold was compared using a Chi-Square test followed by an Adjusted Wald test for the planned comparisons. We compared the speed of the bees using a one or two-tailed Steel method depending upon predictions. For sucrose sensitivity, we conducted comparisons of the total score of bees (0–28 for the sum across times) using a Kruskal-Wallis test. Further, to control for the random initial variation across treatments, we conducted an additional analysis on a normalized score using the sensitivity measured at time 0 h as a baseline [e.g., (score at 3h+1)/(score at 0h+1]. We relied on non-parametric tests using the Steel Method with control to test for deviations from the pattern of change in sucrose sensitivity observed in Control bees across time. In all cases error due to multiple comparisons was controlled using the Classical one stage False Discovery Rate method and adjusted p-values (q-values) are presented [66, 67]. All statistical analyses were conducted using JMP v.15.0 (SAS Institute Inc.).


We thank members of the Riveros’ Lab (Grupo de Investigaciones CANNON) for comments on the design of the work. We thank the technical support of Juan Manuel Quiroga, Lida Triviño, and Amparo Duarte.


  1. 1. van der Sluijs JP, Vaage NS. Pollinators and Global Food Security: the Need for Holistic Global Stewardship. Food Ethics. 2016;1: 75–91.
  2. 2. Manzoor F, Pervez M, Manzoor F, Pervez M. Pesticide Impact on Honeybees Declines and Emerging Food Security Crisis. Global Decline of Insects. IntechOpen; 2021. Available:
  3. 3. Chmiel JA, Daisley BA, Pitek AP, Thompson GJ, Reid G. Understanding the Effects of Sublethal Pesticide Exposure on Honey Bees: A Role for Probiotics as Mediators of Environmental Stress. Front Ecol Evol. 2020;8. Available:
  4. 4. Kathage J, Castañera P, Alonso-Prados JL, Gómez-Barbero M, Rodríguez-Cerezo E. The impact of restrictions on neonicotinoid and fipronil insecticides on pest management in maize, oilseed rape and sunflower in eight European Union regions. Pest Manag Sci. 2018;74: 88–99. pmid:28842940
  5. 5. Castle D, Alkassab AT, Bischoff G, Steffan-Dewenter I, Pistorius J. High nutritional status promotes vitality of honey bees and mitigates negative effects of pesticides. Sci Total Environ. 2022;806: 151280. pmid:34755614
  6. 6. Bernklau E, Bjostad L, Hogeboom A, Carlisle A, H. S. A . Dietary Phytochemicals, Honey Bee Longevity and Pathogen Tolerance. Insects. 2019;10: 14. pmid:30626025
  7. 7. Liao L-H, Pearlstein DJ, Wu W-Y, Kelley AG, Montag WM, Hsieh EM, et al. Increase in longevity and amelioration of pesticide toxicity by natural levels of dietary phytochemicals in the honey bee, Apis mellifera. PLOS ONE. 2020;15: e0243364. pmid:33296402
  8. 8. Mitton GA, Szawarski N, Mitton FM, Iglesias A, Eguaras MJ, Ruffinengo SR, et al. Impacts of dietary supplementation with p-coumaric acid and indole-3-acetic acid on survival and biochemical response of honey bees treated with tau-fluvalinate. Ecotoxicol Environ Saf. 2020;189: 109917. pmid:31776030
  9. 9. Alaux C, Ducloz F, Crauser D, Le Conte Y. Diet effects on honeybee immunocompetence. Biol Lett. 2010;6: 562–565. pmid:20089536
  10. 10. Johnson RM, Mao W, Pollock HS, Niu G, Schuler MA, Berenbaum MR. Ecologically Appropriate Xenobiotics Induce Cytochrome P450s in Apis mellifera. PLOS ONE. 2012;7: e31051. pmid:22319603
  11. 11. Li X, Deng Z, Chen X. Regulation of insect P450s in response to phytochemicals. Curr Opin Insect Sci. 2021;43: 108–116. pmid:33385580
  12. 12. Shahidi F, Yeo J. Bioactivities of Phenolics by Focusing on Suppression of Chronic Diseases: A Review. Int J Mol Sci. 2018;19: 1573. pmid:29799460
  13. 13. Pandey KB, Rizvi SI. Plant polyphenols as dietary antioxidants in human health and disease. Oxid Med Cell Longev. 2009;2: 270–278. pmid:20716914
  14. 14. Riveros AJ, Gronenberg W. The flavonoid rutin protects the bumble bee Bombus impatiens against cognitive impairment by imidacloprid and fipronil. J Exp Biol. 2022;225: jeb244526. pmid:36000283
  15. 15. Nicodemo D, Maioli MA, Medeiros HCD, Guelfi M, Balieira KVB, De Jong D, et al. Fipronil and imidacloprid reduce honeybee mitochondrial activity. Environ Toxicol Chem. 2014;33: 2070–2075. pmid:25131894
  16. 16. Kairo G, Poquet Y, Haji H, Tchamitchian S, Cousin M, Bonnet M, et al. Assessment of the toxic effect of pesticides on honey bee drone fertility using laboratory and semifield approaches: A case study of fipronil. Environ Toxicol Chem. 2017;36: 2345–2351. pmid:28224659
  17. 17. Zaluski R, Kadri SM, Alonso DP, Martins Ribolla PE, de Oliveira Orsi R. Fipronil promotes motor and behavioral changes in honey bees (Apis mellifera) and affects the development of colonies exposed to sublethal doses. Environ Toxicol Chem. 2015;34: 1062–1069. pmid:25703042
  18. 18. El Hassani AK, Dacher M, Gauthier M, Armengaud C. Effects of sublethal doses of fipronil on the behavior of the honeybee (Apis mellifera). Pharmacol Biochem Behav. 2005;82: 30–39. pmid:16102801
  19. 19. Ikeda T, Zhao X, Kono Y, Yeh JZ, Narahashi T. Fipronil Modulation of Glutamate-Induced Chloride Currents in Cockroach Thoracic Ganglion Neurons. NeuroToxicology. 2003;24: 807–815. pmid:14637375
  20. 20. Heisenberg M. What Do the Mushroom Bodies Do for the Insect Brain? An Introduction. Learn Mem. 1998;5: 1–10. pmid:10454369
  21. 21. Pfeiffer K, Homberg U. Organization and Functional Roles of the Central Complex in the Insect Brain. Annu Rev Entomol. 2014;59: 165–184. pmid:24160424
  22. 22. Heinze S, Pfeiffer K. Editorial: The Insect Central Complex—From Sensory Coding to Directing Movement. Front Behav Neurosci. 2018;12. Available: pmid:30104965
  23. 23. Baptista FI, Henriques AG, Silva AMS, Wiltfang J, da Cruz e Silva OAB. Flavonoids as Therapeutic Compounds Targeting Key Proteins Involved in Alzheimer’s Disease. ACS Chem Neurosci. 2013;5: 83–92. pmid:24328060
  24. 24. Williams RJ, Spencer JPE, Rice-Evans C. Flavonoids: antioxidants or signalling molecules? Free Radic Biol Med. 2004;36: 838–849. pmid:15019969
  25. 25. Rahman MdA, Rahman MDH, Biswas P, Hossain MdS, Islam R, Hannan MdA, et al. Potential Therapeutic Role of Phytochemicals to Mitigate Mitochondrial Dysfunctions in Alzheimer’s Disease. Antioxidants. 2020;10: 23. pmid:33379372
  26. 26. Mao W, Schuler MA, Berenbaum MR. A dietary phytochemical alters caste-associated gene expression in honey bees. Sci Adv. 2015;1: e1500795. pmid:26601244
  27. 27. Liao L-H, Wu W-Y, Berenbaum MR. Impacts of Dietary Phytochemicals in the Presence and Absence of Pesticides on Longevity of Honey Bees (Apis mellifera). Insects. 2017;8: 22. pmid:28216580
  28. 28. Farder-Gomes CF, Fernandes KM, Bernardes RC, Bastos DSS, Oliveira LL de, Martins GF, et al. Harmful effects of fipronil exposure on the behavior and brain of the stingless bee Partamona helleri Friese (Hymenoptera: Meliponini). Sci Total Environ. 2021;794: 148678. pmid:34225147
  29. 29. Hosie AM, Baylis HA, Buckingham SD, Sattelle DB. Actions of the insecticide fipronil, on dieldrin-sensitive and- resistant GABA receptors of Drosophila melanogaster. Br J Pharmacol. 1995;115: 909–912.
  30. 30. Zhao X, Yeh JZ, Salgado VL, Narahashi T. Fipronil Is a Potent Open Channel Blocker of Glutamate-Activated Chloride Channels in Cockroach Neurons. J Pharmacol Exp Ther. 2004;310: 192–201. pmid:15014137
  31. 31. Barbara Z, J R, M G, B G. Acetylcholine, GABA and glutamate induce ionic currents in cultured antennal lobe neurons of the honeybee, Apis mellifera. J Comp Physiol A Neuroethol Sens Neural Behav Physiol. 2005;191. pmid:16044331
  32. 32. Grünewald B. Cellular Physiology of the Honey Bee Brain. In: Galizia CG, Eisenhardt D, Giurfa M, editors. Honeybee Neurobiology and Behavior: A Tribute to Randolf Menzel. Dordrecht: Springer Netherlands; 2012. pp. 185–198. Available:
  33. 33. Jacob CRO, Soares HM, Nocelli RCF, Malaspina O. Impact of fipronil on the mushroom bodies of the stingless bee Scaptotrigona postica. Pest Manag Sci. 2015;71: 114–122. pmid:24648377
  34. 34. Raccuglia D, Mueller U. Focal uncaging of GABA reveals a temporally defined role for GABAergic inhibition during appetitive associative olfactory conditioning in honeybees. Learn Mem. 2013;20: 410–416. pmid:23860600
  35. 35. Decourtye A, Devillers J, Genecque E, Menach KL, Budzinski H, Cluzeau S, et al. Comparative Sublethal Toxicity of Nine Pesticides on Olfactory Learning Performances of the Honeybee Apis mellifera. Arch Environ Contam Toxicol. 2005;48: 242–250. pmid:15750780
  36. 36. El Hassani AK, Dupuis JP, Gauthier M, Armengaud C. Glutamatergic and GABAergic effects of fipronil on olfactory learning and memory in the honeybee. Invert Neurosci. 2009;9: 91. pmid:19851797
  37. 37. Devillers J, Decourtye A, Budzinski H, Pham-Delègue MH, Cluzeau S, Maurin G. Comparative toxicity and hazards of pesticides to Apis and non-Apis bees. A chemometrical study. SAR QSAR Environ Res. 2003;14: 389–403. pmid:14758982
  38. 38. Arena M, Sgolastra F. A meta-analysis comparing the sensitivity of bees to pesticides. Ecotoxicology. 2014;23: 324–334. pmid:24435220
  39. 39. Ramírez-Moreno DM, Lubinus KF, Riveros AJ. The flavonoid kaempferol protects the fruit fly Drosophila melanogaster against the motor impairment produced by exposure to the insecticide fipronil. J Exp Biol. 2022;225: jeb244556. pmid:36172773
  40. 40. Mustard JA, Jones L, Wright GA. GABA signaling affects motor function in the honey bee. J Insect Physiol. 2020;120: 103989. pmid:31805284
  41. 41. Irwin RE, Cook D, Richardson LL, Manson JS, Gardner DR. Secondary compounds in floral rewards of toxic rangeland plants: impacts on pollinators. J Agric Food Chem. 2014;62: 7335–7344. pmid:24766254
  42. 42. García LM. Mecanismos de acción cognitiva y neuronal del flavonol rutina. Masther, Pontificia Universidad Javeriana. 2019. Available:
  43. 43. Spencer JPE. Food for thought: the role of dietary flavonoids in enhancing human memory, learning and neuro-cognitive performance. Proc Nutr Soc. 2008;67: 238–252. pmid:18412998
  44. 44. Moghbelinejad S, Nassiri-Asl M, Farivar TN, Abbasi E, Sheikhi M, Taghiloo M, et al. Rutin activates the MAPK pathway and BDNF gene expression on beta-amyloid induced neurotoxicity in rats. Toxicol Lett. 2014;224: 108–113. pmid:24148604
  45. 45. Ueda T, Ito T, Kurita H, Inden M, Hozumi I. p-Coumaric Acid Has Protective Effects against Mutant Copper–Zinc Superoxide Dismutase 1 via the Activation of Autophagy in N2a Cells. Int J Mol Sci. 2019;20: 2942. pmid:31208129
  46. 46. Rudrapal M, Khairnar SJ, Khan J, Dukhyil AB, Ansari MA, Alomary MN, et al. Dietary Polyphenols and Their Role in Oxidative Stress-Induced Human Diseases: Insights Into Protective Effects, Antioxidant Potentials and Mechanism(s) of Action. Front Pharmacol. 2022;13. Available: pmid:35237163
  47. 47. Xu P, Wang S, Yu X, Su Y, Wang T, Zhou W, et al. Rutin improves spatial memory in Alzheimer’s disease transgenic mice by reducing Aβ oligomer level and attenuating oxidative stress and neuroinflammation. Behav Brain Res. 2014;264: 173–180. pmid:24512768
  48. 48. Yoon J-H, Youn K, Ho C-T, Karwe MV, Jeong W-S, Jun M. p-Coumaric acid and ursolic acid from Corni fructus attenuated β-amyloid(25–35)-induced toxicity through regulation of the NF-κB signaling pathway in PC12 cells. J Agric Food Chem. 2014;62: 4911–4916. pmid:24815946
  49. 49. Hussein RM, Mohamed WR, Omar HA. A neuroprotective role of kaempferol against chlorpyrifos-induced oxidative stress and memory deficits in rats via GSK3β-Nrf2 signaling pathway. Pestic Biochem Physiol. 2018;152: 29–37. pmid:30497708
  50. 50. Dolrahman N, Mukkhaphrom W, Sutirek J, Thong-Asa W. Benefits of p-coumaric acid in mice with rotenone-induced neurodegeneration. Metab Brain Dis. 2023;38: 373–382. pmid:36308586
  51. 51. Mao W, Schuler MA, Berenbaum MR. CYP9Q-mediated detoxification of acaricides in the honey bee (Apis mellifera). Proc Natl Acad Sci U S A. 2011;108: 12657–12662. pmid:21775671
  52. 52. Zhang Y-E, Ma H-J, Feng D-D, Lai X-F, Chen Z-M, Xu M-Y, et al. Induction of Detoxification Enzymes by Quercetin in the Silkworm. J Econ Entomol. 2012;105: 1034–1042. pmid:22812145
  53. 53. Griffiths LA, Smith GE. Metabolism of apigenin and related compounds in the rat. Metabolite formation in vivo and by the intestinal microflora in vitro. Biochem J. 1972;128: 901–911. pmid:4638796
  54. 54. Winter J, Popoff MR, Grimont P, Bokkenheuser VD. Clostridium orbiscindens sp. nov., a human intestinal bacterium capable of cleaving the flavonoid C-ring. Int J Syst Bacteriol. 1991;41: 355–357. pmid:1883711
  55. 55. Zabela V, Sampath C, Oufir M, Moradi-Afrapoli F, Butterweck V, Hamburger M. Pharmacokinetics of dietary kaempferol and its metabolite 4-hydroxyphenylacetic acid in rats. Fitoterapia. 2016;115: 189–197. pmid:27810397
  56. 56. Mao W, Schuler MA, Berenbaum MR. Task-related differential expression of four cytochrome P450 genes in honeybee appendages. Insect Mol Biol. 2015;24: 582–588. pmid:26190094
  57. 57. Hernandez-Leon A, González-Trujano ME, Fernández-Guasti A. The anxiolytic-like effect of rutin in rats involves GABAA receptors in the basolateral amygdala. Behav Pharmacol. 2017;28: 303–312. pmid:28145981
  58. 58. Liu Z, Silva J, Shao AS, Liang J, Wallner M, Shao XM, et al. Flavonoid compounds isolated from Tibetan herbs, binding to GABAA receptor with anxiolytic property. J Ethnopharmacol. 2021;267: 113630. pmid:33246118
  59. 59. Scheepens A, Bisson J-F, Skinner M. p-Coumaric acid activates the GABA-A receptor in vitro and is orally anxiolytic in vivo. Phytother Res PTR. 2014;28: 207–211. pmid:23533066
  60. 60. Decourtye A, Henry M, Desneux N. Overhaul pesticide testing on bees. Nature. 2013;497: 188–188. pmid:23657341
  61. 61. El Hassani AK, Schuster S, Dyck Y, Demares F, Leboulle G, Armengaud C. Identification, localization and function of glutamate-gated chloride channel receptors in the honeybee brain. Eur J Neurosci. 2012;36: 2409–2420. pmid:22632568
  62. 62. Riveros AJ, Gronenberg W. Decision-making and associative color learning in harnessed bumblebees (Bombus impatiens). Anim Cogn. 2012;15. pmid:22837045
  63. 63. Matsumoto Y, Menzel R, Sandoz J-C, Giurfa M. Revisiting olfactory classical conditioning of the proboscis extension response in honey bees: A step toward standardized procedures. J Neurosci Methods. 2012;211: 159–167. pmid:22960052
  64. 64. Pankiw T, Page RE. Response thresholds to sucrose predict foraging division of labor in honeybees. Behav Ecol Sociobiol. 2000;47: 265–267.
  65. 65. Penny KI. Appropriate Critical Values When Testing for a Single Multivariate Outlier by Using the Mahalanobis Distance. J R Stat Soc Ser C. 1996;45: 73–81.
  66. 66. Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J R Stat Soc Ser B Methodol. 1995;57: 289–300.
  67. 67. Pike N. Using false discovery rates for multiple comparisons in ecology and evolution. Methods Ecol Evol. 2011;2: 278–282.