ORIGINAL RESEARCH ARTICLE

Larval exposure to thiamethoxam and American foulbrood: effects on mortality andcognition in the honey bee Apis mellifera

Anna Papacha,b, Dominique Fortinib, Stephane Grateaub, Pierrick Aupinelb and Freddie-Jeanne Richarda*

aLaboratoire Ecologie Evolution Symbiose, UMR CNRS 7267 – EBI team Ecologie, Evolution, Symbiose, Universite de Poitiers, Poitiers Cedex 9,France; bINRA Magneraud, UE Entomologie, Surgeres, France

(Received 3 February 2017; accepted 12 May 2017)

Here, we examined the in vitro effects of co-exposure to a pathogen and a common neonicotinoid on honey bee larvaesurvival and on adult learning behavior following a standard olfactory conditioning procedure based on the proboscisextension response paradigm. We exposed or co-exposed honey bee larvae to American foulbrood and to sub-lethaldoses of thiamethoxam (chronic exposure). Our results revealed no additive effects between the two stressors on lar-val mortality. However, the present work provides the first evidence of impaired learning and memory in adult beesthat were fed thiamethoxam (0.6 ng/bee) during the larval stage. We also show no alterations in learning and memoryin bees after infection with American foulbrood at the larval stage. The present study contributes to our knowledge ofthe sub-lethal effects of neonicotinoids on honey bee larvae and adults.

Exposicion larvaria a thiamethoxam y loque americana: efectos sobre la mortalidad y la cognicion en laabeja de miel Apis mellifera

En este estudio se examinaron los efectos in vitro de la co-exposicion a un patogeno y un neonicotinoide comun en lasupervivencia de las larvas de abejas melıferas y sobre el comportamiento de aprendizaje de adultos siguiendo un pro-cedimiento de acondicionamiento olfativo estandar basado en el paradigma de la respuesta de extension de la probos-cide (PER por sus siglas en ingles). Expusimos o co-expusimos las larvas de abejas melıferas a la loque americana y adosis sub-letales de thiamethoxam (exposicion cronica). Nuestros resultados no revelaron efectos aditivos entre losdos estresores sobre la mortalidad larvaria. Sin embargo, el presente trabajo proporciona la primera evidencia de dete-rioro del aprendizaje y la memoria en abejas adultas que fueron alimentadas con thiamethoxam (0,6 38 ng / abeja) dur-ante la fase larvaria. Tambien mostramos que no hay alteraciones en el aprendizaje y la memoria en las abejas despuesde la infeccion con loque americana en la fase larvaria. El presente estudio contribuye a nuestro conocimiento de losefectos sub-letales de los neonicotinoides en larvas de abejas y en adultos.

Keywords: Apis mellifera; environmental interactions; olfactory learning; neonicotinoid; pathogen; sub-lethal effect

Introduction

Health of the European honey bee (Apis mellifera) is of

high importance due to their key role in pollinating crops

and wild plant communities (Aebi et al., 2012). In recent

years losses of managed honey bee colonies were

observed in Northern hemisphere and many factors

were identified to negatively affect honey bee health and

lead to colony failure (vanEnglesdorp & Meixner, 2010).

Among them are land-use intensification, which includes

habitat fragmentation and the heavy usage of pesticides;

climate change and spread of alien species and diseases

(Potts et al., 2010; Vanbergen & The Insect Pollinators

Initiative, 2013). These factors can interact in many dif-

ferent ways and have various effects, many of which are

unknown. Several studies have been done under labora-

tory conditions and show that pesticides can act syner-

gistically with pathogens on honey bee health and cause

higher mortality (Alaux et al., 2010; Aufauvre et al., 2012;

Di Prisco et al., 2013; Doublet, Labarussias, Miranda,

Moritz, & Paxton,2015; Pettis, vanEnglesdorp, Johnson, &

Dively, 2012; Retschnig, Neumann, & Williams, 2014;

Vidau et al., 2011). For example honey bees that are

exposed to the pesticide clothianidin have reduced

immune defenses and are more susceptible to viruses

(Di Prisco et al., 2013), and exposure to sub-lethal con-

centrations of imidacloprid during the larval stage makes

workers more sensitive to the gut parasite Nosema cer-

anae (Pettis et al., 2012; vanEnglesdorp et al., 2009).

Overall, the studies on co-exposure mentioned above

aimed to detect the effects of co-exposure on worker

mortality and immune defenses. None of them focused

on sub-lethal effects co-exposure might have. It has been

noted that in addition to tests aiming to detect bee mor-

tality, we are in need of tests that can determine possible

sub-lethal effects (Schneider, Tautz, Grunewald, & Fuchs,

2012). One of these methods that can be performed

under laboratory conditions is the olfactory conditioning

of the proboscis extension response (PER). It results in

olfactory learning and yields robust olfactory memory.

Considering sub-lethal effects, in the present work in

*Corresponding author. Email: Freddie.jeanne.richard@univ-poitiers.fr

© 2017 International Bee Research Association

Journal of Apicultural Research, 2017

https://doi.org/10.1080/00218839.2017.1332541

addition to mortality, we aimed to study the effects of

co-exposure on honey bee olfactory learning perfor-

mance. Olfactory learning is one of the key components

of successful foraging. Honey bees learn to associate the

floral odor with a nectar reward and then share this

information with newly recruited foragers (Frisch, 1967).

Learned odors can be still remembered several days and

weeks after they are initially encountered (Menzel,

1999). Evidently, olfactory associative behavior is vital for

colony success and survival. Moreover, we decided to

examine the impact of co-exposure on hypopharyngeal

gland (HPG) development. These glands produce the

royal jelly that is required for feeding larvae (Crailsheim

et al., 1992). They are well developed in young bees that

perform nursing duties in the hive, and they degenerate

with age (Deseyn & Billen, 2005).

As a pathogen we chose American foulbrood as it is

considered to be the most detrimental disease for honey

bee larvae (Genersch, 2010). American foulbrood has

been known for more than a century and is a cosmopoli-

tan disease, occurring all around the world wherever

there are the colonies of A. mellifera (Genersch, 2010).

This disease is highly contagious, and the spores are

highly resistant in the environment and can remain dor-

mant for as long as 50 years (Genersch, 2010). The cau-

sative agent is the gram-positive spore-forming bacterium

Paenibacillus larvae, which can only infect honey bee larvae

within 53 h after hatching, while adult bees are safe from

infection (Genersch, Ashiralieva, & Fries, 2005).

The choice of pesticide for our experiment had fallen

on thiamethoxam. This pesticide is used for crop protec-

tion and belongs to the group of neonicotinoids, which

are currently most well-known and widely used group of

pesticides. It is an active ingredient in the pesticides Helix

XTra and Cruiser. In the environment, bees can be

exposed to thiamethoxam while foraging. Thiamethoxam

use is widespread: it is one of the most abundant neoni-

cotinoids and was found to be present in 65% of nectar

samples and 37% of pollen samples examined (Pohorecka

et al., 2012). In treated crops, the amount of thi-

amethoxam in pollen can reach 2–7 ppb (Pilling, Camp-

bell, Coulson, Ruddle, & Tornier, 2013), while in nectar

this number varies between 3.2 and 12.9 ppb (Pohorecka

et al., 2012), and thiamethoxam has also been detected in

wax inside the hives, where it can reach 53.3 ppb (Mullin

et al., 2010). This contaminated pollen and nectar are

brought back to the hives and are used and processed by

nurse bees to feed larvae.

There is a gap our knowledge gained from co-expo-

sure studies on the interaction between pesticides and

pathogens. Moreover, nearly all of the studies con-

ducted on the impact of pesticides on honey bees have

been performed on workers and only a few have looked

at the impact that it might have after the early exposure

of honey bee larvae (Tan et al., 2015; Tavares, Roat,

Carvalho, Silva-Zacarin, & Malaspina, 2015; Yang, Chang,

Wu, & Chen, 2012). Consequently, in our study, we

decided to investigate the effects on honey bee larvae

of exposure to a pesticide and pathogen (both sepa-

rately and when co-exposed). We exposed honey bee

larvae to sub-lethal doses of a pesticide (thiamethoxam)

and a pathogen (P. larvae) and measured their effects on

honey bee larvae survival and emergence, as well as on

olfactory learning and memory abilities and HPG devel-

opment in adults. We also tested whether co-exposure

to this pesticide and pathogen had an additive effect.

Materials and methods

All the experiments were conducted in the entomologi-

cal experimental unit of INRA, le Magneraud, France

between March and August, 2015. In this study, we

recorded larval and pupal mortality and adult emergence

rates. We also measured effects on learning and mem-

ory, as well as total head protein concentrations, in

honey bee workers in the different treatment groups.

We compared all variables between the untreated larvae

used as the control group (control) and the treated

larvae. Treatments were as follows: exposure to 800 P.

larvae spores (Am F) or 0.6 ng thiamethoxam (TMX), in

cases of single exposure, or exposure to 800 P. larvae

spores and 0.6 ng thiamethoxam (Am F × TMX high) or

to 400 P. larvae spores and 0.3 ng thiamethoxam (Am

F × TMX low), in cases of co-exposure. All groups

tested including the control ones were raised in vitro.

Larval rearing and exposure

One-day-old larvae were collected from 3 different

healthy colonies of A. mellifera ligustica and reared in vitro

until reaching the adult stage. For artificial larval rearing,

we followed the method adapted by Aupinel et al.

(2005). To obtain a comb with first instar larvae, the

queen was confined to an empty comb using an exclu-

der cage and left there for 24 h. After the queen was

removed from the cage, a comb with eggs was left in

the hive for two more days to obtain first instar larvae

for grafting. All grafts were performed in the laboratory

at room temperature. Larvae were transferred from the

comb to artificial cells previously filled with royal jelly

and an aqueous solution of D-glucose, D-fructose and

yeast extract. Plates with day-old larvae were trans-

ferred into an incubator at 34.5 ± 0.5 ˚C and 95 ± 5%

relative humidity (RH) (Figure 1). Two sets of grafts

were performed per each colony. A total of 96 larvae

per treatment were obtained during each graft.

On the 8th day, all larvae were transferred into a

desiccator with a lower humidity (80 ± 5%) and kept in

an incubator at 34.5 ± 0.5 ˚C (Figure 1). At the end of

the pupation stage on day 15, plates with pupae were

transferred into alimentary crystal polypropylene boxes

with aerated lids. The boxes were kept in an incubator

at 34.5 ± 0.5 ˚C and approximately 50% RH. After

emergence, honey bee workers were fed with pollen

powder and a solution of 50% sucrose ad libitum and

were exposed to Bee Boost (Phero Tech Inc) and pieces

2 A. Papach et al.

of wax that were added to each box. Pollen was chan-

ged every two days. Mortality was recorded every day

before feeding time from day 3 to day 5 during the lar-

val stage, on day 8 and day 15 during the pupal stage

and from day 18 to day 22 after adult emergence.

Adults were kept in these conditions for 13–14 days

after emergence until the olfactory learning tests.

American foulbrood exposure

American foulbrood is a disease affecting honey bee lar-

vae that is caused by sporulating P. larvae. Spores repre-

sent the infective form of the disease (Genersch, 2010),

and they germinate in the larval midgut of the host, giv-

ing rise to the vegetative forms that causes bacteremia

and death. The P. larvae strain used here was collected

and isolated from an infected colony at the Magneraud

beekeeping research station in 2010. It was purified and

stored in water at 5 ˚C. The genotype of P.larvae that

was used in the present experiment is ERIC I. Honey

bee larvae were exposed to P. larvae spores on day 1

after grafting. Honey bee larvae were exposed to either

400 or 800 spores in their diet according to the

experimental treatment (single or co-exposure).

Thiamethoxam exposure

Thiamethoxam (99% purity) was obtained from the Clu-

zeau Info Labo. Stock solutions of this pesticide were

prepared in water. Thiamethoxam was added to the lar-

val diet. In total, each larva received 0.6 ng or 0.3 ng of

thiamethoxam according to the experimental treatment

(single or co-exposure), which corresponded to 4 and

2 ppb of thiamethoxam in the diet, respectively. Chronic

exposure of larvae to thiamethoxam was achieved by

treating them during 4 days, starting from day 3 and

ending on day 6.

The sub-lethal concentration of thiamethoxam was

determined based on preliminary experiments. For rear-

ing and maintaining honey bee larvae we used the same

method as described above. During the experiments we

exposed 384 larvae per treatment which included the

following: control (no treatment applied), 0.3 ng of thi-

amethoxam per bee, 0.6 ng/bee and 2.4 ng/bee. The lar-

val/adult mortality was recorded until D22. The data

analysis of mortality showed that only exposition at the

higher concentration (2.4 ng/bee) led to significantly

higher mortality (p < 0.001, Figure 2).

Learning and memory behavioral tests

Learning and memory behavioral tests were performed

with workers that were 13–14 days old. At 13–14 days

old, they show better responses to odor and better

learning abilities than younger bees (Laloi et al., 2000).

Prior to the PER tests, honey bee workers were anes-

thetized by cooling them on ice until they became

motionless (Felsenberg, Gehring, Antemann, &

Eisenhardt, 2011). Bees were individually harnessed with

adhesive tape in metal tubes marked for individual iden-

tification. Every honey bee worker was restrained in a

way that allowed its proboscis to extend and its mouth

parts to move freely but that prevents other move-

ments. After harnessing the honey bee workers, they

were left for 3–4 h to recover in the dark at room

temperature (Giurfa & Sandoz, 2012).

Absolute conditioning procedure

In the present study, we followed the revised classical

olfactory conditioning method of the PER protocol

described by Matsumoto, Menzel, Sandoz, & Giurfa,

2012 with positive reinforcement. In most studies, a

Figure 1. Schematic representation of the main steps of the in vitro larval rearing (modified from Medrzycki et al. 2013) andexposure protocols (D = day, RH = relative humidity).Notes: Different letters for diet indicate different diet compositions. At D1 after grafting, larvae were exposed to Paenibacillus larvaespores. D3 to D6 – chronic exposure to thiamethoxam (test solution), which was added to the diet.

Larval exposure to thiamethoxam and American foulbrood 3

30% sugar solution is used (Matsumoto et al., 2012) but

considering the fact that our bees were reared in vitro

and fed with a 50% sugar solution during their develop-

ment, we used a 50% sugar solution as the uncondi-

tioned stimulus (US). As a conditioning stimulus (CS)

during the conditioning trials, we used 1-nonanol (Sigma

Aldrich) that had been freshly prepared before the con-

ditioning. For odor delivery, we used a 20 ml plastic syr-

inge containing a piece of filter paper (10 × 30 mm)

soaked with 5 μl of the odorant.

At first, honey bee workers were tested for their

PER to sucrose by stimulating their antennae with a cot-

ton stick soaked in the sucrose solution. The individuals

who did not extend their proboscis were discarded

from the test. The conditioning site had a ventilation

hood for odor flow regulation and removal. Condition-

ing was performed using the following steps: a

harnessed honey bee worker was placed in the condi-

tioning site, and after 15 s, we presented the odor for

4 s, with a subsequent 4 s of sucrose stimulation and an

inter-stimulus interval (ISI) of 2 s. When the harnessed

honey bee worker was extending its proboscis, it was

allowed to drink. After the CS-US pairing, the harnessed

honey bee worker was left in the conditioning site for

15 s. The inter trial interval (ITI) was 10 min

(Matsumoto et al., 2012). Each bee received three

conditioning trials, and in total, we trained between 88

and 95 individuals per treatment.

Memory retention

Middle-term memory retention tests were performed

1 h after the last conditioning trial. In addition to the CS

(1-nonanol), we also used a novel odor (2-hexanol)

(Sigma Aldrich) as healthy workers are able to discrimi-

nate between those two odors (Guerrieri, Schubert,

Sandoz, & Giurfa, 2005). To test the memory of har-

nessed honey bee workers, the bee was placed in the

conditioning site, and after 15 s the test odor was pre-

sented for 4 s without sugar exposure. The bee was

exposed to both odors at different times, either the

conditioned odor (1-nonanol) or the novel odor (2-hex-

anol), with the two presentation trials separated by an

interval of 10 min and performed in a random order.

After the middle-term memory retention tests, all bees

were checked for their PER to sucrose. Individuals who

did not respond were discarded from the test. For all

treatments, we tested between 50 and 55 bees for mid-

dle-term memory.

To test long-term memory, bees were trained with

five conditioning trials (Menzel, 1999; Matsumoto et al.,

2012). After the last conditioning trial and 6 h before

the memory test, trained bees were fed with a sucrose

solution to avoid starvation and death. Bees were kept

in a dark place at room temperature overnight, and the

memory retention test was performed 24 h later. Each

bee was placed in the conditioning site, and after 15 s

the test odor was presented for 4 s without sugar

exposure. The bee was exposed to both odors, with an

interval of 10 min between trials as described for the

middle-term memory test. After the long-term

memory retention tests all bees were checked for their

PER to sucrose. In total, we tested between 38 and 42

individuals per treatment.

HPG development

One of the most common methods used to assess the

development of the HPG is to measure the protein con-

tent of the glands and to measure the diameter of the

HPG acini (Deseyn & Billen, 2005; DeGrandi-Hoffman,

Chen, Huang, & Huang, 2010; Gupta & Chandel, 1995).

Figure 2. Change in mortality after the chronic application of three concentrations of thiamethoxam from day 4 to day 22.Note: Each mortality was compared with the control sample at day 22 using a χ2 test with 1 df (***, p < 0.001).

4 A. Papach et al.

In the present work, we used an indirect method of

testing HPG development that involved measuring the

total protein content in the head of the honey bee, simi-

lar to method used by Renzi et al. (2016).

To assess HPG activity, we used bees that were

13–14 days after emergence as adults that were killed

by freezing at −20 ˚C and stored until analysis. Total

head protein concentration was measured using the

entire head of the honey bee workers in the Bradford

protein assay, which is based on the reaction of the

proteins with a colorant. Heads were placed individually

in a tube with 300 μl of KH2PO4. To preserve the pro-

teins, tubes were placed on ice. Each head was pre-

ground for five seconds with the help of pellet pestle

and then further ground for one minute. After grinding,

500 μl of KH2PO4 (50 mM, pH 7) was added, and the

tubes were centrifuged for four minutes at 10,000

revolutions/minute at a temperature of 5 ˚C. From each

tube, we extracted 500 μl of supernatant. To measure

the protein content, 20 μl of supernatant was mixed

with 1 ml of the Bradford reactant. Proteins were mea-

sured with a spectrophotometer at 595 nm. In total, we

tested between 88 and 90 honey bees per treatment.

Data analysis

Cumulative larval mortality and adult emergence rates

were compared between the control group and each

treatment using multiple two-by-two χ2 tests with 1 df

and a critical probability level of 0.012 based on a Dunn

Sidak correction of the standard probability level. To test

for a potential additive effect of co-exposure on mortality

on day 22, we used the formula proposed by Aufauvre

et al. (2012): χ2 = (Mo−ME)2/ME, where Mo is the

observed mortality in the group that received both the

pesticide and pathogen and ME is the expected mortality

calculated using the following formula: ME = MAF + MT

(1 − MAF/100), where MAF and MT are the observed per-

cent mortalities caused by P. larvae and thiamethoxam

alone. The results of the calculated χ2 were compared

with the χ2 table values with 1 df. The interactions were

considered synergistic when the calculated χ2 value

exceeded the table value and the difference between Mo

and ME had a positive value.

The results of PER conditioning were analyzed with

a one-way ANOVA followed by Fisher’s least significant

difference (LSD) test (p < 0.05). All analyses were

performed using R (R Core Team 2013).

Results

Development stages and mortality

The mortality at the six time intervals during the test

are presented in Figure 3. During the larval stage (days

3–8), cumulative mortality in the control group was

approximately 6% (Figure 3), which is below the stan-

dard acceptance threshold (≤15%) for in vitro rearing

conditions (Crailsheim et al., 2013).

Cumulative mortality at the end of the larval stage

(D8) was increased in all groups fed P. larvae spores com-

pared to that of the control group (Am F: χ2 = 27.2, df:1,

p < 0.01; Am F × TMX high: χ2 = 30.4, df:1, p < 0.01; Am

F × TMX low: χ2 = 17.8, df:1, p < 0.01).

The adult emergence rate in the control group was

82% (Figure 4), which is above the acceptance threshold

(≥70%) for in vitro rearing conditions (Crailsheim et al.,

2013). Feeding honey bee larvae only thiamethoxam

(0.6 ng) had no significant effect on worker emergence

rate when compared to that of the control group

(χ2 = 1.75, df:1, p > 0.01; Figure 4). However, we

observed the lowest adult emergence rate (58%) when

the larvae were co-exposed to Am F × TMX high

(χ2 = 77.2, df:1, p < 0.01). The exposure of honey bee

larvae to Am F and the co-exposure with Am F × TMX

low also resulted in significantly lower adult emergence

rates (χ2 = 48.8, df:1, p < 0.01 and χ2 = 37.3, df:1,

p < 0.01, respectively).

No additive effect of thiamethoxam was detected on

the larval mortality caused by American foulbrood

(Mo = 41.8%, ME = 45.9%, χ2 = 1.82).

Learning and memory behavioral performance

To test the impact of thiamethoxam and/or American

foulbrood on honey bee learning abilities, restrained

workers were subjected to a PER assay. Figure 5 shows

the percentage of bees responding to the CS during three

learning trials in the control group and in the treated

groups (N = 95 control, N = 88 Am F, N = 90 TMX,

N = 95 Am F × TMX high and N = 94 Am F × TMX low).

At the end of the conditioning (trial 3), the highest PER

rate to the CS was recorded in the control group (61%),

while the lowest PER rate was observed in the group that

was treated with the high rate of TMX (42%). Overall,

feeding larvae thiamethoxam and/or P. larvae spores did

not result in a statistically significant change in the propor-

tion of non-responding adult bees (F4,457 = 2.12,

p = 0.07). Nonetheless, a pairwise comparison revealed

reduced PER rates to the CS in the groups exposed to

TMX (Fisher LSD: p = 0.01) and to Am F × TMX low

(Fisher LSD: p = 0.03).

Middle-term memory

In memory tests 1 h after the last conditioning trial,

bees in all groups showed more responses to the CS

(1-nonanol) than to the novel odor (2-hexanol) (McNe-

mar test: control: χ2 = 8.05, p < 0.005; Am F: χ2 = 8.33,

p < 0.005; TMX: χ2 = 28.12, p < 0.001 Am F × TMX

high: χ2 = 13.33, p < 0.001; Am F × TMX low: χ2 = 8.33,

p < 0.005). This confirms that the honey bee workers

had a CS-specific memory and the ability to distinguish

between odors was not affected by exposure to any of

the treatments.

CS-specific memory was compared among the

groups by computing the percentage of individuals

Larval exposure to thiamethoxam and American foulbrood 5

responding to the CS (Figure 6(a)). Feeding larvae the

pesticide and/or pathogen did not affect the percentage

of bees responding to the CS (F4,257 = 2.23, p = 0.06).

However, a pairwise comparison showed that feeding

larvae thiamethoxam led to the impairment of middle-

term memory (Fisher LSD: p = 0.004).

Long-term memory

The effects of the treatments on long-term memory

performance were tested 24 h after the conditioning

(Figure 6(b)). The bees used for these tests received

two extra trials, and in total, we used 200 bees for the

long-term memory tests (control N = 42; Am F N = 35;

Figure 3. Change in mortality after the application of the treatments from day 3 to day 22.Notes: Am F: larvae exposed to Paenibacillus larvae spores, causing American foulbrood disease. TMX: larvae exposed tothiamethoxam. Am F × TMX high: larvae co-exposed to 800 Paenibacillus larvae spores and 0.6 ng thiamethoxam. Am F × TMX low:larvae co-exposed to 400 Paenibacillus larvae spores and 0.3 ng thiamethoxam.

Figure 4. Effect of the treatments applied during the larval stage on adult emergence rate. Each emergence rate was comparedwith that of the control group using a χ2 test with 1 df (*p < 0.01).Notes: Am F: larvae exposed to Paenibacillus larvae spores, causing American foulbrood disease. TMX: larvae exposed tothiamethoxam. Am F × TMX high: larvae co-exposed to 800 Paenibacillus larvae spores and 0.6 ng thiamethoxam. Am F × TMX low:larvae co-exposed to 400 Paenibacillus larvae spores and 0.3 ng thiamethoxam.

6 A. Papach et al.

TMX N = 38; Am F × TMX high N = 41; Am F × TMX

low N = 44). The PER rate to the CS stayed at the same

level during trials four and five in all groups and was the

same as that at the end of the third trial (control:

χ2 = 0.05, p > 0.05; Am F: χ2 = 0.06, p > 0.05; TMX:

χ2 = 0, p > 0.05; Am F × TMX high: χ2 = 0.05, p > 0.05;

Am F × TMX low: χ2 = 0.05, p > 0.05).

Exposing honey bee larvae to thiamethoxam and/or

American foulbrood resulted in different CS response

rates between the groups (F4,195 = 2.42, p = 0.04). Bees

from the control group showed better performance

compare to bees that were fed TMX (Fisher LSD:

p = 0.02), Am F × TMX high (Fisher LSD: p = 0.02), and

Am F × TMX low (Fisher LSD: p = 0.003) during the

larval stage. In bees that were fed Am F during the larval

stage, we noticed a tendency for impaired long-term

memory (Fisher LSD: p = 0.06).

HPGs development

To evaluate HPG development in adult bees, we com-

pared the total head protein concentrations in adult’s

that received different treatments at the larval stage.

Comparisons between our control group and the

groups exposed the either thiamethoxam or P. larvae

spores or co-exposed to both show that there was no

significant impact of exposure at the larval stage on

adult bees HPGs (F4,439 = 2.31, p = 0.056, Figure 7).

Discussion

Honey bee survival at the different stages

All honey bee workers were reared in vitro since the

larval stage and were either not exposed (control) or

were exposed to a pesticide (the neonicotinoid thi-

amethoxam), to a pathogen (spores of P. larvae, the cau-

sative agent of American foulbrood) or to both at two

different concentrations. The observed larval mortality

and adult emergence rates in the control group were

comparable to the mortality and emergence rates

observed when larvae were exposed to thiamethoxam

at the highest concentration (0.6 ng/larvae), which cor-

responded to a sub-lethal dose in our experiments. To

put this value into perspective, we compared it with the

sub-lethal dose of thiamethoxam for adult honey bees

(1.34 ng/bee) (European Food and Safety Authority,

2013). The doses used in our experiment are lower

than that value; however, it has been suggested that

honey bee larvae are more tolerant to some of neoni-

cotinoids than adults (Tavares et al., 2015; Yang et al.,

2012). Interestingly, the sub-lethal effects on the devel-

opment of Africanized honey bee larvae of chronic

exposure to thiamethoxam has been demonstrated at

higher doses (56.4 ng/larva) than those used in our

study for the sub-species Apis mellifera ligustica (Tavares

et al., 2015). Then again, it seems that larvae of the stin-

gless bee (Scaptotrigona aff. depilis) are much more sensi-

tive to thiamethoxam than the honey bee (Apis mellifera

Figure 5. Learning curves showing the proportion of restrained honey bee workers responding with a proboscis extension to theconditioned odor during 3 conditioning trials. Number of responding workers in each treatment was compared with the number ofresponding workers in the control group. Fisher LSD (*p < 0.05).Notes: Am F: larvae exposed to Paenibacillus larvae spores, causing American foulbrood disease. TMX: larvae exposed tothiamethoxam. Am F × TMX high: larvae co-exposed to 800 Paenibacillus larvae spores and 0.6 ng thiamethoxam. Am F × TMX low:larvae co-exposed to 400 Paenibacillus larvae spores and 0.3 ng thiamethoxam.

Larval exposure to thiamethoxam and American foulbrood 7

ligustica). In this species, thiamethoxam induces

increased mortality and shorter development times at a

concentration 0.044 ng/larva (Rosa et al., 2016). Notice-

ably, the observed and reported values are highly vari-

able, suggesting a considerable variation in tolerance

among species and sub-species. Moreover, several stud-

ies have concluded that susceptibility to neonicotinoids

depends on genetics, the specific life cycle stage, and dif-

ferences in nesting activity and foraging behavior (God-

fray et al., 2014; Laurino, Porporato, & Patetta, 2011;

Sandrock et al., 2014; vanEnglesdorp et al., 2009), fur-

ther suggesting that it can vary considerably. For that

reason, the permissible doses of pesticides should be

reconsidered after taking into account the substantial

variation that exists in susceptibility among species.

We also exposed honey bee larvae to P. larvae

spores alone and at two different concentrations along

with the pesticide. As variations in virulence are geno-

type dependent (Genersch et al., 2005), we calculated

the larval mortality caused by the strain used in our

study. A mortality rate of up to 42% (24% more than in

the control group) occurred primarily within 5–15 days

post-infection. The mortality rate was not influenced by

either the quantity of the spores alone or by the addi-

tional exposure to the pesticide, even when the quantity

of spores was reduced by half. The other strain, JT-79,

has been shown to cause mortality rates ranging from

25 to 55%, depending on spore dose (ranging from 3 to

24) (Brødsgaard, Ritter, Hansen, & Brødsgaard, 2000).

The time course of infection in our study was not

(a)

(b)

Figure 6. Memory retrieval of the conditioned proboscis response. Bars represent the proportion of restrained honey beeworkers that showed the conditioned response to the presented odor 1 h after the last conditioning trial (a) and 24 h after the lastconditioning trial (b). Fisher LSD test (*p < 0.05 and **p < 0.01).Notes: Am F: larvae exposed to Paenibacillus larvae spores, causing American foulbrood disease. TMX: larvae exposed tothiamethoxam. Am F × TMX high: larvae co-exposed to 800 Paenibacillus larvae spores and 0.6 ng thiamethoxam. Am F × TMX low:larvae co-exposed to 400 Paenibacillus larvae spores and 0.3 ng thiamethoxam.

8 A. Papach et al.

influenced by spore concentration, as has been shown

in a previous study (Genersch et al., 2005).

No additive effect of thiamethoxam was found on

the mortality caused by American foulbrood. A study of

the interactions between American foulbrood, acute

paralysis virus (APV) and Varroa jacobsoni in various

inoculation combinations at the larval stage revealed no

additive mortality caused by P. larvae spores (Brødsgaard

et al., 2000). Among the other studies testing the inter-

actions between pesticides and pathogens on honey bee

species, the majority have revealed additive effects on

mortality (Alaux et al., 2010; Di Prisco et al., 2013;

Doublet et al., 2015; Pettis et al., 2012; Retschnig et al.,

2014; Vidau et al., 2011) however, this is not always the

case. It has been noted that there is lack of studies

assessing the importance of the sequence of exposure

to the different factors (Holmstrup et al., 2010), which

was later demonstrated in a co-exposure study on the

interaction between one pathogen and one insecticide

on the mortality of honey bee adults by Aufauvre et al.

(2012). In their study, they paid attention to the

sequence of exposure and noted that the most signifi-

cant interaction between the factors was detected when

they were applied simultaneously.

Learning and memory behavioral performance

Successful learning and memory ensures successful for-

aging behavior, which is important for colony survival.

Impaired learning negatively affects individual bee forag-

ing, and it may jeopardize colony survival (Bryden, Gill,

Mitton, Raine, & Jansen, 2013; Gill & Raine, 2014; San-

drock et al., 2014). In the present work, we provide the

first evidence of impaired learning and memory in adult

bees that were exposed to thiamethoxam during the

larval stage. Chronic larval exposure to sub-lethal doses

of this neonicotinoid resulted in alterations of associa-

tive behavior in adults. Impaired learning in adults has

also been revealed when honey bee larvae are chroni-

cally exposed to sub-lethal doses of imidacloprid (Yang

et al., 2012).

Impaired learning and memory in honey bee adults

exposed to sub-lethal doses of neonicotinoids has been

shown in many previous studies (Aliouane et al., 2009;

Blacquiere, Smagghe, van Gestel, & Mommaerts, 2012;

Decourtye et al., 2004; Potts et al., 2010; Tan et al.,

2015; Vanbergen & The Insect Pollinators Initiative,

2013). For example, impaired long-term memory in

honey bee workers due to chronic thiamethoxam

exposure was observed at a concentration of 0.1 ng/bee

(Aliouane et al., 2009).

Pathogens/parasites are known to alter a host’s

behavior in many different ways (Combes, 2001), while

minimal attention has been devoted to the indirect

effects of pathogens on learning and memory. For exam-

ple, learning and memory can be altered by endosym-

bionts such as the bacteria Wolbachia in terrestrial

isopods (Temple & Richard, 2015), and the parasitic

mite Varroa destructor can impair non-associative learning

in honey bee adults (Kralj, Brockmann, Fuchs, & Tautz,

2007). In the present study, we tested both learning and

memory in individuals treated with P. larvae spores.

P. larvae spores are known to be highly virulent at the

larval stage, killing the host within very short time (Gen-

ersch et al., 2005). In our study, 64% of the treated lar-

vae survived P. larvae spore treatment to potentially

show physiological alterations or were asymptomatic

and had the potential to become contagious individuals.

Figure 7. Protein concentration in the heads of honey bee workers at 13–14 days old (mean ± SEM).Notes: Number of replicates is between 87 and 90 individuals per treatment. Am F: larvae exposed to Paenibacillus larvae spores,causing American foulbrood disease. TMX: larvae exposed to thiamethoxam. Am F × TMX high: larvae co-exposed to 800 Paenibacil-lus larvae spores and 0.6 ng thiamethoxam. Am F × TMX low: larvae co-exposed to 400 Paenibacillus larvae spores and 0.3 ngthiamethoxam.

Larval exposure to thiamethoxam and American foulbrood 9

When honey bee larvae are infected, P. larvae germinate

in the midgut, causing possible side effects. For example,

it has been shown that gut bacteria (Lactobacillus strain)

can induce alterations in the brains of mice (Bravo

et al., 2011). However, in our experimental conditions,

individuals exposed to P. larvae spores revealed no

impairments in learning or memory compared to

untreated individuals.

A reduced positive PER rate was observed during

learning performance and long-term memory tests in the

group exposed to both thiamethoxam and P. larvae

spores at lower concentrations. Surprisingly, we did not

observe alterations in learning behavior in the group co-

exposed to the higher doses. A possible explanation for

this is that selection occurred during the development

stages. The honey bee larvae that were treated with both

0.6 ng of thiamethoxam and 800 P. larvae spores had the

highest mortality rate (42%). Therefore, in this treatment,

only the strongest individuals survived, and this selection

could mask the negative impact of thiamethoxam.

HPG activity

In the present study, we also looked at the activity of

the HGP glands. These glands secrete the “brood food”

which is rich in proteins and is used to feed larvae of all

castes (Sagili & Pankiw, 2007). HPG activity is age

dependent: HPGs are well developed in nursing bees

that feed the brood (Crailsheim et al., 1992), but with

age, they stop producing proteins. In bees reared in

cages, HPGs are not very well developed relative to

those of bees from colonies (Crailsheim et al., 1992). In

nurse bees in a colony the maximum development and

productivity of HPGs are on days 8–12 after emergence,

and they subsequently start to decrease in size (Deseyn

& Billen, 2005). In our experiment, we measured the

total concentration of proteins in the workers’ heads as

a measure of HPG activity at 13–14 days after emer-

gence (Renzi et al., 2016). HPG activity was not altered

by exposure to 4 ppb of thiamethoxam, by infection

with P. larvae or by co-exposure with both of these

stressors. In contrast, previous studies have shown that

exposing honey bees to neonicotinoids negatively affects

HPG productivity and size. For example, chronic expo-

sure of honey bees to sub-lethal concentrations of thi-

amethoxam (10 and 40 ppb) has been shown to result

in decreased amounts of total head protein and acini

size (Renzi et al., 2016). The acute or chronic exposure

of honey bees to sub-lethal doses of imidacloprid causes

a reduction acini size (Hatjina et al., 2013; Heylen,

Gobin, Arckens, Huybrechts, & Billen, 2011).

General conclusion

Our results suggest that there are no interactions

between the common neonicotinoid thiamethoxam and

the widespread disease American foulbrood in exposure

sequence tested here. Still, the present work provides

the first evidence of impaired learning and memory in

adult bees that were fed thiamethoxam (0.6 ng/bee) dur-

ing the larval stage. Our work did not reveal any signifi-

cant changes in HPG activity in bees that were infected

with American foulbrood or exposed to thiamethoxam

(0.6 ng/bee) during the larval stage.

Authors contributions

SG performed beekeeping activities and provided biolog-

ical material. AP performed the larval rearing and expo-

sure protocols and the learning and memory tests. AP

and DF performed the HPG analysis. AP, PA and FJR

analyzed the data. FJR, PA, DF and SG designed the

experiments and supervised the work in the respective

laboratories. AP and FJR wrote the manuscript. All

authors read and approved the final manuscript.

Acknowledgements

We would like to thank Carole Moreau-Vauzelle for Paenibacil-lus larvae maintenance. The authors would also like to thankAmerican Journal Experts for their assessment of Englishquality.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This project was financially supported by the European Com-mission through the program Erasmus Mundus Master Course– International Master in Applied Ecology (EMMC-IMAE) [grantnumber FPA 532524-1-FR-2012-ERA MUNDUS-EMMC].

ORCID

Freddie-Jeanne Richard http://orcid.org/0000-0002-2796-1181

References

Aebi, A., Vaissiere, B.E., vanEnglesdorp, D., Delaplane, K., Rou-bik, D.W., & Neumann, P. (2012). Back to the future: Apisversus non-Apis pollination—a response to Ollerton et al.Trends in Ecology & Evolution, 27, 142–143.

Alaux, C., Brunet, J.L., Dussaubat, C., Mondet, F., Tchamitchan,S., Cousin, M., … Le Conte, Y. (2010). Interactions betweenNosema microspores and a neonicotinoid weaken honeybees (Apis mellifera). Environmental Microbiology, 12, 774–782.

Aliouane, Y., El Hassani, A.K., Gary, V., Armengaud, C., Lambin,M., & Gauthier, M. (2009). Subchronic exposure of honeybees to sublethal doses of pesticides: Effects on behavior.Environmental Toxicology and Chemistry, 28, 113–122.

Aufauvre, J., Biron, D.G., Vidau, C., Fontbonne, R., Roudel, M.,Diogon, M., … Blot, N. (2012). Parasite-insecticide interac-tions: A case study of Nosema ceranae and fipronil synergyon honey bee. Scientific Reports, 2(326), 1–7.

Aupinel, P., Fortini, D., Dufour, H., Tasei, J.N., Michaud, B.,Odoux, J.-F., & Pham-Delegue, M.H. (2005). Improvementof artificial feeding in a standard in vitro method for rearingApis mellifera larvae. Bulletin of Insectology, 58, 107–111.

Blacquiere, T., Smagghe, G., van Gestel, C.A.M., & Mommaerts,V. (2012). Neonicotinoids in bees: A review on concentra-

10 A. Papach et al.

tions, side-effects and risk assessment. Ecotoxicology, 21,973–992.

Bravo, J.A., Forsythe, P., Chew, M.V., Escaravage, E., Savignac,H.M., Dinan, T.G., … Cryan, J.F. (2011). Ingestion of Lacto-bacillus strain regulates emotional behavior and centralGABA receptor expression in a mouse via the vagusnerve. Proceedings of the National Academy of Sciences, 108(38), 16050–16055.

Brødsgaard, C.J., Ritter, W., Hansen, H., & Brødsgaard, H.F.(2000). Interactions among Varroa jacobsoni mites, acuteparalysis virus, and Paenibacillus larvae larvae and their influ-ence on mortality of larval honey bees in vitro. Apidologie,31, 543–554.

Bryden, J., Gill, R.J., Mitton, R.A.A., Raine, N.E., & Jansen,V.A.A. (2013). Chronic sublethal stress causes bee colonyfailure. Ecology Letters, 16, 1463–1469.

Combes, C. (2001). Parasitism : The ecology and evolution of inti-mate interactions. Chicago, IL: University of Chicago Press.

Crailsheim, K., Brodschneider, R., Aupinel, P., Behrens, D.,Genersch, E., Vollmann, J., & Riessberger-Galle, U. (2013).Standard methods for artificial rearing of Apis mellifera lar-vae. In V. Dietemann, J.D. Ellis, & P. Neumann (Eds.), TheCOLOSS BEEBOOK, Volume I: standard methods for Apismellifera research. Journal of Apicultural Research, 52(1).doi:10.3896/IBRA.1.52.1.05

Crailsheim, K., Schneider, L.H.W., Hrassnigg, N., Buhlmann,G., Brosch, U., Gmeinbauer, R., & Schoffmann, B. (1992).Pollen consumption and utilization in worker honey bees(Apis mellifera carnica): Dependence on individual age andfunction. Journal of Insect Physiology, 38, 409–419.

Decourtye, A., Armengaud, C., Renou, M., Devillers, J.,Cluzeau, S., Gauthier, M., & Pham-Delegue, M.H. (2004).Imidacloprid impairs memory and brain metabolism in thehoney bee (Apis mellifera L.). Pesticide Biochemistry andPhysiology, 78, 83–92.

DeGrandi-Hoffman, G., Chen, Y., Huang, E., & Huang, M.H.(2010). The effect of diet on protein concentration,hypopharyngeal gland development and virus load inworker honey bees (Apis mellifera L.). Jouranl of Insect Physi-ology, 56, 1184–1191.

Deseyn, J., & Billen, J. (2005). Age-dependent morphology andultrastructure of the hypopharyngeal gland of Apis melliferaworkers (Hymenoptera, Apidae). Apidologie, 36, 49–57.

Di Prisco, G., Cavaliere, V., Annoscia, D., Varricchio, P.,Caprio, E., Nazzi, F., Gargiulo, G., & Pennacchio, F. (2013).Neonicotinoid clothianidin adversely affects insect immu-nity and promotes replication of a viral pathogen in honeybees. Proceedings of the National Academy of Sciences, 110,18466–18471.

Doublet, V., Labarussias, M., Miranda, J.R., Moritz, R.F.A., &Paxton, R.J. (2015). Bees under stress: Sublethal doses of aneonicotinoid pesticide and pathogens interact to elevatehoney bee mortality across the life cycle. EnvironmentalMicrobiology, 17, 969–983.

Felsenberg, J., Gehring, K.B., Antemann, V., & Eisenhardt, D.(2011). Behavioural pharmacology in classical conditioningof the proboscis extension response in honey bees (Apismellifera). Journal of Visualized Experiments, (47), e2282.doi:10.3791/2282

European Food and Safety Authority. (2013). Conclusion on thepeer review of the pesticide risk assessment for bees for theactive substance thiamethoxam. EFSA Journal, 11, 68.

Frisch, K. (1967). The dance language and orientation of bees.Cambridge, MA: Harvard University Press.

Genersch, E. (2010). American foulbrood in honey bees andits causative agent, Paenibacillus larvae. Journal of InvertebratePathology, 103, S10–S19.

Genersch, E., Ashiralieva, A., & Fries, I. (2005). Strain- andgenotype-specific differences in virulence of Paenibacillus

larvae subsp. larvae, a bacterial pathogen causing americanfoulbrood disease in honey bees. Applied and EnvironmentalMicrobiology, 71, 7551–7555.

Gill, R.J., & Raine, N.E. (2014). Chronic impairment of bumblebee natural foraging behaviour induced by sublethal pesti-cide exposure. Functional Ecology, 28, 1459–1471.

Giurfa, M., & Sandoz, J.-C. (2012). Invertebrate learning andmemory: Fifty years of olfactory conditioning of theproboscis extension response in honey bees. Learning &Memory, 19, 54–66.

Godfray, H.C.J., Blacquiere, T., Field, L.M., Hails, R.S., Petrokof-sky, G., Potts, S.G., … McLean, A.R. (2014). A restatementof the natural science evidence base concerning neonicoti-noid insecticides and insect pollinators. Proceedings.Biological sciences/The Royal Society, 281, 20140558.

Guerrieri, F., Schubert, M., Sandoz, J.C., & Giurfa, M. (2005).Perceptual and neural olfactory similarity in honey bees.PLoS Biology, 3, 0718–0732.

Gupta, P.R., & Chandel, R.S. (1995). Effects of diflubenzuronand penfluron on workers of Apis cerana indica F and Apismellifera L. Apidologie, 26, 3–10.

Hatjina, F., Papaefthimiou, C., Charistos, L., Dogaroglu, T.,Bouga, M., Emmanouil, C., & Arnold, G. (2013). Sublethaldoses of imidacloprid decreased size of hypopharyngealglands and respiratory rhythm of honey bees in vivo.Apidologie, 44, 467–480.

Heylen, K., Gobin, B., Arckens, L., Huybrechts, R., & Billen, J.(2011). The effects of four crop protection products onthe morphology and ultrastructure of the hypopharyngealgland of the European honey bee, Apis mellifera. Apidologie,42, 103–116.

Holmstrup, M., Bindesbøl, A.M., Oostingh, G.J., Duschl, A.,Scheil, V., Kohler, H.R., … Spurgeon, D.J. (2010). Interac-tions between effects of environmental chemicals and nat-ural stressors: A review. Science of The Total Environment,408, 3746–3762.

Kralj, J., Brockmann, A., Fuchs, S., & Tautz, J. (2007). The para-sitic mite Varroa destructor affects non-associative learningin honey bee foragers, Apis mellifera L. Journal of Compara-tive Physiology A, 193, 363–370.

Laloi, D., Bailez, O., Blight, M. M., Roger, B., Pham-Delegue,M.H., & Wadhams, L.J. (2000). Recognition of complexodors by restrained and free-flying honey bees, Apis mellif-era. Journal of Chemical Ecology, 26, 2307–2319.

Laurino, D., Porporato, M., & Patetta, A. (2011). Toxicity ofneonicotinoid insecticides to honey bees: Laboratory tests.Bulletin of Insectology, 64, 107–113.

Matsumoto, Y., Menzel, R., Sandoz, J.C., & Giurfa, M. (2012).Revisiting olfactory classical conditioning of the proboscisextension response in honey bees: A step toward stan-dardized procedures. Journal of Neuroscience Methods, 211(1), 159–167.

Medrzycki, P., Giffard, H., Aupinel, P., Belzunces, L.P., Chauzat,M.-P., Claßen, C, … Vidau, C. (2013). Standard methodsfor toxicology research in Apis mellifera. In V. Dietemann,J.D. Ellis & P. Neumann (Eds.), The COLOSS BEEBOOK,Volume I: standard methods for Apis mellifera research.Journal of Apicultural Research (Vol. 52). Retrieved fromhttp://dx.doi.org/10.3896/IBRA.1.52.4.14

Menzel, R. (1999). Memory dynamics in the honey bee. Journalof Comparative Physiology A: Sensory, Neural, and BehavioralPhysiology, 185, 323–340.

Mullin, C.A., Frazier, M., Frazier, J.L., Ashcraft, S., Simonds, R.,vanEnglesdorp, D., & Pettis, J.S. (2010). High levels of miti-cides and agrochemicals in North American apiaries: Impli-cations for honey bee health. PLoS ONE, 5(3), 1–19.

Pettis, J.S., vanEnglesdorp, D., Johnson, J., & Dively, G. (2012). Pes-ticide exposure in honey bees results in increased levels of thegut pathogenNosema.Naturwissenschaften, 99, 153–158.

Larval exposure to thiamethoxam and American foulbrood 11

Pilling, E., Campbell, P., Coulson, M., Ruddle, N., & Tornier, I.(2013). A four-year field program investigating long-termeffects of repeated exposure of honey bee colonies to flower-ing crops treated with thiamethoxam. PLoS ONE, 8(10), 1–14.

Pohorecka, K., Skubida, P., Miszczak, A., Semkiw, P., Sikorski,P., Zagibajo, K., … Zdanska, D. (2012). Residues of neoni-cotinoid insecticides in bee collected plant materials fromoilseed rape crops and their effect on bee colonies. Journalof Apicultural Science, 56, 115–134.

Potts, S.G., Biesmeijer, J.C., Kremen, C., Neumann, P.,Schweiger, O., & Kunin, W.E. (2010). Global pollinatordeclines: Trends, impacts and drivers. Trends in Ecology &Evolution, 25, 345–353.

Renzi, M.T., Rodrıguez-Gasol, N., Medrzycki, P., Porrini, C.,Martini, A., Burgio, G., Maini, S., & Sgolastra, F. (2016).Combined effect of pollen quality and thiamethoxam onhypopharyngeal gland development and protein content inApis mellifera. Apidologie, 47, 779–788.

Retschnig, G., Neumann, P., & Williams, G.R. (2014). Thiacloprid–Nosema ceranae interactions in honey bees: Host survivorshipbut not parasite reproduction is dependent on pesticide dose.Journal of Invertebrate Pathology, 118, 18–19.

Rosa, A., Teixeira, J.S.G., Vollet-Neto, A., Queiroz, E.P., Blochtein,B., Pires, C.S.S., & Imperatriz-Fonseca, V.L. (2016). Consump-tion of the neonicotinoid thiamethoxam during the larvalstage affects the survival and development of the stinglessbee, Scaptotrigona aff. depilis. Apidologie, 47, 729–738.

Sagili, R.R., & Pankiw, T. (2007). Effects of protein-constrainedbrood food on honey bee (Apis mellifera L.) pollen foragingand colony growth. Behavioral Ecology and Sociobiology, 61,1471–1478.

Sandrock, S., Tanadini, M., Tanadini, L.G., Fauser-Misslin, A.,Potts, S.G., & Neumann, P. (2014). Impact of chronicneonicotinoid exposure on honey bee colony performanceand queen supersedure. PLoS ONE, 9(8), 1–13.

Schneider, C.W., Tautz, J., Grunewald, B., & Fuchs, S. (2012).RFID tracking of sublethal effects of two neonicotinoidinsecticides on the foraging behavior of Apis mellifera. PLoSONE, 7(1), 1–9.

Tan, K., Chen, W., Dong, S., Liu, X., Wang, Y., & Nieh, J.C.(2015). A neonicotinoid impairs olfactory learning in Asianhoney bees (Apis cerana) exposed as larvae or as adults.Scientific Reports, 5, 1–8.

Tavares, D. A., Roat, T.C., Carvalho, S.M., Silva-Zacarin,E.C.M., & Malaspina, O. (2015). In vitro effects ofthiamethoxam on larvae of Africanized honey bee Apis mel-lifera (Hymenoptera: Apidae). Chemosphere, 135, 370–378.

Temple, N., & Richard, F.-J. (2015). Intra-cellular bacterialinfections affect learning and memory capacities of aninvertebrate. Frontiers in Zoology, 12(1), 145.

Vanbergen, A., & The Insect Pollinators Initiative. (2013).Threats to an ecosystem service: Pressures on pollinators.Frontiers in Ecology and the Environment, 11, 251–259.

vanEnglesdorp, D., Evans, J., Saegerman, C., Mullin, C.,Haubruge, E., Nguyen, B., … Pettis, J. (2009). Colony col-lapse disorder: A descriptive study. PLoS ONE, 4(8), 1–17.

vanEnglesdorp, D., & Meixner, M.D. (2010). A historical reviewof managed honey bee populations in Europe and the Uni-ted States and the factors that may affect them. Journal ofInvertebrate Pathology, 103, S80–S95.

Vidau, C., Diogon, M., Aufauvre, J., Fontbonne, R., Vigues, B.,Brunet, J.L., … Delbac, F. (2011). Exposure to sublethaldoses of fipronil and thiacloprid highly increases mortalityof honey bees previously infected by Nosema ceranae. PLoSONE, 6(6), 1–8.

Yang, E.C., Chang, H.C., Wu, W.Y., & Chen, Y.W. (2012).Impaired olfactory associative behavior of honey beeworkers due to contamination of imidacloprid in the larvalstage. PLoS ONE, 7(11), 1–8.

12 A. Papach et al.