Parasitol Res (2014) 113:4601–4610 DOI 10.1007/s00436-014-4150-z

ORIGINAL PAPER

Protecting honey bees: identification of a new varroacide by in silico, in vitro, and in vivo studies Fabienne Dulin & Céline Zatylny-Gaudin & Céline Ballandonne & Bertrand Guillet & Romain Bonafos & Ronan Bureau & Marie Pierre Halm

Received: 10 April 2014 / Accepted: 23 September 2014 / Published online: 31 October 2014 # Springer-Verlag Berlin Heidelberg 2014

Abstract Varroa destructor is the main concern related to the gradual decline of honeybees. Nowadays, among the various acaricides used in the control of V. destructor, most presents increasing resistance. An interesting alternative could be the identification of existent molecules as new acaricides with no effect on honeybee health. We have previously constructed the first 3D model of AChE for honeybee. By analyzing data concerning amino acid mutations implicated in the resistance associated to pesticides, it appears that pirimicarb should be a good candidate for varroacide. To check this hypothesis, we characterized the AChE gene of V. destructor. In the same way, we proposed a 3D model for the AChE of V. destructor. Starting from the definition of these two 3D models of AChE in honeybee and varroa, a comparison between the gorges of the active site highlighted some major differences and particularly different shapes. Following this result, docking studies have shown that pirimicarb adopts two distinct positions with the strongest intermolecular interactions with VdAChE. This result was confirmed with in vitro and in vivo data for which a

clear inhibition of VdAChE by pirimicarb at 10 μM (contrary to HbAChE) and a 100 % mortality of varroa (dose corresponding to the LD50 (contact) for honeybee divided by a factor 100) were observed. These results demonstrate that primicarb could be a new varroacide candidate and reinforce the high relationships between in silico, in vitro, and in vivo data for the design of new selective pesticides.

Database Nucleotide sequence data are available in the DDBJ/EMBL/ GenBank databases under the accession number(s) KJ499538

Many biotic and abiotic factors, alone or in combination, are suspected to be involved in the gradual decline of honeybee (Apis mellifera), named Colony Collapse Disorder (Vanengelsdorp et al. 2009). Among them, the ectoparasitic mite Varroa destructor (Acari: Varroidae) is the main concern related to this syndrome (Rosenkranz et al. 2010; Dainat et al. 2012). In the face of the straight challenges presented by this ectoparasite, beekeepers have become dependent on management techniques for controlling mite infestations, apicultural acaricides playing a predominant role (Rosenkranz et al. 2010). The extensive use of these treatments has lead to the development of resistant population in several countries worldwide (Hillier et al. 2013; Maggi et al. 2011, 2010) with additionally an accumulation of acaricides inside the hive (Smodis Skerl et al. 2009; Wiest et al. 2011). Lambert and collaborators revealed the widespread occurrence of multiple

F. Dulin (*) : C. Zatylny-Gaudin : C. Ballandonne : R. Bureau : M. P. Halm University of Caen Basse-Normandie, Caen, France e-mail: [email protected] F. Dulin : C. Ballandonne : R. Bureau : M. P. Halm UNICAEN, CERMN (Centre d’Etudes et de Recherche sur le Médicament de Normandie, EA 4258, FR CNRS 3038 INC3M – SF 4206 ICORE, Université de Caen Basse - Normandie, U.F.R. des Sciences Pharmaceutiques), F-14032 Caen, France C. Zatylny-Gaudin Université de Caen Basse-Normandie UMR BOREA MNHN, UPMC, UCBN, CNRS-7208, IRD-207, F-14032 Caen, France B. Guillet : R. Bonafos Montpellier SupAgro, USAE, Montpellier, France

Keywords Apis mellifera . Varroa destructor . Acaricide . Acetylcholinesterase . Pirimicarb Abbreviations HbAChE Honey bee acetylcholinesterase VdAChE Varroa destructor acetylcholinesterase DmAChE Drosophila melanogaster acetylcholinesterase

Introduction

4602

chemical residues in beehive matrices from apiaries located in different landscapes of Western France (Lambert et al. 2013). Chemical analysis shows that the ubiquitous contaminants of bees and bee products introduced by beekeepers to control the ectoparasitic mite V. destructor are tau-fluvalinate and coumaphos (Johnson et al. 2013; Mullin et al. 2010). Actually, there is an urgent need for an extended range of acaricides and, particularly, to characterize new chemical control agents that selectively kill an arthropod pest of an arthropod host. In the last few years, drug repositioning (drug repurposing), using old drugs for new therapeutic indications, has been growing up (Cragg et al. 2014). This technique has been suggested to be a more efficient strategy for drug development than the current standard with novel agents. A significant advantage of this concept is that the molecule safety is known and the risk of failure for reasons of adverse toxicology is reduced (Chong and Sullivan 2007; Wei et al. 2012). Limited economic resources are actually allocated for the fight of V. destructor. For these reasons, the “drug repositioning” (“pesticide repositioning” in this case) can be applied for the search of new acaricides against these acari. The chemical used against V. destructor acts on different molecular targets. The most commonly used are nerve and muscle targets: acetylcholinesterase (coumaphos for example), sodium channel (bifenthrine, etc.), octopamine receptor (amitraz, essential oils, etc.), GABA-gated chloride channels (endosulfan, etc.), etc. Others are the respiration targets or growth and development targets, but there are still many acaricides for which the mode of action is still unknown (Insecticide Resistance Action Committee 2012). Natural products are also used for varroa control, including the monoterpenoid thymol (ApilifeVar® and ApiGuard®) and the organic acids oxalic acid (Oxivar®) and formic acid (MiteAwayQuickStrips®), but often, their modes of action remain unknown. Two important classes of pesticides are carbamates and organophosphates corresponding to acetylcholinesterase inhibitors. These classes of pesticides inhibit the acetylcholinesterase enzyme from breaking down acetylcholine, thereby increasing the level of neurotransmitter that results in prolonged activation of cholinergic receptors, followed by their desensitivation (Williamson and Wright 2013; Williamson et al. 2013). To be selective against varroa keeping the honeybee away from the toxic effect, it is essential to understand the molecular interactions between pesticides and AChEs. In an attempt to apprehend these interactions at the catalytic site of honeybee and varroa enzyme, molecular modeling on the complexes was carried out. Up to now, the experimental 3D structures of AChE in honeybee and varroa are unknown. We have previously constructed the first 3D model of AChE for honeybee (HbAChE) with Drosophila melanogaster AChE (DmAChE) crystallographic structure as a template (Dulin et al. 2012). In the same way, in this publication, we proposed a 3D model for

Parasitol Res (2014) 113:4601–4610

the AChE of V. destructor (VdAChE). Moreover, for this last purpose, a molecular genetic approach was done to isolate the VdAChE gene and to determine its primary sequence. This work presents the comparison between the two AChE active sites. Among AChE inhibitors, some of them are good insecticides whereas others are more effective as acaricides. Previous QSAR studies on AChE inhibitor pesticides highlighted two molecules (formetanate and pirimicarb) which have structural features leading to non-toxic effects on honeybee (Dulin et al. 2012). To understand this non-toxicity, mutation data was available only for pirimicarb, one mutation associated to a resistance for a species named Myzus Persicae. In this case, a single substitution of an amino acid (Ser431Phe) was found in the pirimicarb-resistant strains (Nabeshima et al. 2003) and this Phe residue is straight observed for honeybee (Phe386 for HbAChE). Sequence alignment between the honeybee, varroa, and My. persicae AChE highlights the presence of different amino acids in this position (Phe/Leu/Ser, respectively). Our docking results show that molecular interactions involved in the VdAChE or HbAChE/pirimicarb complexes are different and in favor of a better interaction between pirimicarb and VdAChE. According to these in silico results, the kinetics of AChE inhibition of pirimicarb on VdAChE and HbAChE were compared. In the same way, its toxic effect on varroa confirms the potentiality of this candidate as varroacide.

Materials and methods Insects and mites collection Honeybee The experiments were carried out in some apiaries of honeybees derived from A. mellifera, placed in standard hives in the region of Caen (France) (for the AChE inhibitors assays). Varroas Varroas were picked up from brood that beekeepers gave us (i) in the region of Caen (Normandy, France) for the 5′-rapid amplification of cDNA ends (5′-RACE) experiments and the inhibition tests or (ii) in the region of Montpellier (Languedoc Roussillon, France) for the toxicity tests. Adult mites were collected from larvae and non-pigmented pupae. Mites were not taken from adult bees in order to minimize mite mortality in the untreated control vials (Milani 1995). The nymphs of bees intended to insure the nutrition of the adults of V. destructor during the bio-assays are chosen from frames where mites were extracted. They are taken at the stage of nymph with the white or pink eyes.

Parasitol Res (2014) 113:4601–4610

Genome data

4603

and sequenced by genomic platform of University of Nantes (France).

Sequence analysis Homology modeling of the 3D V. destructor AChE Bioinformatics treatment of the genome sequences of V. destructor (Cornman et al. 2010) (accession number BRL_Vdes_1.0 (genome size 294.13 Mb)) was made using pipeline pilot (http://accelrys.com/products/pipeline-pilot/). Protein sequence alignments were conducted using the Multalin program (Corpet 1988) and drawn with the ESPript server (version 2.2) (Gouet et al. 2003) (http://espript.ibcp.fr/ ESPript/cgi-bin/ESPript.cgi). Total RNA preparation Total RNA was isolated from the 400 acari V. destructor obtained from our collection using Tri-Reagent method (Sigma-Aldrich, St. Louis, MO, USA) according to the manufacturer’s instructions. RT-PCR and 5′-RACE PCR Oligonucleotide primers were purchased from Genecust (Luxembourg). Primers were designed based on sequences of acetylcholinesterase VDK0013955-3704 obtained from analysis of the varroa genome. The assembly of the contig VDK0013955-3704 was verified by RT-PCR using primers AChE–F 5′-AGCCAGACCTGGTTGCCTAC-3′ and AChER 5′-GCTCGGTGTCTTCATGCCG-3′. Amplification was realized with Platinum TaqDNA Polymerase High fidelity (Invitrogen) under the following conditions: 2 min at 94 °C, 30 cycles of 30 s at 94 °C, 30 s at 55 °C, 1 min 20 at 72 °C, and 8 min at 72 °C. To complete the sequence, RACE PCR approach was realized, only 5′-RACE PCR was described. The transcriptional start site was determined by oligo-capping RACE methods using a GeneRacer kit (Invitrogen). First-strand complementary DNA (cDNA) was synthesized from mRNA with the GeneRacer Oligo dT Primer supplied in the GeneRacer kit (Invitrogen) according to the manufacturer’s instructions. The first PCR was performed using the GeneRacer 5′ primer and primer AChE5 (5′-TGTGACTTCGTTGGGGTCGCCT CCG-3′) with Platinum TaqDNA Polymerase High fidelity, under the following conditions: 2 min at 94 °C, 5 cycles of 30 s at 94 °C and 2 min at 72 °C, 5 cycles of 30 s at 94 °C and 2 min at 68 °C, and 25 cycles of 30 s at 94 °C, 30 s at 62 °C, and 2 min at 68 °C (7 min for the last cycle). The second PCR was performed using the GeneRacer 3′ nested primer and AChE5N (5′-GTAGGCAACCAGGTCTGGCTCTCGTG GC-3′) under the following conditions: 94 °C for 5 min, 35 cycles of 30 s at 94 °C, 30 s at 62 °C, and 2 min at 72 °C (7 min for the last cycle). All PCR products were subcloned

The @TOME server (Pons and Labesse 2009), by using screening methods like FUGUE (Shi et al. 2001), SP3 (Zhou and Zhou 2005), PSIBLAST (Altschul et al. 1997), and HHSEARCH (Soding 2005), identified the 2.7-Å-resolution crystal structure (PDB: 1DX4) of Dr. melanogaster AChE (DmAChE) complex with tacrine derivative as the better 3D experimental template (sequence identity=29 %) (Harel et al. 2000). The alignment between the two sequences has been manually optimized and was used as the basis for the homology modeling with the Modeller software (Eswar et al. 2008). The folding quality of the VdAChE model was estimated by two 3D evaluation tools: Verify3D (Eisenberg et al. 1997) and Eval23D (Gracy et al. 1993). The modeled structure was further minimized with Discovery studio. Qualitative Model Energy Analysis (QMEAN) scores were also determined to estimate the quality of each homology model generated (Benkert et al. 2008). QMEAN scoring function is based on the linear combination of six structural descriptors and reflects the predicted global model reliability ranging from 0 to 1. Docking studies Docking calculations were performed with Genetic Optimisation for Ligand Docking (GOLD) starting from the default parameters (Jones et al. 1995, 1997). The active site of VdAChE was defined as any atom that lies within a 12-Å radius of γ-oxygen of the catalytic Ser 220 (CB3190). Protein flexibility was handled by allowing side chains to rotate with a pre-defined rotamer library for the different conformations of side chain (GOLD). Arg90, Trp143, Gln219, Ser220, Lys305, Tyr307, Glu343, Phe346, Leu347, and His460 were considered to be flexible. Docking solutions were classified by GoldScore values. AChE inhibition assays AChE inhibitors Three acaricides were tested: pirimicarb, formetanate, and coumaoxon. Pirimicarb and formetanate were obtained from Sigma-Aldrich. The coumaphos-oxon was synthesized by our laboratory (Briseño-Roa et al. 2006). The compounds were dissolved in DMSO to produce stock solutions of 10-mM concentration. These stock solutions were then diluted to the working

4604

concentration to 100 μM in sodium phosphate buffer (0.1 M, pH 7.4) for the first set of experiments.

Enzyme assays The pesticides were evaluated on enzyme activities in honeybees as previously described by Loucif-Ayad et al. (W.Loucif-ayad et al. 2008). The AChE activity was carried out following the method of Ellman et al. (1961) using acetylthiocholine as a substrate. Pooled head of bees or whole varroa was homogenized in the following solution containing 38.03 mg ethylenediaminetetraacetic acid (EDTA), 1 ml Triton X-100, 5.845 g NaCl, and 80 ml Tris buffer (10 Mm, pH 7). After centrifugation (5,000g, 5 min), the AChE activity was measured in aliquots (50 μl) of resulting supernatants added to 100 μl of 5-5’-dithiobis-(2-nitrobenzoic acid) (0.705 mM in sodium phosphate buffer 0.1 M, pH 7.4) and 50 μl of sodium phosphate buffer (0.1 M, pH 7.4) or of AChE inhibitors at appropriate concentrations. After 5 min, 50 μl acetylthiocholine (1 mM in sodium phosphate buffer 0.1 M, pH 7.4) was added. The rate of absorbance change was measured at a wavelength of 412 nm. IC50 values were determined graphically from six-point inhibition curves using the Origin software and expressed as IC50 ±SD with three replicates.

Parasitol Res (2014) 113:4601–4610

Statistical analysis Results were expressed as the mean ± SD. The significance of the differences between the mean values was estimated using Kruskall-Wallis and Mann-Whitney tests.

Toxicity tests Experiment protocol A Petri dish with a 3-cm diameter served as experimental device. Lids fenced, placed on every Petri dish, were used to avoid the leaks of mites but especially to maintain a favorable hygrometry for nymphs. Fluon was put on the top of the side wall of the Petri dish to force varroa staying in contact to the chemicals. Treatments were realized by using a Potter tower (Potter 1952). The deposit of each active substance was uniform on the support (Petri dish), homogeneous and equal to 1.5 mg±0,2 mg/cm2. Treated Petri boxes were then placed under an extractor fan during 20 to 30 min to allow the drying of the deposit. After the treatment of supports, adults of V. destructor extracted from the brood were deposited according to four individuals on the bottom of the treated boxes with four replicates. Acarids were left in direct contact with the treatment during 10 min, and then a nymph of bee was

Fig. 1 Nucleotide and deduced amino acid sequences of the VdAChE precursor. Peptide signal is represented in gray and italics

Parasitol Res (2014) 113:4601–4610

4605

Fig. 2 Sequence alignment between VdAChE from our 5′-RACE work (VARROA-RTPCR) and from literature (VARROA_GENOME). The two main differences between the VdAChE are highlighted by green boxes in the N-terminal and C-terminal parts

extracted from the brood and put in every Petri dishes. When experiments were achieved, the experiments were maintained in a climatic chamber 35±1.5 °C and 50±15 % HR. A distilled water control was systematically realized, for correcting the mortality according to Abbot (1925).

l’alimentation de l’environnement et du travail (Anses), AGRITOX. http://www.dive.afssa.fr/agritox/index.php)) and ECOTOX (US Environmental Protection Agency (EPA)), the ECOTOXicology database. http://www.ipmcenters.org/Ecotox/ index.cfm). A control with distilled water was also treated.

Determination of mortality—statistical analysis Varroas were considered dead if they did not move legs when touched by a smooth contact with a paintbrush. Varroa mortality was evaluated after 24 h. As statistical analysis, we made a probit transformation on mortality data and adjusted it as a linear model by using WinDL50 software. Pirimicarb and formetanate were applied at concentrations of 0.2496 and 0.2176 g/l, respectively, corresponding to LD50 honeybee/100. The contact LD50 values of pirimicarb and formetanate to honeybee were respectively 18.7 and 14 μg/bee (according to database: ePesticide Manual (Tomlin 1994), Agritox (Agence Nationale de Sécurité Sanitaire de

Since recently, the whole genome of V. destructor is available (Cornman et al. 2010). In their study, Cornman et al. analyzed the Korean (K) halotype of V. destructor from A. mellifera, the predominant halotype presently found in North America. They have identified over 13,000 contigs with sequence homologous to other species. Many of these have recognized domains and/or functional annotations transfered from other

Fig. 3 Homology modeling of the VdAChE. a 3D cristallographic structure of the DmAChE. b Sequence alignment between DmAChE and VdAChE used for the model construction. β-strands are represented

by green arrow and α-helix by pink arrow. Amino acids of the catalytic triad, oxanion hole, and anionic binding site are respectively marked by different colors. c 3D model of the VdAChE

Results/discussion Sequence analysis of the V. destructor AChE

4606

arthropods (like Ixodes scapularis). After bioinformatic treatment of the raw sequence data and by successive Psiplast against acari known AChE, we identified several genes encoding enzymes close to acetylcholinesterase. By a comparison to other AChEs identified in arthropods, a contig (identified as VDK0013955-3704) of 1,470 bp (partial cDNA sequence RACE PCR for the approach) was considered. Starting from amplification of VdAChE, a RACE-PCR was carried out in parallel (Fig. 1). This approach leads to a correct agreement for the sequences (identity of 93 %, Fig. 2) between the published contig and our data except for some residues in the 5′ end and the 3′ end. The signal peptide of the precursor was identified (in gray on Fig. 1). The 5′ end sequencing (accession KJ499538) allowed to determine completely and correctly the N-terminal extremity of AChE precursor. For the 3′ end, no clear data was currently obtained. In conclusion of this study, the 93 % identity between the two sequences, with differences in the 5′ and 3′ ends, suggests (i) a likely intraspecific variability between the North American and the French varroa despite the fact that the K halotype is the most widespread halotype in the world or (ii) a problem in the assembling of genome by considering the published contig existing. In the next studies, our primary sequence was considered. Molecular modeling The structure of AChE is actually experimentally solved for six species (according to the Protein Data Bank (PDB)): Torpedo californica (80 structures in the PDB), Mus musculus (70 structures), Homo sapiens (20 structures), Dendroaspis angusticeps (10 structures), Dr. melanogaster (3 structures), Electrophorus electricus (3 structures), and Rattus norvegicus (2 structures). Based on the Dr. melanogaster AChE crystallographic structure (template used for the HbAChE 3D model), the 3D model of VdAChE was constructed by homology modeling (sequence identity of 29 %) (Fig. 3). We have also considered alignments with other species to perform the best sequence alignment. A comparison of DmAChE and VdAChE sequences shows some amino acid insertions and deletions (Fig. 3). AChE belongs to the a0b hydrolase fold family, with a core of β-sheets connected by α-helices (Ollis et al. 1992). Superimposition of the representative VdAChE homology model with the DmAChE crystal structure showed a root-mean-square (RMS) of 0.726 Å taking into account all intracellular and extracellular loops. The insertions/ deletions (1 to 16 amino acids) are located in neighboring surface loops and have no effects on the secondary structure elements. A QMEAN score of 0.62 (>0.50) suggests that the homology model produced is reliable. Like in DmAChE (used as a template for the homology modeling), the active site of VdAChE is located at the base of a long and narrow 20-Å gorge (Table 1) which consists of two subsites: an “esterasic” one containing the catalytic machinery and an “anionic” subsite responsible for binding the

Parasitol Res (2014) 113:4601–4610 Table 1 The active site gorge residues of DmAChE, HbAChE, and VdAChE DmAChE

HbAChE

VdAChE

Ser238 His 480 Glu 367 Gly150

Ser253 His496 Glu382 Gly167

Ser220 His460 Glu343 Gly139

“Esterasic” subsite cataytic triade

Gly151 Ala239 Trp83 Thr154 Tyr162

Gly168 Ala254 Trp116 Ser171 Tyr179

Ala140 Ala221 Arg90 Trp143 Tyr151

oxanion hole

Ile484 Tyr148 Glu237 Gly481 Gly149 Gly155 Thr154 Trp472 Glu80 Trp271 Tyr370 Met153 Phe371 Leu328 Phe330 Arg70a Tyr71 Glu69

Val500 Tyr165 Glu252 Gly497 Gly166 Gly172 Ser171 Trp488 Glu113 Trp286 Tyr385 Met170 Phe386 Leu343 Phe345 Arg103 Tyr104 Glu102

Ile464 Tyr137 Gln219 Asn461 Gly138 Gly144 Trp143 Leu452 Lys87 Phe254 Phe346 Ile142 Leu347 Lys305 Tyr307 Glu82 Phe83 Leu81

Tyr374 Trp321 Glu72 Tyr324 Tyr73 Asp375

Tyr389 Trp336 Glu105 Tyr339 Tyr106 Asp390

Ala350 Leu298 Leu298 Glu301 Tyr85 Glu354

“Esterasic” subsite

“Anionic” binding subsite Active site gorge from its bottom to its entrance

Amino acids conserved between species are in bold

quaternary trimethylammonium tail group of acetylcholine (ACh). The active site triad, essential catalytic functional unit, is composed of Ser220, His460, and Glu343 in VdAChE (corresponding to Ser238, His480, and Glu367 in DmAChE). Another important functional unit of the esterasic subsite, named the oxyanion hole, is composed by residues Gly150, Gly151, and Ala239 in DmAChE and Gly139, Ala140, and Ala221 in VdAChE. A co-crystallographic structure for the enzyme-substrate (ACh) complex is not available, but it was established that the choline moiety of ACh interacts with the anionic subsite by a cation-pi interaction with Trp83 (DmAChE) (Dougherty 2013; Scrutton and Raine 1996). This

Parasitol Res (2014) 113:4601–4610

4607

Fig. 4 Molecular surface of the active site gorge of the HbAChE and the VdAChE. Two crucial amino acids are highlighted in ball and sticks. This figure was made with Discovery 3.5

amino acid is conserved in HbAChE but is replaced by Arg 90 in VdAChE. This mutation should not alter the binding of ACh because another tryptophan (Trp143 in VdAChE instead of a Ser171 in HbAChE) is spatially closed to this position leading to a conservation of this cation-pi interaction (Fig. 5a, c). We can then postulate that this Trp143 could play the same role for the interaction of ACh with VdAChE. In AChEs, the active site gorge is coated with aromatic side chains (13 residues in DmAChE), which interact with its various inhibitors via hydrophobic interactions (Harel et al. 1993, 2000). Their side chains render the gorge flexible in accommodating inhibitors by swinging to assume different conformations. According to the list of amino acids described by Harel et al. for the 3D structure of DmAChE, the corresponding residues in VdAChE and HbAChE were reported in Table 1. The composition of this gorge is different between varroa and honeybee with only 5 amino acids conserved between the two organisms on a set of 25 amino acids. This difference leads to a change in the shape of the binding pocket: The cavity associated to the active is site larger in the VdAChE than in HbAChE (678 and 581 Å3, respectively (Fig. 4)). The presence of a leucine (Leu347 in VdAChE) instead of a phenylalanine (Phe386 in HbAChE) contributes especially to this enlargement of the cavity. Docking studies In the previous study, we identified an amino acid (Phe386) in HbAChE which could explain the low toxicity of pirimicarb

to this organism (Dulin et al. 2012). Indeed, it appears that the Ser/Phe mutation on My. persicae AChE leads to a strong decrease in pirimicarb affinity for AChE (IC50) and to a resistance for this insecticide (Nabeshima et al. 2003). This phenylalanine (Phe386) is located between the active site triad and the peripheral anionic site (PAS) in HbAChE. If this Phe386 is mutated by a Serine in HbAChE (like in My. Persicae), the cavity volume increases leading to a new polar cavity and a new position of pirimicarb. The two positions of the pesticide are illustrated in Fig. 5a, b. According to the sequence and the 3D available model of the VdAChE, this phenylalanine (in HbAChE) is replaced by a leucine (Leu347) in VdAChE. The docking of pirimicarb was made with the same parameters as in HbAChE and Phe/Ser386 HbAChE. Interestingly, the results show that, in VdAChE, pirimicarb adopts the same position as expected with the mutated HbAchE (replacement of Phe386 by a serine), whereas the plane conformation persists. In the HbAChE model, the pirimicarb binds to the active site with a parallel pi stacking between the Trp116 and the pyrimidine cycle of pirimicarb (Fig. 5a). Whereas, in the VdAChE, the position of pirimicarb flipped compared to the previous one, the carbamate function is close to the catalytic serine (Fig. 5c). So, the presence of a leucine (Leu347 in VdAChE) instead of a phenylalanine (Phe386 in HbAChE) and the replacement of the Trp116 (in HbAChE) by Arg 90 (in VdAChE) allow pirimicarb to adopt a position close to the position observed in the mutated HbAChE. In VdAChE, the two methyl groups of the amine

Fig. 5 Docking poses of pirimicarb. Model of pirimicarb bound to the HbAChE (a), the Phe/Ser386 HbAChE (b), and the VdAChE (c). This figure was made with pymol

4608

Parasitol Res (2014) 113:4601–4610

Table 2 Inhibition of varroa and honeybee AChE for coumaoxon, formetanate, and pirimicarb, n=3

Varroa AChE Compound Coumaoxon Formetanate Pirimicarb

Honeybee AChE

Percent inhibition, 10–5 M 100±0 % 87±3 % 66±7 %

function of pirimicarb interact with the hydrophobic side chain of Leu 347, and the pyrimidine cycle of pirimicarb interacts with Arg90 by a pi-cation interaction. The presence of these interactions between pirimicarb and VdAChE and an orientation in agreement with the docking position of pirimicarb in Phe/Ser386 HbAChE complex show clear differences with the pirimicarb/HbAChE complex.

AChE inhibition of pirimicarb, formetanate, and coumaoxon for honeybee and varroa Biochemical assays were performed to assess the ability of pirimicarb to inhibit AChE extracted from honeybee brain and varroa. If the inhibition of AChE activity test was performed on honeybees (Badiou et al. 2008; Williamson et al. 2013), no reference is available on V. destructor. Three compounds have been tested: pirimicarb, formetanate, and coumaoxon. For the last one, coumaphos is inactive as an AChE inhibitor and requires metabolic conversion to coumaoxon. We decided to add coumaoxon to these tests as a reference for a VdACHE inhibitor with low/middle toxicity for honeybee, and formetanate, another compound, emerged from our previous QSAR study as low toxic for honeybee. The in vitro inhibitory effect (at 10−5 M) and the IC50 of the three compounds on VdAChE and HbAChE activity are shown in Table 2. The inhibition percentage of AChE differs significantly between V. destructor and honeybee for pirimicarb (p25 (Ruffinengo et al. 2005). If we compare the efficacy obtained with pirimicarb and formetanate with the acaricidal effects of usual molecules against varroa, values are similar. For example, Elzen et al. reported LC50 values of amitraz, coumaphos, and fluvalinate of 16.4, 6.3, and 0.56 μg by Petri dish, respectively, for susceptible mite populations (Elzen et al. 2000). The applied quantities of Table 3 Contact toxicity of formetanate and pirimicarb on Varroa destructor after 24-h exposition 24 h Rep 1 Rep 2 Rep 3 Rep 4 Nb live varroas Nb tot. Mortality Perc.

Control 4 4 3 4 15 16 6.25 %

Formetanate 0 0 0 0 0 16 100.00 %

Pirimicarb 0 0 0 0 0 16 100.00 %

Nb live varroas number of live varroas in the Petri dish, Nb tot. number of total varroas in the Petri dish, Mortality Perc. mortality percentage

Parasitol Res (2014) 113:4601–4610

pirimicarb and formetanate in our study are 2.6 and 1.97 μg/Petri dish, respectively (corresponding to concentrations of 0.2496 and 0.2176 g/l). The mite populations in which chemical toxic effects are analyzed, in this study, should be also considered as susceptible to pirimicarb and formetanate. Nevertheless, as the percentage of mortality is 100 % at the tested concentration, it should be interesting to test at lower concentrations. Moreover, to preserve honeybee health and to avoid potential sublethal effect on honeybee, applied concentration should be as low as possible. Indeed, Palmer et al. suggest that AChE inhibitors as coumaoxon inhibit also nicotinic responses in honeybees suggesting cognitive impairment (Palmer et al. 2013). Recently, Williamson et al. conducted AChE activity assays on honeybee to demonstrate that exposure to AChE inhibitors (coumaphos, coumaoxon, chlorpyrifos, chlorpyrifos-oxon, aldicarbe) alters the physiology and motor function on honeybees (Williamson et al. 2013). Nevertheless, these inhibitors increase the amount and frequency of grooming behavior promoting the natural removal of varroa by honeybees (Williamson et al. 2013). The challenge is to find a balance between a beneficial alteration of the behavior of bees (grooming behavior) and deleterious sublethal effects. We cannot observe a difference between pirimicarb and formetanate toxicities on V. destructor and honeybees in vivo whereas the AChE inhibition profile is different. Formetanate has a profile close to coumaoxon (precursor of coumaphos) for AChE. Formetanate (Palmer et al. 2013) was also described as a potential inhibitor of the octopamine receptor (G-protein-coupled receptor), a target of another well-known varroacide, amitraz (Evans and Gee 1980).

Conclusion In conclusion, among the various acaricides used in the fight against V. destructor, most presents increasing resistance to varroa. An original alternative is the identification of existent molecules as new acaricides with no effect on honeybee health: pesticide repositioning. Within this context, according to our previous study, we hypothesized that pirimicarb should be a good varroacide candidate. In order to confirm this assumption, based on RT-PCR results, we isolated the VdAChE sequence, constructed the first VdAChE 3D model, and built the pirimicarb/HbAChE and VdAChE complexes. The in vitro and in vivo results reinforce our in silico hypothesis: The pirimicarb affinity for AChE is higher in varroa than in honeybee leading to a difference in its toxicity. Although pirimicarb is not toxic for honeybee by acute contact, sublethal experiments need to be conducted. This new research results should be interesting according to the physicochemical properties of pirimicarb. Indeed, logP (the logarithm of the partition coefficient between n-octanol and water) of reference molecules amitraz, tau-fluvalinate, and

4609

coumaphos is respectively 5.5, superior to 3.8 and 4.13 (according to the ePesticide manual (Tomlin 1994)). The pirimicarb logP (1.7) highlights that pirimicarb is less lipophile than the others, and we can expected that it will not accumulate in the honeybee and its products. Actual damage to bee health is a function of toxicity and exposure of the compound, but in combination with the mode of application. Special interest must be focused now on the formulation of this new varroacide candidate. As we noticed in the introduction, the advantage of the pesticide repositioning is that the molecule safety is known. Indeed, pirimicarb is a selective insecticide used to control aphids on vegetable, cereal, and orchard crops, and the overall conclusion of the risk assessment by EFSA is that there is no genotoxic potential for this molecule (European Food Safety Authority 2005). Finally, this approach needs to be applied to other modes of action (octopamine receptor, nicotinic receptor, etc.) in order to avoid resistance. Acknowledgments This work was supported by a fellowship from the Conseil Régional of Basse Normandie (DARM). The authors thank the beekeepers of Abeille Normande du Calvados for their assistance in animal care and for the varroa harvest.

References Abbot SW (1925) A method of computing the effectiveness of an insecticide. J Econ Entomol 18:265–267 Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25:3389–3402 Badiou A, Meled M, Belzunces LP (2008) Honeybee Apis mellifera acetylcholinesterase: a biomarker to detect deltamethrin exposure. Ecotoxicol Environ Saf 69:246–253 Benkert P, Tosatto SCE, Schomburg D (2008) QMEAN: a comprehensive scoring function for model quality assessment. Proteins 71: 261–277 Briseño-Roa L, Hill J, Notman S, Sellers D, Smith AP, Timperley CM, Wetherell J, Williams NH, Williams GR, Fersht AR, Griffiths AD (2006) Analogues with fluorescent leaving groups for screening and selection of enzymes that efficiently hydrolyze organophosphorus nerve agents. J Med Chem 49:246–255 Chong CR, Sullivan DJ (2007) New uses for old drugs. Nature 448:645– 646 Cornman SR, Schatz MC, Johnston SJ, Chen Y-P, Pettis J, Hunt G, Bourgeois L, Elsik C, Anderson D, Grozinger CM, Evans JD (2010) Genomic survey of the ectoparasitic mite Varroa destructor, a major pest of the honey bee Apis mellifera. BMC Genomics 11:602 Corpet F (1988) Multiple sequence alignment with hierarchical clustering. Nucleic Acids Res 16:10881–10890 Cragg GM, Grothaus PG, Newman DJ (2014) New horizons for old drugs and drug leads. J Nat Prod 77(3):703–723 Dainat B, Evans JD, Chen YP, Gauthier L, Neumann P (2012) Dead or alive: deformed wing virus and Varroa destructor reduce the life span of winter honeybees. Appl Environ Microbiol 78:981–987 Dougherty DA (2013) The cation-π interaction. Acc Chem Res 46:885–893 Dulin F, Halm-Lemeille M, Lozano S, Lepailleur A, Sopkova-de Oliveira Santos J, Rault S, Bureau R (2012) Interpretation of honeybees

4610 contact toxicity associated to acetylcholinesterase inhibitors. Ecotoxicol Environ Saf 79:13–21 Eisenberg D, Luthy R, Bowie JU (1997) VERIFY3D: assessment of protein models with three-dimensional profiles. Methods Enzymol 277:396–404 Ellman GL, Courtney KD, Andres V, Featherstone RM (1961) A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem Pharmacol 7:88–95 Elzen PJ, Baxter JR, Spivak M, Wilson WT (2000) Control of Varroa jacobsoni Oud. resistant to fluvalinate and amitraz using coumaphos. Apidologie 31:437–441 Eswar N, Eramian D, Webb B, Shen MY, Sali A (2008) Protein structure modeling with MODELLER. Methods Mol Biol 426:145–159 European Food Safety Authority (2005) Conclusion regarding the peer review of the pesticide risk assessment of the active substance: pirimicarb. EFSA Scientific Report 43: 1–76 Evans PD, Gee JD (1980) Action of formamidine pesticides on octopamine receptors. Nature 287:60–62 Gouet P, Robert X, Courcelle E (2003) ESPript/ENDscript: extracting and rendering sequence and 3D information from atomic structures of proteins. Nucleic Acids Res 31:3320–3323 Gracy J, Chiche L, Sallantin J (1993) Improved alignment of weakly homologous protein sequences using structural information. Protein Eng 6:821–829 Harel M, Schalk I, Ehret-Sabatier L, Bouet F, Goeldner M, Hirth C, Axelsen PH, Silman I, Sussman JL (1993) Quaternary ligand binding to aromatic residues in the active-site gorge of acetylcholinesterase. Proc Natl Acad Sci U S A 90:9031–9035 Harel M, Kryger G, Rosenberry TL, Mallender WD, Lewis T, Fletcher RJ, Guss JM, Silman I, Sussman JL (2000) Three-dimensional structures of Drosophila melanogaster acetylcholinesterase and of its complexes with Two potent inhibitors. Protein Sci 9:1063–1072 Hillier NK, Frost EH, Shutler D (2013) Fate of dermally applied miticides fluvalinate and amitraz within honey bee (Hymenoptera: Apidae) bodies. J Econ Entomol 106:558–565 Insecticide Resistance Action Committee (2012) IRAC MoA Classification Scheme Johnson RM, Dahlgren L, Siegfried BD, Ellis MD (2013) Acaricide, fungicide and drug interactions in honey bees (Apis mellifera). PLoS ONE 8:e54092 Jones G, Willett P, Glen RC (1995) Molecular recognition of receptor sites using a genetic algorithm with a description of desolvation. J Mol Biol 245:43–53 Jones G, Willett P, Glen RC, Leach AR, Taylor R (1997) Development and validation of a genetic algorithm for flexible docking. J Mol Biol 267:727–748 Lambert O, Piroux M, Puyo S, Thorin C, L’hostis M, Wiest L, Buleté A, Delbac F, Pouliquen H (2013) Widespread occurrence of chemical residues in beehive matrices from apiaries located in different landscapes of Western france. PLoS ONE 8:e67007 Loucif-ayad W, Aribi1 N, Smagghe G, NS (2008) Evaluation of secondary effects of some acaricides on apis—disease drug information sharing. African Entomology 18 Maggi MD, Ruffinengo SR, Negri P, Eguaras MJ (2010) Resistance phenomena to amitraz from populations of the ectoparasitic mite Varroa destructor of Argentina. Parasitol Res 107:1189–1192 Maggi MD, Ruffinengo SR, Mendoza Y, Ojeda P, Ramallo G, Floris I, Eguaras MJ (2011) Susceptibility of Varroa destructor (Acari: Varroidae) to synthetic acaricides in Uruguay: Varroa mites’ potential to develop acaricide resistance. Parasitol Res 108:815–821 Milani N (1995) The resistance of Varroa jacobsoni Oud to pyrethroids: a laboratory assay. Apidologie 26:415–429

Parasitol Res (2014) 113:4601–4610 Mullin CA, Frazier M, Frazier JL, Ashcraft S, Simonds R, vanEngelsdorp D, Pettis JS (2010) High levels of miticides and agrochemicals in north american apiaries: implications for honey bee health. PLoS ONE 5: e9754 Nabeshima T, Kozaki T, Tomita T, Kono Y (2003) An amino acid substitution on the second acetylcholinesterase in the pirimicarbresistant strains of the peach potato aphid, Myzus persicae. Biochem Biophys Res Commun 307:15–22 Ollis DL, Cheah E, Cygler M, Dijkstra B, Frolow F, Franken SM, Harel M, Remington SJ, Silman I, Schrag J (1992) The alpha/beta hydrolase fold. Protein Eng 5:197–211 Palmer MJ, Moffat C, Saranzewa N, Harvey J, Wright GA, Connolly CN (2013) Cholinergic pesticides cause mushroom body neuronal inactivation in honeybees. Nat Commun 4:1634 Pons JL, Labesse G (2009) @TOME-2: a new pipeline for comparative modeling of protein-ligand complexes. Nucleic Acids Res 37: W485–W491 Potter C (1952) An improved laboratory apparatus for applying direct sprays and surface films, with data on the electrostatic charge on atomized spray fluids. Ann Appl Biol 39:1–28 Rosenkranz P, Aumeier P, Ziegelmann B (2010) Biology and control of Varroa destructor. J Invertebr Pathol 103(Suppl):S96–S119 Ruffinengo S, Eguaras M, Floris I, Faverin C, Bailac P, Ponzi M (2005) LD50 and repellent effects of essential oils from Argentinian wild plant species on Varroa destructor. J Econ Entomol 98:651–655 Scrutton NS, Raine AR (1996) Cation-pi bonding and amino-aromatic interactions in the biomolecular recognition of substituted ammonium ligands. The Biochemical j 319(Pt 1):1–8 Shi J, Blundell TL, Mizuguchi K (2001) FUGUE: sequencestructure homology recognition using environment-specific substitution tables and structure-dependent gap penalties. J Mol Biol 310:243–257 Smodis Skerl MI, Velikonja Bolta S, Basa Cesnik H, Gregorc A (2009) Residues of pesticides in honeybee (Apis mellifera carnica) bee bread and in pollen loads from treated apple orchards. Bull Environ Contam Toxicol 83:374–377 Soding J (2005) Protein homology detection by HMM-HMM comparison. Bioinformatics 21:951–960 Tomlin C (1994) The pesticide manual crop protection publication U. S. Environmental Protection Agency (EPA), n.d. the ECOTOXicology database. http://www.ipmcenters.org/Ecotox/index.cfm Vanengelsdorp D, Evans JD, Saegerman C, Mullin C, Haubruge E, Nguyen BK, Frazier M, Frazier J, Cox-Foster D, Chen Y, Underwood R, Tarpy DR, Pettis JS (2009) Colony collapse disorder: a descriptive study. PLoS ONE 4:e6481 Wei Y, Ma Y, Zhao Q, Ren Z, Li Y, Hou T, Peng H (2012) New use for an old drug: inhibiting ABCG2 with sorafenib. Mol Cancer Ther 11:1693– 1702 Wiest L, Buleté A, Giroud B, Fratta C, Amic S, Lambert O, Pouliquen H, Arnaudguilhem C (2011) Multi-residue analysis of 80 environmental contaminants in honeys, honeybees and pollens by one extraction procedure followed by liquid and gas chromatography coupled with mass spectrometric detection. J Chromatogr 1218(34):5743–5756 Williamson SM, Wright GA (2013) Exposure to multiple cholinergic pesticides impairs olfactory learning and memory in honeybees. The J of exp biol 216:1799–1807 Williamson SM, Moffat C, Gomersall MAE, Saranzewa N, Connolly CN, Wright GA (2013) Exposure to acetylcholinesterase inhibitors alters the physiology and motor function of honeybees. Front Physiol 5:4–13 Zhou H, Zhou Y (2005) Fold recognition by combining sequence profiles derived from evolution and from depth-dependent structural alignment of fragments. Proteins 58:321–328