Toxicon 71 (2013) 105–112

Contents lists available at SciVerse ScienceDirect

Toxicon

journal homepage: www.elsevier.com/locate/toxicon

Top-down sequencing of Apis dorsata apamin by MALDI-TOF MS and evidence of its inactivity against microorganisms

D. Baracchi a,*, G. Mazza a, E. Michelucci b, G. Pieraccini b, S. Turillazzi a,b, G. Moneti b a Department of Biologia Evoluzionistica “Leo Pardi”, University of Firenze, Via Romana 17, 50125 Firenze, Italy b Mass Spectrometry Centre, University of Firenze, Viale G. Pieraccini 6, 50139 Firenze, Italy article info abstract

Article history: Apis mellifera is one of the best characterized among , yet Received 13 March 2013 relatively little is known about venom belonging to other species in the Apis. Received in revised form 18 May 2013 , one of the most important bioactive peptides, has been isolated and characterized Accepted 22 May 2013 in A. mellifera, , Apis dorsata and Apis florea, while apamin has been only Available online 7 June 2013 characterized in A. mellifera and A. cerana. At present, no information is available about the sequence of A. dorsata apamin. Moreover, while the antiseptic properties of melittin and Keywords: MCD peptides are well documented, the antimicrobial activity of apamin has never been MALDI-TOF Venom tested. In the present study, we isolated and characterized apamin from the venom of the Honeybee giant honeybee A. dorsata. We tested the activity of apamin against bacteria and yeasts in a Antimicrobial peptides microbiological assay to gain a more complete understanding of the antimicrobial Peptide sequencing competence of the medium molecular weight venom fraction. We show that A. dorsata apamin toxin has the same primary sequence as apamin in A. mellifera and A. cerana,yet with a different C-terminal amidation. We did not find any antiseptic activity of apamin against any of the tested microorganisms. We discuss the evolutionary processes con- nected to the ecological context of venom use that drove the generation of Apis venom complexity. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction 2006; Nascimento et al., 2006) that target a myriad of re- ceptors and ion channels, making them ideal for pharma- such as scorpions, spiders and stinging hy- ceutical and agrochemical research (King and Hardy, 2013; menopterans have developed venom glands that mainly Morgenstern and King, 2013). Many studies have been produce proteins including several enzymes, peptides, conducted to understand the evolutionary processes that several neurotoxic, antimicrobial and cytolytic compounds drove the generation of this venom complexity (for reviews and low-molecular mass substances such as ionic salts, see Fry et al., 2009). biogenic amines, amino acids and alkaloids (Bettini, 1978; Apis mellifera venom, one of the best characterized Piek, 1986; Kuhn-Nentwig, 2003). Venoms are extremely venoms among Hymenoptera, is composed of a wide spec- complex blends of diverse substances (Escoubas et al., trum of biomolecules, the structure and function of which have been determined in great detail (Kreil, 1973; Hoffman, 1996). Recently, a straightforward characterization of the * Corresponding author. Current address: Queen Mary University of peptidic fraction ranging from 1000 to 4000 Da of the pure London, Research Centre for Psychology, School of Biological and Chem- venom by MALDI-TOF MS allowed the detection of the three ical Sciences, Mile End Road, London E1 4NS, UK. Tel.: þ44 (0)20 7882 4807. major types; apamin, mast cell degranulating peptide E-mail address: [email protected] (D. Baracchi). (MCD) and melittin (Francese et al., 2009). Furthermore,

0041-0101/$ – see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.toxicon.2013.05.020 106 D. Baracchi et al. / Toxicon 71 (2013) 105–112 other minor compounds were detected in the honeybee Hence, this study focuses on the chemical character- venom and they still remain to be identified (Baracchi and ization of apamin isolated from the venom of the giant Turillazzi, 2010; Baracchi et al., 2011). Melittin is a basic honeybee A. dorsata. The peptide sequence was entirely peptide of 26 residues (MWw2.8 kDa), accounts for 45%– obtained by means of top down MALDI-TOF MS experi- 50% of the venom dry weight (de Lima and Brochetto-Braga, ments. Additionally, to complete the framework of the 2003; Ownby et al., 1997) and exhibits an amphipathic antimicrobial competence of the medium MW venom structure; its polar and non-polar ends allow it to interact fraction, a microbiological assay was used to test the ac- with lipid membranes ultimately increasing its perme- tivity of apamin against Gram-positive and Gram-negative ability. Two other peptides are considered major venom bacteria and yeasts. toxins: MCD and apamin. The former causes mast cell break down, accounts for about 2% of the venom dry weight (de 2. Material and methods Lima and Brochetto-Braga, 2003) and comprises 22 residues (MWw2.6 kDa). Apamin is a well character- 2.1. Venom collection ized small peptide (18 amino acids, MWw2.0 kDa) with two disulphide bridges connecting the four cysteine residues in Sixty adult worker were collected from a wild an overlapping pattern that recurs in at least one other colony of A. dorsata located in Bukit Katil (Melaka State, venom peptide component (Gauldie et al., 1978). The sec- ) and killed by freezing. ondary or tertiary structure of apamin is exceptionally sta- All specimens were dissected and drops of their venom ble with respect to pH, temperature, and denaturants were extracted directly from the tip of the sting with a (Miroshnikov et al., 1978; Pease and Wemmer, 1988). Apa- small capillary glass tube after gently squeezing the venom min accounts for less than 2% of venom dry weight, presents sac with a glass plate. All the collected venom were a neurotoxic action (de Lima and Brochetto-Braga, 2003) reunited in one single tube and 2 ml methanol added. The and possesses unusual functional as well as structural tube with the sample was transported to the Italian mass properties (Habermann, 1972). It is remarkable among spectrometry centre at the university of Firenze, Italy, and peptides in its ability to cross the blood–brain barrier and stored at 20 C until MS analyses. act on the . Apamin is known to block þ calcium dependent potassium fluxes by binding to a Ca2 - 2.2. Chemical analysis dependent potassium channel (Banks et al., 1979; Castle et al., 1989; Labbé-Jullié et al., 1991; Ishii et al., 1997). 2.2.1. Chemicals While the antiseptic properties of melittin (from A. melli- Methanol, acetonitrile (ACN), and n-pentane were of fera) and many MCD peptides (from other arthropods) are chromatography grade and purchased of Riedel de Haen well documented (Kuhn-Nentwig, 2003; Konno et al., 2001, (Sigma Aldrich Italia, Milan, Italy). Purified and deionized 2006; Mendes et al., 2004; Souza et al., 2005; Turillazzi et al., water was prepared using a Milli-Q system (Millipore, 2006; Xu et al., 2006), the antimicrobial action for apamin Bedford, MA, USA). Formic acid and trifluoroacetic acid has never been tested. Although apamin lacks the surfactant (TFA) were purchased from Fluka (Sigma Aldrich Italia). character of melittin (Bettini, 1978), this peptide shares at The a-cyano-4-hydroxycinnamic acid (a-CHCA) and least three similar characteristics with a group of membrane 1,5-diaminonaphthalene (1,5-DAN) were obtained from modulators and molecules implicated in defence against Bruker Daltonics (Bremen, Germany). Unless stated other- pathogens, including: scorpion toxin, and scorpion wise, all other reagents were of analytical grade and used as defensins, snake sarafotoxins, plant thionins and human supplied. endothelins (Froy and Gurevitz, 1998). In particular, they have an effect on the membrane potential, have similar gene 2.2.2. General procedure organization, and they share a similar cysteine-stabilized a- Apamin was isolated from the crude venom by HPLC helical (CSH) motif, which involves a Cys-X-X-X-Cys stretch and detected by ESI-TOF. To confirm the presence of cystins, of the a-helix bonded through two disulphide bridges to a we compared alkylated apamin to native apamin using Cys-X-Cys triplet of a b-strand (Kobayashi et al., 1991; MALDI-TOF MS. Top-down sequencing of apamin was done Bonmatin et al., 1992; Bruix et al., 1993; Bulet et al., 2004). with MALDI-TOF instruments using Anchor–chips plate. In contrast to the A. mellifera venom, which has one of the We measured the exact mass of apamin using LTQ- best characterized venoms amongst all Hymenoptera, rela- Orbitrap. tively little is known about the venom composition of other Apis species (reviewed in Schmidt, 1995). The three main 2.2.3. Peptide fractionation venom peptides melittin, apamin and MCD are also present in Extracts were taken to dryness and then resuspended in the venom of Apis cerana, Apis dorsata,andApis andreniformis 3 ml of a solution of water and ACN (70:30 v/v containing (Schmidt, 1995; Baracchi et al., 2011). In a pioneering study 0.5% formic acid). The peptide was separated from the other Kreil (1973) showed that melittin has the same amino acid components of the venom by HPLC using a Series 200 sequence in A. mellifera and A. cerana; melittin of A. dorsata (Perkin–Elmer, Boston, MA, USA) HPLC system including an differs in three amino acid residues (Kreil, 1975). autosampler, a quaternary pump, and an UV–VIS detector Apamin also has the same sequence in A. mellifera and A. coupled with a fraction collector (Biologic BioFrac, BioRad, cerana (Haux et al., 1967; Zhang et al., 2003). However, no Hercules, CA, USA). The RP-HPLC column was a Luna C8, information is available on the sequence of apamin in other 150 4.6 mm, 5 mm (Phenomenex, Torrance, CA, USA), Apis species. operating at a flow rate of 0.75 ml/min. The elution D. Baracchi et al. / Toxicon 71 (2013) 105–112 107 gradient program started at 20% ACN, then to 50% in 3 min, the top-down experiment the matrix was a solution of 1,5- and to 80% ACN in 15 min; deionized water and ACN, both DAN in 50% ACN containing 0.1% TFA at saturation (<20 mg/ containing 0.5% formic acid, were used as the eluents. After ml). The 1,5-DAN solution was prepared shortly before MS the UV detector, the eluate was split in a 1:4 ratio and about experiments because of the instability of 1,5-DAN in ACN. 200 ml/min were directed to the ESI interface of an ESI–TOF One droplet of 0.50 ml of analyte solution and one droplet of mass spectrometer (Mariner, Applied Biosystems, Foster 0.5 ml of matrix solution were deposited on the MALDI MTP City, CA, USA). The collected fractions were dried and AnchorChip target and allowed to dry at ambient temper- resuspended in a solution of water and ACN (70:30 v/v ature, leading to the crystallization of the sample. A total of containing 0.1% formic acid) for further MS analyses. 2500 laser shots were manually acquired and accumulated. Biotools ver. 3.0 (Bruker Daltonics) was used for fragment 2.2.4. Mass spectrometry ion assignment (knowing the sequence from A. mellifera In the ESI–TOF mass spectrometer (Mariner, Applied apamin available on UniProt database [Swiss-Prot code: Biosystems, Foster City, CA, USA) the spray tip and nozzle P01500], [TrEMBL code: B7UUK0]) and for top-down potentials were set to 3.8 kV and 120 V, respectively. The sequencing of the peptide. positive ion ESI mass spectra were recorded. The collected fractions of the venom, prepared as above described, were 2.3. Microbiological assays analysed by MALDI-TOF. The experiments were performed on an Ultraflex-TOF/TOF instrument (Bruker Daltonics) by 2.3.1. Determination of antimicrobial activity of synthetic using Flex ControlÔ (ver. 2.4). Positive ion spectra were apamin acquired in reflectron mode, setting the Ion Source 1 at Both apamin isolated from the A. dorsata venom blend 25 kV, the Ion Source 2 at 21.9 kV and the delay time at 20 and A. mellifera synthetic apamin (product number A9459- ns. The matrix solution was a 10 mg/ml solution of a-CHCA 5 MG, Sigma Aldrich Italia) was used for the microbiological in ACN/Water 70:30 containing 0.1% TFA. One ml of the assays. The Gram-positive strain Bacillus subtilis ATCC 6633 sample was mixed with 1 ml matrix solution and then and the Gram-negative strain Escherichia coli JM109 were spotted on a stainless steel MALDI target plate. Mass used as bacterial indicators. Petri dish with Luria Bertani spectra were obtained by manually accumulating 800 shots (LB) solid medium was plated with 106 E. coli cells from an and acquiring in the mass range m/z 800–3000. External overnight culture in LB medium. Petri dish with Nutrient calibration was performed with the Bruker Standard Pep- Agar (NA, Oxoid) medium was overlaid with 7 ml of NA tide Calibration kit (m/z 1000–3500). inoculated with 106 B. subtilis spores. Saccharomyces cer- ESI-MS spectra were recorded via direct infusion (at evisiae S288C was used as yeast indicator. Petri dish with 5 ml/min flow rate) of the sample solution in the conven- Malt Extract Agar (MEA, Oxoid) was plated with 106 spores tional ESI source of a high-resolution mass spectrometer of the yeast. Five ml of the aqueous solution of the peptide (in LTQ-Orbitrap (Thermo Scientific, San Jose, CA, USA). The a concentration of 100 mg/ml for A. mellifera apamin and working conditions were: spray voltage 3.1 kV, capillary approximately the same concentration for A. dorsata apa- voltage 45 V and capillary temperature 220 C. The sheath min) was spotted onto the plates as well as water solvent as and the auxiliary gases were set, respectively, at 17 and 1 the control. All the plates were incubated for 24 h at 30 C. (arbitrary units). For acquisition, Xcalibur software (ver. 2.0, Penicillin G was used as standard for B. subtilis, while Thermo Scientific) was used; monoisotopic and average Kanamycin was the standard for E. coli and S. cerevisiae. deconvoluted masses were obtained by using the inte- Standards were spotted onto the plates as 1 ml of solution grated Xtract tool. A nominal resolution (at m/z 400) of 50 mg/ml. Antimicrobial activity was indicated by clear 100.000 FWHM was used for spectrum acquisition. MALDI- zones of growth inhibition on the plates. For a more firm TOF MS and ESI-MS analyses was also performed on syn- result the microbiological assays were repeated three times. thetic Apamin of A. mellifera (product number A9459-5 MG, Sigma Aldrich Italia) to obtain comparable spectra. 3. Results

2.2.5. Peptide sequencing 3.1. Chemical analysis A fraction of the isolated peptide was modified by reduction and carboxyamidomethylation before amino 3.1.1. Peptide fractionation and mass spectrometry acid sequencing. A chromatographic peak attributable to apamin was The peptide was taken to dryness and then resuspended detected at 19.37 min retention time during HPLC-ESI-TOF in reduction and alkylation buffer (100 mM ammonium analysis of the extract (Fig. 1). bicarbonate buffer). Reduction was accomplished by the The peptide was characterized by the presence of an addition of 40 ml of DTT solution (5 ml 2M DTT þ 895 ml intense doubly charged species (m/z 1014.6), indicating a water þ 100 ml 1M ammonium bicarbonate), and the so- molecular weight of 2027.2 Da. The corresponding lution was incubated at 56 C for 45 min. The solution was collected fraction was used for further MS analyses. The cooled to room temperature for 10 min, and 40 mlof55mM MALDI-TOF analysis of the venom fraction containing the iodoacetamide solution (10 mg iodoacetamide þ 100 ml1M peptide shows a singly charged signal at m/z 2028.27 (Fig. A ammonium bicarbonate þ 900 ml water) was added to a supplementary material). Synthetic Apamin of A. mellifera final concentration of 3-fold molar excess over DTT and shows a singly charged signal at m/z 2027.23. The mass incubated for 30 min in the dark. The reduced/alkylated difference of 1 Da is attributed to the amidation of A. mel- peptide was analysed by MALDI-TOF as described above. In lifera apamin (see below). The reduction/alkylation process 108 D. Baracchi et al. / Toxicon 71 (2013) 105–112

Fig. 1. HPLC–ESI–TOF analysis of methanol extract of A. dorsata venom. The peak at 19.37 min corresponds to apamin. confirms the presence of four cysteines in the amino acid carried out on the Gram-positive B. subtilis, the Gram- sequence of apamin causing a shift of 232 Da in the MALDI- negative E. coli and the yeast S. cerevisiae. By contrast, TOF spectrum (Fig. A supplementary material). Penicillin G, antibiotic used as positive control for B. subtilis, The experimental data obtained excluded the presence and Kanamycin, aminoglycoside antibiotic used as positive of a modification of a carboxylic group (at the C-terminal control for E. coli and S. cerevisiae gave clear zones of peptide) with a primary amide group as, instead, usually growth inhibition on the plates. The lack of antibiotic ac- occur in venom peptides of others arthropods including tivity of apamin peptides at the tested concentration aculeata hymenopterans species and A. mellifera apamin. against bacteria and yeast was always confirmed. The exact molecular weight mass of A. dorsata apamin was determined using ESI-HRMS. Fig. 2 reports theoretical and 4. Discussion experimental isotopic patterns for the doubly charged ion of apamin; a difference of 18 ppm was observed between the We isolated and characterized apamin toxin from the measured and theoretical masses. The experimental mon- venom of the honeybee A. dorsata. The primary sequence is oisotopic mass of A. dorsata apamin was 2026.83 Da, while similar to that reported for A. mellifera and A. cerana: this of synthetic apamin of A. mellifera was 2025.85 Da. CNCKAPETALCARRCQQH (Haux et al., 1967; Zhang et al., 2003). Moreover, the mass shift obtained in the MALDI- 3.1.2. Peptide sequencing TOF spectra after peptide reduction and carbox- The comparison of the signals registered in the MS/MS yamidomethylation confirms the presence of the two spectrum of extracted apamin with the known A. mellifera disulphide bridges in the intact apamin. We can therefore apamin sequence allowed an almost complete sequence assume that also in A. dorsata apamin the cysteine- coverage; using Biotools 3.0 most fragment ions were stabilized a-helical motif probably involves the Cys1-X- assigned to predicted peptide fragments. A. dorsata apamin Cys3 triplet of the b-strand bonded through two cross is composed of 18 amino acids and its sequence is disulphide bridges to a Cys11–X–X–X–Cys15 stretch of the CNCKAPETALCARRCQQH, which is identical to that of A. a-helix in which the pairs are Cys1–Cys11 and Cys3–Cys15 mellifera and A. cerana (Fig. 3; Table 1). (Miroshnikov et al., 1978; Habermann, 1984). The pharmacological similarity of all the honeybees’ 3.2. Microbiological assays venoms (Schmidt, 1995) and the evolutionary preservation of components like melittin and apamin are not surprising 3.2.1. Determination of antimicrobial activity of synthetic considering the relatively recent radiation of the honeybee apamin group (A. cerana–A. mellifera group divergence occurred No growth inhibition was found in three replicates of A. less than 1Myr during the Pleistocene, Culliney, 1983; dorsata and A. mellifera apamin microbiological tests Ruttner, 1988). Apamin amino acid sequences are the D. Baracchi et al. / Toxicon 71 (2013) 105–112 109

Fig. 2. Theoretical (left) and experimental (right) isotopic pattern for the doubly charged ion of apamin. same in A. mellifera, A. cerana and A. dorsata. Moreover, A. According to the MALDI-TOF and LTQ Orbitrap analyses, mellifera and A. cerana possesses an identical melittin a slight difference was found between the C-terminal amino acid sequence and A. dorsata and Apis florea differ carboxyl group of A. dorsata apamin, which is not amidated, from the former in only 3 and 5 amino acids, respectively. and those of A. mellifera and A. cerana which instead are. At the same time, the evolutionary preservation of specific Several studies have found that the C-terminals amidation core motifs in melittin and apamin illuminates the role that in a number of scorpion toxins is essential for full expres- these motifs have in providing important biological func- sion of their biological activities (Auguste et al., 1990; tions. Indeed, there is evidence to suggest that differences Sabatier et al., 1993; Benkhadir et al., 2004). In particular, in amino acid sequences of melittins are insignificant in the presence of C-terminal carboxyl amidation often in- terms of biological activity (Kreil, 1973, 1975). duces a considerable strengthening of the toxin to receptor

Fig. 3. Scoring of all possible sequences with regard to the top-down spectrum of apamin. Biotools 3.0 assigned most fragment ions to predicted peptide fragments to create an almost complete sequence coverage. The glutamic acid in position 7 is not assigned to any fragment peptide. 110 D. Baracchi et al. / Toxicon 71 (2013) 105–112

Table 1 Top-down MALDI-TOF fragment ions. In bold assigned fragments by Biotools 3.0 to predicted peptides fragment masses. The glutamic acid in position 7 is not assigned to any fragment peptide.

N-term Ion a b C y z C-term Ion 1 C 133.043 161.038 178.064 155.093 140.082 18 H 2 N 247.086 275.081 292.107 283.151 268.140 17 Q 3 C 407.117 435.111 452.138 411.210 396.199 16 Q 4 K 535.212 563.206 580.233 571.241 556.230 15 C 5 A 606.249 634.244 651.270 727.342 712.331 14 R 6 P 703.301 731.296 748.323 883.443 868.432 13 R 7 E 832.344 860.339 877.365 954.480 939.469 12 A 8 T 933.392 961.387 978.413 1114.511 1099.500 11 C 9 A 1004.429 1032.424 1049.450 1227.595 1212.584 10 L 10 L 1117.513 1145.508 1162.534 1298.632 1283.621 9 A 1 C 1277.544 1305.538 1322.565 1399.679 1384.668 8 T 12 A 1348.581 1376,576 1393.602 1528.722 1513.711 7 E 13 R 1504.682 1532.677 1549.703 1625.775 1610.764 6 P 14 R 1660.783 1688.778 1705.804 1696.812 1681.801 5 A 15 C 1820.814 1848.808 1865.835 1824.907 1809.896 4 K 16 Q 1948.872 1976.867 1993.894 1984.937 1969.927 3 C 17 Q 2076.931 2104.926 2121.952 2098.980 2083.969 2 N 18 H 2213.006 2241.000 2258.27 2259.011 2244.000 1 C

interaction giving an irreversible binding and a higher high mass of brood, are rewarding targets for many to mammals (Auguste et al., 1990; Sabatier et al., predators and microorganisms. In response, the evolu- 1993; Benkhadir et al., 2004). Contrasting results are tion of honeybee venom must have tended toward fast- however reported for apamin. While Labbé-Jullié et al. acting in deterring vertebrate aggressors but also to- (1991) found that the deletion of the C-terminal histidine ward strong-acting in killing pathogens. In the honeybee residue only slightly reduced binding and toxicity of apa- venom, melittin has optimized its structure and its min, Devaux et al. (1995) reported a central role of the quantity (that accounts for the 50% of the venom dry carboxamide group in the toxicity of the molecule. If these weight) to fullfil the defence against pathogens. Apamin differences in toxicity will be confirmed by future studies, it lacks of antimicrobial competence and has optimized its would be interesting to discuss this in the light of the structure for enhancing the cytotoxic and neurotoxic different level of vertebrate predation faced by different activity against vertebrates (de Lima and Brochetto- honeybees species. In fact A. dorsata, undoubtedly more Braga, 2003). The presence of these highly specialized protected by its habits to nest on cliff faces and higher peptidesinthevenommaybetheresultofthevarious branches of trees then the cavity dwelling species like A. circumstances of venom use (Free, 1961; Collins and mellifera and A. cerana, would possess venom with a lower Kubasek, 1982; Baracchi et al., 2010b, 2011). Indeed, toxicity to mammals. there are evidences that the exploitation of the venoms Evidence suggests that the antimicrobial activity of for multiple purposes in diverse ecological contexts have different venom peptides is also highly influenced by the resulted in different evolutionary trajectories for the presence of the amide in their C-terminal (Dathe et al., venom toxins of many arthorpods (reviewed in 2001; Sforça et al., 2004; dos Santos Cabrera et al., Morgenstern and King, 2013). 2008). Yet this appears of little importance in the case Many studies have been conducted to understand the of apamin, as we did not find any activity against the evolutionary processes that drove the generation of this tested microorganisms (i.e. Gram positive/negative bac- venom complexity (Fry et al., 2009) and the increasing teria and yeasts) by both the C-terminal amidated and not knowledge on venoms shows how the reduction amidated apamin. Even if we tested only one concentra- of energy costs for biosynthesis of the biomolecules, the tion of both apamins the used quantity is ecological and synergisms between highly effective substances, and the physiological relevant as decidedly higher than that pre- existence of a wide spectrum of specialized substances are sent on the cuticle and the reservoir of individual bees all promoted by natural selection. and in the wax of the comb (Baracchi and Turillazzi, 2010; Baracchi et al., 2011). Acknowledgements In social insects the venom is a multifunctional secretion mainly used as potent weapon for prey The authors thank Dr. Francesca Romana Dani for her immobilization, predator deterrence and pathogen help in chemical analyses. We also thank Prof. Rosly Bin neutralization (Turillazzi, 2006; Moreau, 2013). Indeed, Hashim and Dr. Nurul Atika of the Institute of Biological bees, and ants use venom not only during stinging, Sciences of the University of Malaya and Mr Simon Hok for but also to coat their own cuticle, nest material or brood their support in Malaysia. Our grateful thanks are also to protect them against infections (Obin and Vander extended to the anonymous referees for their useful com- Meer, 1985; Turillazzi et al., 2006; Baracchi et al., ments and suggestions. Research was co-funded by the 2010a, 2012; Tragust et al., 2013). In particular - project APENET (DM 19735/7303/08 del 29/12/2008), and colonies, with their rich store of honey, pollen and PRIN 2008 (prot. 2008KZ82RE). D. Baracchi et al. / Toxicon 71 (2013) 105–112 111

Appendix A. Supplementary data Escoubas, P., Sollod, B., King, G.F., 2006. Venom landscapes: mining the complexity of spider venoms via a combined cDNA and mass spec- trometric approach. Toxicon 47, 650–663. Supplementary data related to this article can be found Francese, S., Lambardi, D., Mastrobuoni, G., la Marca, G., Moneti, G., at http://dx.doi.org/10.1016/j.toxicon.2013.05.020. Turillazzi, S., 2009. Detection of honeybee venom in envenomed tissues by direct MALDI MSI. J.Am.Soc. Mass Spectrom. 20, 112–123. Free, J.B., 1961. The stimuli releasing the stinging response of honeybees. Conflict of interest Anim. Behav. 9, 193–196. Froy, O., Gurevitz, M., 1998. Membrane potential modulators: a thread of – The authors declare that there are no conflicts of scarlet from plants to humans. FASEB J. 12, 1793 1796. Fry, B.G., Roelants, K., Champagne, D.E., Scheib, H., Tyndall, J.D.A., interest. King, G.F., Nevalainen, T.J., Norman, J.A., Lewis, R.J., Norton, R.S., Renjifo, C., de la Vega, R.C., 2009. The toxicogenomic multiverse: convergent recruitment of proteins into venoms. Annu. Rev. References Genom. Hum. Genet. 10, 483–511. Gauldie, J., Hanson, J.M., Shipolini, R.A., Vernon, C.A., 1978. The structures Auguste, P., Hugues, M., Grave, B., Gesquiere, J.C., Maes, P., Tartar, A., of some peptides from bee venom. Eur. J. Biochem. 83, 405–410. Romey, G., Schweitz, H., Lazdunski, M., 1990. Leiurotoxin I (scylla- Habermann, E., 1972. Bee and venoms. Science 177, 314–322. þ toxin), a peptide ligand for Ca2(þ) – activated K channels. Chemical Habermann, E., 1984. Apamin. Pharmacol. Ther. 25, 255–270. synthesis, radiolabeling, and receptor characterization. J. Biol. Chem. Haux, P., Sawerthal, H., Habermann, E., 1967. Sequence analysis of bee 265, 4753–4759. venom (apamine) from its tryptic and chymotryptic Banks, B.E.C., Brown, C., Burgess, G.M., Burnstock, G., Claret, M., cleavage products. Z. Physiol. Chem. 348, 737–738. Cocks, T.M., Jenkinson, D.H., 1979. Apamin blocks certain Hoffman, D.R., 1996. Hymenoptera venom proteins. Nat. Toxin. 2, 169– neurotransmitter-induced increases in potassium permeability. Na- 186. ture 282, 415–417. Ishii, T.M., Maylie, J., Adelman, J.P., 1997. Determinants of apamin and d- Baracchi, D., Turillazzi, S., 2010. Differences in venom and cuticular pep- tubocurarine block in SK potassium channels. J. Biol. Chem. 272, tides in individuals of Apis mellifera (Hymenoptera: ) deter- 23195–23200. mined by MALDI-TOF MS. J. Physiol. 56, 366–375. King, G.F., Hardy, M.C., 2013. Spider-venom peptides: structure, phar- Baracchi, D., Dapporto, L., Teseo, S., Hashim, R., Turillazzi, S., 2010a. Me- macology, and potential for control of insect pests. Annu. Rev. Ento- dium molecular weight polar substances of the cuticle as tools in the mol. 58, 475–496. study of the , systematics and chemical ecology of tropical Kobayashi, Y., Takashima, H., Tamaoki, H., Kyogoku, Y., Lambert, P., hover wasps (Hymenoptera: Stenogastrinae). J. Zool. Syst. Evol. Res. Kuroda, H., Chino, N., Watanabe, T., Kimura, T., Sakakibara, S., 48, 109–114. Moroder, L., 1991. The cysteine-stabilized ahelix: a common structural Baracchi, D., Cusseau, G., Pradella, D., Turillazzi, S., 2010b. Defence re- motif of ion-channel blocking neurotoxic peptides. Biopolymers 31, actions of Apis mellifera ligustica against attacks from the European 1213–1220. Vespa crabro. E. E. E. 22, 281–294. Konno, K., Hisada, M., Fontana, R., Lorenzi, C.C., Naoki, H., Itagaki, Y., Baracchi, D., Francese, S., Turillazzi, S., 2011. Beyond the antipredatory Miwa, A., Kawai, N., Nakata, Y., Yasuhara, T., Ruggiero, N.J., de Jr., A.W., defence: honey bees venom functions as a component of social im- Palma, M.S., Nakajima, T., 2001. Anoplin, a novel antimicrobial peptide munity. Toxicon 58, 219–229. from the venom of the solitary wasp Anoplius samariensis. Biochim. Baracchi, D., Mazza, G., Turillazzi, S., 2012. From individual to collective Biophys. Acta 1, 70–80. immunity. The role of the venom as antimicrobial agent in the Konno, K., Hisada, M., Naoki, H., Itagaki, Y., Fontana, R., Rangel, M., Stenogastrinae wasp societies. J. Insect Physiol. 58 (1), 188–193. Oliveira, J.S., Cabrera, M.P., Neto, J.R., Hide, I., Nakata, Y., Yasuhara, T., Benkhadir, K., Kharrat, R., Cestele, S., Mosbah, A., Rochat, H., El Ayeb, M., Nakajima, T., 2006. Eumenitin, a novel antimicrobial peptide from the Karoui, H., 2004. Molecular cloning and functional expression of the venom of the solitary eumenine wasp Eumenes rubronotatus. Peptides alpha-scorpion toxin BotIII: pivotal role of the C-terminal region for 11, 2624–2631. its interaction with voltage-dependent sodium channels. Peptides 25, Kreil, G., 1973. Structure of melittin isolated from two species of honey 151–161. bees. FEBS Lett. 33, 241–244. Bettini, S. (Ed.), 1978. Arthropod Venoms. Springer, Heidelberg. Kreil, G., 1975. The structure of Apis dorsata melittin: phylogenetic re- Bonmatin, J.M., Bonnat, J.L., Gallet, X., Vovelle, F., Ptak, M., Reichhart, J.M., lationships between honeybees as deduced from sequence data. FEBS Hoffmann, J.A., Keppi, E., Legrain, M., Achstetter, T., 1992. Two- Lett. 54, 100–102. dimensional 1H NMR study of recombinant insect defensin A in Kuhn-Nentwig, L., 2003. Antimicrobial and cytolytic peptides of water: resonance assignments, secondary structure and global venomous arthropods. Cell Mol. Life Sci. 60, 2651–2668. folding. J. Biomol. NMR 2, 235–256. Labbé-Jullié, C., Granier, C., Albericio, F., Defendini, M.L., Ceard, B., Bruix, M., Jimenez, M.A., Santoro, J., Gonzalez, C., Colilla, F.J., Mendez, E., Rochat, H., Rietschoten, J., 1991. Binding and toxicity of apamin. Eur. J. Rico, M., 1993. Solution structure of gamma 1-H and gamma 1-P Biochem. 196, 639–645. thionins from barley and wheat endosperm determined by 1H-NMR: Mendes, M.A., de Souza, B.M., Marques, M.R., Palma, M.S., 2004. a structural motif common to toxic arthropod proteins. Biochemistry Structural and biological characterization of two novel peptides 32, 715–724. from the venom of the neotropical social wasp Agelaia pallipes Bulet, P., Stocklin, R., Menin, L., 2004. Anti-microbial peptides: from in- pallipes. Toxicon 1, 67–74. vertebrates to vertebrates. Immunol. Rev. 198, 169–184. Miroshnikov, A.I., Elyakova, E.G., Kudelin, A.B., Senyavina, L.B., 1978. Castle, N.A., Haylett, D.G., Jenkinson, D.H., 1989. Toxins in the character- Physicochemical characteristics of apamin, a neurotoxin from honey- ization of potassium channels. Trends Neurosci. 12, 59–65. bee. Apis mellifera. Bioorg. Khim 4, 1022–1028. Collins, A.M., Kubasek, K.J., 1982. Field test of (Hymenoptera: Moreau, S.J.M., 2013. “It stings a bit but it cleans well”: venoms of hy- Apidae) colony defensive behavior. Ann. Entomol. Soc. Am. 74, 383–387. menoptera and their antimicrobial potential. J. Insect Physiol. 59, Culliney, T.W., 1983. Origin and evolutionary history of the honeybees, 186–204. Apis. Bee World 64, 29–38. Morgenstern, D., King, G.F., 2013. The venom optimization hypothesis Dathe, M., Nikolenko, H., Meyer, J., Beyermann, M., Bienert, M., 2001. revisited. Toxicon 63, 120–128. Optimization of the antimicrobial activity of magainin peptides by Nascimento, D.G., Rates, B., Santos, D.M., Verano-Braga, T., Barbosa- modification of charge. FEBS Lett. 501, 146–150. Silva, A., Dutra, A.A.A., Biondi, I., Martin-Eauclaire, M.F., Lima, M.E.D., de Lima, P.R., Brochetto-Braga, M.R., 2003. Hymenoptera venom review Pimenta, A.M.C., 2006. Moving pieces in a taxonomic puzzle: venom focusing on Apis mellifera. J. Venom. Anim. Toxin. Incl. Trop. Dis. 9, 2D-LC/MS and data clustering analyses to infer phylogenetic re- 149–162. lationships in some scorpions from the Buthidae family (Scorpions). Devaux, C., Knibiehler, M., Defendini, M.L., Mabrouk, K., Rochat, H., Toxicon 47, 628–639. Rietschoten, J., Baty, D., Granier, C., 1995. Recombinant and chemical Obin, M.S., Vander Meer, R.K., 1985. Aster flagging by fire ants (Solenopsis derivatives of apamin. Eur. J. Biochem. 231, 544–550. spp.): functional significance of venom dispersal behavior. J. Chem. dos Santos Cabrera, M.P., Costa, S.T.B., de Souza, B.M., Palma, M.S., Ecol. 11, 1757–1768. Ruggiero, J.R., Ruggiero Neto, J., 2008. Selectivity in the mechanism of Ownby, C.L., Powell, J.R., Jiang, M.S., Fletcher, J.E., 1997. Melittin and action of antimicrobial mastoparan peptide Polybia-MP1. Eur. Bio- Phospholipase A, from bee (Apis mellifera) venom cause necrosis of phys. J. 37, 879–891. murine skeletal muscle in vivo. Toxicon 35, 67–80. 112 D. Baracchi et al. / Toxicon 71 (2013) 105–112

Pease, J.H., Wemmer, D.E., 1988. Solution structure of apamin determined and functional characterization of two novel peptide toxins isolated by nuclear magnetic resonance and distance geometry. Biochemistry from the venom of the social wasp Polybia paulista. Peptides 11, 2157– 27, 8491–8498. 2164. Piek, T., 1986. In: Piek, T. (Ed.), Venoms of the Hymenoptera. Academic Tragust, S., Mitteregger, B., Barone, V., Konrad, M., Ugelvig, L.V., Cremer, S., Press, London. 2013. Ants disinfect fungus-exposed brood by oral uptake and spread Ruttner, F.,1988. Biogeography and Taxonomy of Honeybees. Springer, Berlin. of their . Curr. Biol. 23, 76–82. Sabatier, J.M., Zerrouk, H., Darbon, H., Mabrouk, K., Benslimane, A., Turillazzi, S., 2006. Polistes venom: a multifunctional secretion. Ann. Zool. Rochat, H., Martin-Eauclaire, M.F., Van Rietschoten, J., 1993. P05, a Fenn 43, 488–499. new leiurotoxin I-like scorpion toxin: synthesis and structure/activity Turillazzi, S., Mastrobuoni, G., Dani, F.R., Moneti, G., Pieraccini, G., la, M.G., þ relationships of the a-amidated analog, a ligand of Ca2 -activated K’ Bartolucci, G., Perito, B., Lambardi, D., Cavallini, V., Dapporto, L., 2006. channels with increased affinity. Biochemistry 32, 2763–2770. Dominulin A and B: two new antibacterial peptides identified on the Schmidt, J.O., 1995. of venoms from the honeybee genus Apis. cuticle and in the venom of the social paper wasp Polistes dominulus Toxicon 33, 917–927. using MALDI-TOF, MALDI-TOF/TOF, and ESI-ion trap. J. Am. Soc. Mass Sforça, M.L., Oyama Jr., S., Canduri, F., Lorenzi, C.C., Pertinhez, T.A., Spectrom. 3, 376–383. Konno, K., Souza, B.M., Palma, M.S., Neto, J.R., Azevedo, W.F., Spisni, A., Xu, X., Li, J., Lu, Q., Yang, H., Zhang, Y., Lai, R., 2006. Two families of 2004. How C-terminal carboxyamidation alters the biological activity antimicrobial peptides from wasp (Vespa magnifica) venom. Toxicon of peptides from the venom of the eumenine solitary wasp. 2, 249–253. Biochemistry 43, 5608–5617. Zhang, S.F., Shi, W.J., Cheng, J.A., Zhang, C.X., 2003. Cloning and compar- Souza, B.M., Mendes, M.A., Santos, L.D., Marques, M.R., Cesar, L.M., ison of the genes encoding precursor of apamin from the venom of Almeida, R.N., Pagnocca, F.C., Konno, K., Palma, M.S., 2005. Structural honeybees and wasps. Insect Sci. 10, 21–26.