Peptides 26 (2005) 2157–2164

Structural and functional characterization of two novel peptide toxins isolated from the venom of the social wasp Polybia paulista Bibiana M. Souza a , Maria A. Mendes a , Lucilene D. Santos a , Maur´ıcio R. Marques a , Lilian M.M. C´esar a , Roberta N.A. Almeida a , Fernando C. Pagnocca a,b , Katsuhiro Konno c , Mario S. Palma a,∗

b

a CEIS, Department of Biology/IBRC-UNESP (CAT-CEPID/FAPESP), Institute of Immunological Investigations (Millennium Institute-MCT/CNPq), Rio Claro, SP 13506-900, Brazil Department of Biochemistry and Microbiology, IBRC-UNESP, Rio Claro, SP, Brazil c Institute Butantan (CAT-CEPID/FAPESP), S˜ ao Paulo, SP, Brazil

Received 14 February 2005; received in revised form 19 April 2005; accepted 20 April 2005 Available online 29 August 2005

Abstract Two novel inflammatory peptides were isolated from the venom of the social wasp Polybia paulista. They had their molecular masses determined by ESI-MS and their primary sequences were elucidated by Edman degradation chemistry as: • Polybia-MPI: I D W K K L L D A A K Q I L-NH2 (1654.09 Da), • Polybia-CP: I L G T I L G L L K S L-NH2 (1239.73 Da). Both peptides were functionally characterized by using Wistar rat cells. Polybia-MPI is a mast cell lytic peptide, which causes no hemolysis to rat erythrocytes and presents chemotaxis for polymorphonucleated leukocytes (PMNL) and with potent antimicrobial action both against Gram-positive and Gram-negative bacteria. Polybia-CP was characterized as a chemotactic peptide for PMNL cells, presenting antimicrobial action against Gram-positive bacteria, but causing no hemolysis to rat erythrocytes and no mast cell degranulation activity at physiological concentrations. © 2005 Elsevier Inc. All rights reserved. Keywords: Polybia paulista; Hymenoptera insect; Polycationic peptide; Wasp venom; Mastoparans; Chemotactic peptides

1. Introduction Stings by Hymenoptera insects such as hornets, yellow jackets and honeybees are manifested by symptoms of pain, local edema and cardiovascular disturbances [5]. The Hymenoptera venoms are constituted of biogenic amines (histamine, serotonin, dopamine, and norepinephrine), some proteins (phospholipases, hyaluronidase, antigen 5) and a series of biologically active polycationic peptides such as: mastoparans, chemotactic peptides and waspkinins among ∗

Corresponding author. Tel.: +55 193526 4163; fax: +55 193534 8523. E-mail address: [email protected] (M.S. Palma).

0196-9781/$ – see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.peptides.2005.04.026

the wasps; mellitin, MCD-peptide and apamine among the bees [1,15]. Mast cell degranulating peptides and chemotactic peptides are among the major components of vespid venoms [13]. The mastoparans are amidated tetradecapeptides responsible for histamine release from the mast cells, of serotonine from platelets, of catecholamines and adenylic acids from adrenal chromafin cells [24] and even causes exocytosis in rat pituitary cells [26] and pancreatic ␤-cells [34]. Mastoparans are thought to cause the formation of ion channels in lipid membranes leading to cell lysis [17], and they are also known to increase permeability of ions and small molecules through the biological membranes by

2158

B.M. Souza et al. / Peptides 26 (2005) 2157–2164

forming pores at high peptide concentrations [25]. These peptides also promote an increase of intracellular Ca2+ concentration in neutrophils [27]. Investigations directed toward elucidation of the regulatory mechanism of mastoparans have shown that these peptides are involved with the modulation of the activities of proteins as diverse as GTP-binding proteins (G proteins), phospholipase A2 and phospholipase C [10,32]. The second most important group of biologically active peptides from wasp venoms are the chemotactic peptides, which are tridecapeptides presenting many hydrophobic amino acid residues and generally a single basic residue; they generally attract macrophages and polymorphonuclear leukocytes to the region around the site of stinging [33]. Depending on their primary sequences, these peptides can cause hemolysis and exhibit a reduced activity of mast cell degranulation [11,23,12]. Polycationic peptides usually contain from 12 to 50 amino acids residues, and a net positive charge from +2 to +7 due to the excess of basic amino acid residues; the hydrophobic residues represent more than 50% of their amino acid sequences [8]. These structural features contribute to the formation of amphipathic, ␣-helical conformations, making them able to interact with the anionic components of the bacterial membranes, providing their assembly in these membranes with consequent pore formation [7]. Because of this, some of these peptides also can present antimicrobial activities [15,19]. The amphipathic, ␣-helical conformations also may permit the assembling of some of these cationic peptides with the zwitterionic membranes of mammalian cells, making some of these peptides act as hemolysins of this cells [15]. Recently, novel pharmacologically active peptides from tropical wasp venoms have been reported: three novel mastoparan peptides were described in the venom of Protopolybia exigua, with two of them characterized as modulators of mast cell degranulation by virtue of their interaction with G-protein receptors [21]. Micro-scale bioassay guided low abundant inflammatory peptides, naturally occurring in the venom of the wasp Agelaia pallipes pallipes were sequenced by using tandem mass spectrometry protocols, permitting the identification of novel mastoparan and chemotactic peptides [20]. Two novel inflammatory peptides were described in the venom of the social wasp Polybia paulista, presenting acetylation at the N-terminus, responsible by the modulation of the PMNL cells chemotaxys and histamine release from mast cells [28]. In the present work we report two novel inflammatory peptides isolated from the venom of the social wasp P. paulista. They were purified by HPLC under reversed phase conditions, had their molecular masses determined by ESI-MS and their sequences were determined by automated Edman dedradation chemistry. Their functional characterization revealed that both peptides exhibit inflammatory and antimicrobial activities.

2. Materials and methods 2.1. Sample preparation The wasps collected in Rio Claro-SP, southeast Brazil, were immediately frozen and stored at −20 ◦ C. The venom reservoirs of 3000 worker wasps were removed by dissection with surgical microscissors and washed with 1:1 acetonitrile (MeCN, Aldrich): water containing 0.1% (v/v) trifluoroacetic acid (TFA, Aldrich) to solubilize the peptides. The extract was then centrifuged at 8000 × g during 15 min at 4 ◦ C; the supernatant was collected and used to purify the peptides. 2.2. Peptide purification The supernatant described above was chromatographed in a CAPCELL PACK C-18 UG120 column (10 mm × 250 mm, 5 ␮m, Shisheido) under a linear gradient from 5 to 60% (v/v) MeCN [containing 0.1% (v/v) trifluoroacetic acid], at a flow rate of 2.0 mL/min over 60 min by monitoring at UV 214 nm. The extracts were manually collected and dried by using a lyophilizer (MLW LGA-05, Heto). The peaks of interest (fractions 11 and 12) were re-chromatographed under reversed phase chromatography with the same column described above, by using isocratic elution with 41% (v/v) MeCN for the fraction 11 and 45% (v/v) MeCN for the fraction 12 (containing 0.1% TFA in both situations) at a flow rate of 2.0 mL/min during 20 min at 30 ◦ C. The elution was monitored at 215 nm and fractions were manually collected in 5 mL glass vials. 2.3. ESI mass spectrometry analysis All the mass spectrometric analysis were performed in a triple quadrupole mass spectrometer (MICROMASS, mod. Quattro II). The experimental protocol was based in details described in a previous publication [19] and adapted for the present investigation. The mass spectrometer was outfitted with a standard probe electrospray (ESI—Micromass, Altrinchan). The samples were injected into electrospray transport solvent by using a micro syringe (500 ␮L) coupled to a micro infusion pump (KD Scientific) at a flow rate of 4 ␮L/min. The mass spectrometer was calibrated with intact horse heart myoglobin and its typical cone-voltage induced fragments to operate at resolution 4000. The samples were dissolved in 50% (v/v) acetonitrile [containing 0.1% (v/v) formic acid] to be analyzed by positive electrospray ionization (ESI+) using typical conditions: a capillary voltage of 3.5 kV, a cone voltage of 30 V, a dessolvation gas temperature of 80 ◦ C and flow of nebulizer gas (nitrogen) about 15 L/h and drying gas (nitrogen) 200 L/h. The spectra were obtained in the continuous acquisition mode, scanning from m/z 100 to 2500 at a scan time of 5 s. The acquisition and treatment of data were performed with MassLynx software.

B.M. Souza et al. / Peptides 26 (2005) 2157–2164

2.4. Amino acid sequencing The amino acid sequence was performed by using a gasphase sequencer PPSQ-21 A (Shimadzu) based on automated Edman Degradation Chemistry. 2.5. Biological activities 2.5.1. Mast cell degranulation activity Degranulation was determined by measuring the release of the granule marker, ␤-d-glucosaminidase, which colocalizes with histamine, as previously described [9]. Mast cells was obtained by peritoneal washing of female adult Wistars rats. The mast cells were washed three times by re-suspension and centrifugation in a mast cell medium (150 Mm NaCl (MERCK), 4 mM KCl (MERCK), 4 mM NaH2 PO4 (SYNTH), 3 mM KH2 PO4 (SYNTH), 5 mM glucose (SYNTH), 15 ␮M BSA (SIGMA), 2 mM CaCl2 (MERCK), 50 ␮L Liquemine (5000 UI/0.250 mL, ROCHE)). The cells were incubated with various peptide concentrations for 15 min at 37 ◦ C and after centrifugation the supernatants were sampled for ␤-d-glucosaminidase. Briefly, 50 ␮L substrate, 5 mM p-nitrophenyl-N-acetyl-␤-dglucosaminide in 0.2 M citrate, pH 4.5, and 50 ␮L of the samples of the medium were incubated in 96-well plates for 6 h at 37 ◦ C to yield the chromophore, p-nitrophenol. After incubation, 50 ␮L of the previous solution were added to 150 ␮L of 0.2 M Tris and absorbance was measured at 405 nm. The values were expressed as the percentage of total ␤-d-glucosaminidase activity from rat mast cell suspensions, determined in lysed mast cells in presence of 0.1% (v/v) Triton X-100 (considered as 100% reference). Results are expressed as mean ± S.D. of five experiments. 2.5.2. Mast cell lysis This method measures the leakage of lactate dehydrogenase (LDH, EC 1.1.1 27) from the cytoplasm of rat mast cells into the surrounding medium as an indicator of mast cell lysis by the peptide toxins. LDH catalyzes the reversible reduction of pyruvate to lactate with NADH as coenzyme. The activity of lactate dehydrogenase was assayed with the supernatants of rat peritoneal mast cells incubated with the toxin peptides also used for mast cell degranulation assay as described above (Section 2.5.1). The LDH activity was assayed by using the UV-LDH Assay Kit from Biobras Diagnostics; 20 ␮L each supernatant was pre-incubated with 800 ␮L of LDH buffer (50 mM Tris pH 7.4, containing 1.2 mM pyruvate, 5 mM EDTA) during 5 min, at 25 ◦ C. The standard spectrophotometric LDH assay measured the decreasing of absorbance at 340 nm due to the consumption of NADH in this reaction. The reaction was initiated by the addition of 200 ␮L of LDH substrate (0.15 mM NADH); the kinetics of NADH consumption was monitored by acquiring the decreasing of the absorbance at 340 nm during 12 min (
A340 ) at 25 ◦ C. The results were initially calculated as catalytic units (micromoles

2159

of NADH min−1 at 25 ◦ C and pH 7.4) and then converted into relative activity by using the total LDH activity of rat mast cells lysed in presence of 0.1% (v/v) Triton X-100 (considered as 100% reference). Results are expressed as mean ± S.D. of five experiments. 2.5.3. Hemolytic activity Five hundred microliters of washed rat red blood cells (WRRBC) were washed three times with physiological saline solution and suspended in 50 mL of physiological saline solution [0.9% (m/v) NaCl]. Ninety microliters of this cell solution were incubated with 10 ␮L of peptide solution in different concentrations, at 37 ◦ C for 2 h. The samples were then centrifuged and the absorbance of the supernatants was measured at 540 nm. The absorbance measured from lysed WRRBC in presence of 1% (v/v) Triton X-100 was considered to be 100% and physiological saline solution activity as being 0%. Results are expressed as mean ± S.D. of five experiments. 2.5.4. Chemotaxis assays Chemotaxis was assayed in a specific multi-chamber apparatus (NEURO PROBE) [6] by using polymorphonucleated leukocytes (PMNL), obtained from subcutaneous inflammatory induction in Wistar rats. The upper chambers were filled with 200 ␮L of a PMNL suspension (∼2.7 × 105 cells/mL in 0.9% NaCl solution–physiological saline solution) and the lower chambers were filled with 400 ␮L of physiological saline solution containing various concentrations of the peptides (10−7 to 10−4 M). A polycarbonate membrane containing pores of 10 ␮m of diameter (NEURO PROBE) was placed between both chambers. The chemotaxis chamber was incubated at 37 ◦ C for 1 h. After incubation, cells in the lower chamber were counted by using a Ne¨ubauer chamber under a microscope after violet crystal staining. Results are expressed as mean ± S.D. of five experiments. 2.5.5. Antimicrobial activity The minimal inhibitory concentrations (MIC) of the peptides were determined based on methods previously described elsewhere [2,18]. The microorganisms used were: Bacilus subtilis (CCT 2576), Staphylococcus aureus (ATCC 6538), Escherichia coli (ATCC 25922) and Pseudomonas aeruginosa (ATCC 15422). The assays were performed in 96-well plates. The inocula were prepared in saline solution [0.9% (v/v) NaCl] by suspending a 18 h old culture in M¨uller–Hinton broth to the 0.5 value of a McFarland scale, a concentration equivalent to 1 × 104 cells/mL and 50 ␮L were inoculated into the wells, containing 50 ␮L of solution of each peptide solubilized in M¨uller–Hinton broth medium. The concentration range of the peptides was from 0.5 to 1000 ␮g/mL. The plates were incubated at 37 ◦ C for 24 h and after 20 ␮L of a triphenyltetrazolium chloride solution (TTC) 0.5% (w/v) was added. The plates were then incubated for an additional

2160

B.M. Souza et al. / Peptides 26 (2005) 2157–2164

period of 2 h at 37 ◦ C. The minimal inhibitory concentration was that where the dye was not reduced and tetracycline in a concentration range from 0.2 to 600 ␮g/mL was used as control. Results are expressed as mean ± S.D. of five experiments.

3. Results 3.1. Peptides purification and sequencing The crude venom of P. paulista was initially fractionated under reversed phase-HPLC by using a linear gradient from 5 to 60% (v/v) acetonitrile [containing 0.1% (v/v) trifluoracetic acid]. The chromatographic profile (Fig. 1) shows the existence of 13 peaks, which were submitted to preliminary biological assays. Fractions 1 and 2 were constituted by complex mixtures of endogenous biogenic amines and neurotransmitters, while fractions 3–12 contained unidentified venom components; the fraction 13 consisted of a mixture of two acetylated inflammatory peptides already characterized both structurally and functionally [28] and unknown venom components. The fraction 11 presented some mast cell degranulation, chemotaxis for PMNL cells and potent inhibitory activity against both Gram-positive and Gram-negative bacteria, whereas the fraction 12 presented some mast cell degranulation, chemotaxis for PMNL cells and potent inhibitory activity against Gram-positive bacteria. Afterward, the purity of fractions were tested by ESI-MS. These observations lead to further purification of the peptide components of these fractions by reverse-phase HPLC under isocratic conditions (Fig. 2a and b). The biological activities described above were found associated to the peaks 11C (Fig. 2a) and 12A (Fig. 2b); ESI-MS spectra revealed the high purity of the isolated peptides with molecular ion peaks at m/z 1654.09 and 1239.73 Da for fractions 11C and 12A, respectively (not shown results). Therefore, both peptides were considered pure enough to be sequenced by Edman Degradation Chemistry and to be biologically characterized.

Fig. 2. (a) Chromatogram profile of re-fractionation of fraction 11 under reverse-phase HPLC with a C-18 (ODS) CAPCELL PACK UG120 column (250 mm × 10 mm, 5 ␮m), under isocratic condition (41% (v/v) MeCN (containing 0.1% TFA)), at a flow rate of 2.0 mL/min over 20 min by monitoring at UV 214 nm. (b) Chromatogram profile of re-purification of fraction 12 under reverse-phase HPLC with a C-18 (ODS) CAPCELL PACK UG120 column (250 mm × 10 mm, 5 ␮m), under isocratic condition (45% (v/v) MeCN (containing 0.1% TFA)), at a flow rate of 2.0 mL/min over 20 min by monitoring at UV 214 nm.

The primary sequences determined for both peptides were: • Fraction 11C: I D W K K L L D A A K Q I L-NH2 (1654.09 Da) • Fraction 12A: I L G TI L G L L K S L-NH2 (1239.73 Da) The sequences above just fit to the experimental values of the respective molecular masses if the C-terminal residues were considered in the amidated form, as with most of the peptide toxins from the venoms of Hymenoptera [14,30]. The primary sequence of the peptide present in the fraction 11C was compared to other similar peptides from the venom of social wasps as shown in Table 1, revealing some Table 1 Amino acid sequences of Polybia-MPI compared to other mastoparan peptides from social wasp venoms

Fig. 1. Chromatogram profile of fractionation of venom extract of the Polybia paulista venom under reverse-phase HPLC with a C-18 (ODS) CAPCELL PACK UG120 column (250 mm × 10 mm, 5 ␮m), under linear gradient from 5 to 60% (v/v) MeCN (containing 0.1% TFA), at a flow rate of 2.0 mL/min over 60 min by monitoring at UV 214 nm.

Peptide

Primary sequence

Wasp species

Polybia-MPI Mastoparan-A Mastoparan-M Protonectarina-MP Mastoparan-C Protopolybia-MPI Parapolybia-MP Protopolybia-MPII Agelaia-MP Protopolybia-MPIII

IDWKKLLDAAKQIL IKWKAILDAVKKVL INLKAIAALAKKLL INWKALLDAAKKVL INWKALLAVAKKIL INWLKLGKKVSAIL INWKKMAATALKMI INWKAIIEAAKQAL INWLKLGKAIIDAL INWLKLGKAVIDAL

P. paulista Vespa analis Vespa mandarina P. sylveirae Vespa crabro P. exigua P. indica P. exigua A. p. pallipes P. exigua

B.M. Souza et al. / Peptides 26 (2005) 2157–2164

2161

Table 2 Primary sequence of Polybia-CP compared to other chemotactic peptides from social wasps venoms Peptide

Primary sequence

Wasp species

Paulista-CP Protonectin Ves-CP-T Ves-CP-M Ves-CP-X Ves-CP-P

ILGTILGLLKSL ILGTILGLLKGL FLPILGKILGGLL FLPIIGKLLSGLL FLPIIAKLLGGLL FLPIIAKLVSGLL

P. paulista P. sylveirae/A. p. pallipes V. tropica V. mandarina V. xanthoptera Paravespula lewisi

conservation in relation to the sequences of the mastoparan peptides, particularly in relation to Protonectarina-MP isolated from the venom of the tropical social wasp Protonectarina sylveirae, which has 71% of similarity in relation to the peptide component of fraction 11C. Thus, this peptide was classified as a mastoparan (MP) and was named Polybia-MPI. When the sequence of peptide component from fraction 12A is compared to the sequences available in the literature, as shown in Table 2, it is possible to observe 90% of similarity in relation to the chemotactic peptide previously reported as Protonectin in the venoms of two different species of neotropical social wasps, Protonectarina sylveirae [4] and A. p. pallipes [20]. Thus, the peptide isolated from fraction 12A will be referred as Polybia-CP. 3.2. Biological activities Biological activities of Polybia-MPI and Polybia-CP were investigated by assaying mast cell degranulation, release of LDH activity, hemolysis, chemotaxis and antimicrobial activities. Fig. 3 shows the results of rat peritonial mast cell degranulation for Polybia-MPI and Polybia-CP, revealing that both peptides presented a reduced activity at 10−5 M, while a reasonable activity was observed when the peptides concentration was 10−4 M (62 and 52% degranulation for Polybia-MPI

Fig. 3. Degranulation activity in rat peritoneal mast cell. The activity was determined by measuring the release of the granule marker, ␤-dglucosaminidase, which co-localizes with histamine and the values for ␤-d-glucosaminidase released in the medium were expressed in the percentage of total ␤-d-glucosaminidase. Values are mean ± S.D. (n = 5).

Fig. 4. LDH activity in rat peritoneal mast cell. The activity was determined by measuring the presence of the lactate dehydrogenase activity to the medium; the results were shown as relative activity by using the total LDH activity contents of rat mast cells lysed in presence of 0.1% (v/v) Triton X-100 (considered as 100%). Results are expressed as mean ± S.D. (n = 5).

and Polybia-CP, respectively); in fact, this concentration is too high to speculate about any important action under physiological conditions. Fig. 4 shows the results of rat peritoneal mast cell lysis by LDH measurement in the medium. Polybia-MPI presented about 50% of LDH activity at 4.5 × 10−5 M and PolybiaCP only reach this percentage at concentration higher than 10−4 M. Hemolysis of rat erythrocytes was also examined for both peptides and the results are represented in Fig. 5. Polybia-MPI presented no hemolytic activity, while Polybia-CP presented a very reduced hemolysis at 10−5 M, but caused 78% hemolysis at the concentration of 10−4 M. Polybia-MPI and Polybia-CP presented a high chemotaxys of PMNL cells (2.4 × 104 and 1.7 × 104 cells/ml, respectively) at the concentration of 10−5 M (Fig. 6). The antimicrobial activity was examined and the results summarized in Table 3. Polybia-MPI presented a potent antimicrobial activity against both Gram-positive and Gram-negative bacteria, while Polybia-CP demonstrated a significant antimicrobial activity only against Gram-positive bacteria.

Fig. 5. Hemolytic activity in washed rat red blood cells (WRRBC). The absorbance measured at 540 nm from lysed WRRBC in presence of 1% (v/v) Triton X-100 was considered as 100%. Values are mean ± S.D. (n = 5).

B.M. Souza et al. / Peptides 26 (2005) 2157–2164

2162

• A central cationic Lys residue is observed at the seventh position in the chemotactic peptides from cold climate species, while Polybia-CP and Protonectin contain a cationic Lys residue at the 10th position. • The only structural difference between Protonectin and Polybia-CP is the replacement of the Gly residue at the 11th position in the sequence of Protonectin by a Ser in Polybia-CP.

Fig. 6. Chemotaxis of PMNL cells for Polybia-MPI and Polybia-CP. Values are mean ± S.D. (n = 5). Table 3 Antimicrobial activities of the peptides Polybia-CP and Polybia-MPI Peptides

Polybia-MPI Polybia-CP Tetracycline (reference)

Minimal inhibitory concentration (␮g/mL) E. coli

P. aeruginosa

B. subtilis

S. aureus

8 250 2

8 250 18

4 15 18

15 15 0.5

4. Discussion The results above reveal two novel peptide toxins from the venom of the social wasp P. paulista, presenting a series of different biological activities and the following primary sequences: • Polybia-MPI: I D W K K L L D A A K Q I L-NH2 • Polybia-CP: I L G T I L G L L K S L-NH2 One of the characteristic structural features of the mastoparan peptides is the existence of some basic amino acid residues (generally lysine), which are responsible for the formation of positive charges, generally at the positions 4, 5, 11 and 12. In most mastoparan peptides from two to four Lys residues are located in the positions mentioned above. Thus, Polybia-MPI presents the Lys residues at the positions 4, 5 and 11, while its similar natural analogue ProtonectarinaMP presents the basic residues at the positions 4, 11 and 12 (Table 1). The primary sequences of Polybia-CP and other chemotactic peptides isolated from venoms of other species of social wasps [22,24] are compared in Table 2. This comparison reveals some interesting structural aspects: • Most chemotactic peptides of the wasps endemic to cold regions of the planet (Ves-CP-T, Ves-CP-M, Ves-CP-X and Ves-CP-P) present a sequence of 13 amino acid residues, while the chemotactic peptides from the venom of tropical species (Polybia-CP and Protonectin) present 12 residues. • Most chemotactic peptides of wasp species endemic to cold regions of the planet present a characteristic FLP tripeptide at the amino terminal side, which is missing in Polybia-CP and Protonectin.

Delivery of stored compounds from mast cell granules may occur either due to the cytolytic effect of the peptides or due to exocytosis activated by the binding of the peptides to G-protein coupled receptors, which in turn activate a cascade of molecular events, resulting in mast cell degranulation [10]. Some mastoparan peptides seem to be involved in guanylate cyclase activation, either interacting directly with the enzyme or through other proteins [32]. Despite classification as a mastoparan peptide and to present a high similarity of primary sequence when compared to Protonectarina-MP, the rat mast cell degranulation activity of Polybia-MPI is reduced (EC50 = 4.5 × 10−5 M) when compared to Protonectarina-MP (EC50 = 4 × 10−6 M) [3]. Polybia-MPI caused the release of 62% of the contents from rat mast cells at 4.5 × 10−5 M (Fig. 3), however it also promoted the delivering of 50% of LDH activity from these cells at the same concentration range. Therefore, taking into account both results together and the standard deviation of both experiments, it may be concluded that the peptide caused the lysis of 50% of rat mast cells, which in turn led to the delivering the granules contents of these cells. Polybia-MPI induced no hemolytic suggesting that it demonstrates a poor interaction with the zwitterionic membranes of rat erythrocytes (Fig. 5), despite application of increasingly high doses in the assays. Thus, Polybia-MPI seems to be present a much higher affinity to interact with the membranes of mast cells than of erythrocytes. Polybia-CP presented both a reduced mast cell degranulation activity at 10−5 M (Fig. 3) and mast cell lysis (Fig. 4) as expected for the most of chemotactic peptides from wasp venoms [22]; this peptide also has a poor hemolytic activity in rat erythrocytes in the same concentration range (Fig. 5), as well as its natural analogue Protonectin [3,20]. The activities observed at the concentration of 10−4 M is too high to have any physiological meaning. Thus, the peptide PolybiaCP does not seems to interact with the zwiterionic membranes of rat erythrocytes and rat mast cells. Chemotaxis is the phenomenon in which bacteria or single cells of multicellular organisms direct their movements according to the presence of specific chemical signals in their environment. The recruitment of leukocytes to a site of tissue injury caused by a wasp sting, constitutes a cause for inflammatory responses. Mechanistically, it involves a cascade of cellular events precisely regulated by temporal and spatial presentation of a repertoire of molecules in the migrating leukocytes and their surroundings (microenvironments) [16]. Eukaryotic cells in general sense the presence of

B.M. Souza et al. / Peptides 26 (2005) 2157–2164

chemotactic stimuli through stereospecific 7-transmembrane heterotrimeric G-protein coupled receptors; the activity of the chemotactic peptides generally is dependent on specific signals mediated by G-proteins located on the plasma membrane of the chemoattracted cells, making the cell/peptide interaction e relatively selective [22]. Although the mechanism of induction of PMNL cells attraction by the peptides Polybia-MPI and Polybia-CP is not known, these peptides probably elicit their effect by interacting with G-proteins at the level of the PMNL cell membrane, as previously reported for other chemotatic peptides from the venom of the tropical social wasp P. exigua [21]. Generally, the peptide wasp toxins are polycationic peptides which may adopt a ␣-helix conformation and present amphiphylic properties, which are essential to exhibit their biological activities [30]. Thus, they can interact with the anionic components of the bacterial membranes in different ways, sometimes resulting in irreversible damage to the cell. Among the known toxins from wasps venoms the peptides Crabrolin [15], Anoplin [14] and Protonectin [20] were also reported to present antimicrobial activity. Polybia-MPI presented antimicrobial activity against both Gram-positive and Gram-negative bacteria, while PolybiaCP presented antimicrobial action only against the Grampositive bacteria. In order to explain these effects based on the possible interactions of each peptide with the bacterial membranes, the secondary structures of PolybiaMPI and Polybia-CP were predicted by using the software “Consensus Secondary Structure Prediction”, from the Network Protein Sequence Analysis (http://npsa-pbil.ibcp.fr/cgibin/npsa automat.pl?page=/NPSA/npsa seccons.html). The algorithm predicted that Polybia-MPI has about 71.43% of ␣-helix conformation and 28.57% of coil structure, while Polybia-CP was predicted to present 50% of random coil and 50% of ambiguous conformations. Thus, it seems clear that the high content of ␣-helix of Polybia-MPI must aid its assembly in the bacterial membrane, reflecting the lower MIC values observed for this peptide, when compared to PolybiaCP. The cell wall is a complex structure, fundamentally different in Gram-positive and Gram-negative bacteria. The cell wall of bacteria consists of a polymer of disaccharides cross-linked by short chain peptides, forming a type of peptidoglycan called murein. In the Gram-positive bacteria, the cell wall is thick (15–80 nm), consisting of several layers of peptidoglycan complexed with molecules of teichoic acids. In the Gram-negative bacteria, the cell wall is relatively thin (10 nm) and composed of a single layer of peptidoglycan surrounded by a membranous structure, the outer membrane, which invariably also contains lipopolysaccharides. Thus, the outer membrane is more hydrophobic in the Gram-negative than in the Gram-positive bacteria and constitutes the target for attack by hydrophobic agents and other antibiotic agents [29,31]. According to Mendes et al. [19] Protonectin has antimicrobial action against both Gram-positive and Gram-negative

2163

bacteria, while the peptide Polybia-CP was effective only against Gram-positive bacteria. The difference between the sequences of both peptides is the replacement of Gly residue at the 11th of Protonectin, by a Ser residue in Polybia-CP. The side chain for the Gly residue is hydrogen, while for the Ser residue is a hydroxy methyl group. Thus, the C-terminal region of Polybia-CP is more hydrophylic than the equivalent region of Protonectin, which could at first, explains the better interaction of Protonectin with the more hydrophobic characteristics of the outer membrane of the Gram-negative bacteria. Thus, the present investigation describes the isolation, purification, sequencing and biological characterization of two novel biologically active peptide toxins from the venom of the tropical social wasp P. paulista. These toxins constitute members of polycationic, linear peptides deprived of cysteine residues, and demonstrating multifunctional activities such as: mast cell degranulation, chemotaxys of PMNL cells, cytolysis of erythrocytes and antimicrobial activity.

Acknowledgements This work was supported by a grant from the S˜ao Paulo State Research Foundation (FAPESP); M.A.M. is Postdoctoral fellow from FAPESP (Proc. 01/05060-4), B.M.S. and M.R.M are Doctoral students fellows from FAPESP; L.D.S. and L.M.M.C are Doctoral student fellow from CAPES. Mario Sergio Palma (Proc. 300377/2003-5) and Fernando Carlos Pagnocca are researching for the Brazilian Council for Scientific and Technological Development (CNPq).

References [1] Argiolas A, Pisano J. Facilitation of phospholipase A2 activity by mastoparans, a new class of mast cell degranulating peptides from wasp venom. J Biol Chem 1983;258:13697–702. [2] Coyle B, Kavanagh K, McCann M, Devereux M, Geraghty M. Mode of anti-fungal activity of 1,10-phenanthroline and its Cu(II), Mn(II) and Ag(I) complexes. BioMetals 2003;16:321–9. [3] Dohtsu K, Hagiwara K, Palma MS, Nakajima T. Isolation and sequence analysis of peptides from the venom of Protonectarina sylveirae (Hymenoptera Vespidae). Nat Toxins 1993;1:271–6. [4] Dohtsu K, Hagiwara K, Palma MS, Nakajima T. Determination of the structure and mechanism of action of novel peptides from the venom of Protonectarina sylveirae (II). Peptide chemistry. Holland: ESCON Printing Co; 1992. [5] Eno AE. Pharmacological Investigation of Oedema Induced by venom from the Wasp Polistes fuscatus. Toxicon 1997;35:1691–8. [6] Falk W, Goodwin Jr RH, Leonard EJ. Chemotaxis assay. J Immunol Meth 1980;33:239–47. [7] Gallo RL, Huttner KM. Antimicrobial peptides: an emerging concept in cutaneous biology. J Investig Dermatol 1998;111:739–43. [8] Hancock REW, Chapple DS. Peptide antibiotics. Antimicrob Agents Chemother 1999;43:1317–23. [9] Hide I, Bennett JP, Pizzey A, Boonem GM, Bar-Sagi D, Gomperts BD, Tatham PER. Degranulation of individual mast cells in response to Ca2+ and guanine nucleosides: all-or-nothing event. J Cell Biol 1993;123:585–93.

2164

B.M. Souza et al. / Peptides 26 (2005) 2157–2164

[10] Higashijima T, Burnier J, Ross EM. Regulation of Gi and G0 by mastoparan related peptides and hydrophilic amines. J Biol Chem 1990;265:14176–86. [11] Hirai Y, Yasuhara T, Yoshida H, Nakajima T, Fujino M, Kitada C. A new mast cell degranulating peptide “mastoparan” in the venom of Vespula lewisii. Chem Pharm Bull 1979;27:1942–4. [12] Ho CL, Hwang LL. Structure and biological activities of a new mastoparan isolated from the venom of the hornet (Vespa basalis). Biochem J 1991;274:453–6. [13] Ho CL, Shih YP, Wang KT, Yu HM. Enhancing the hypotensive effect and diminishing the cytolytic activity of hornet mastoparan B by d-amino acid substitution. Toxicon 2001;39:1561–6. [14] Konno K, Hisada M, Fontana R, Lorenzi CCB, Naoki H, Itagaki Y, Miwa A, Kawai N, Nakata Y, Yasuhara T, Ruggiero J, Azevedo WF, Palma MS, Nakajima T. Anoplin a novel antimicrobial peptide from the venom of the solitary wasp Anoplius samariensis. Biochim Biophys Acta 2001;1550:70–80. [15] Krishnakumari V, Nagaraj P. Antimicrobial and hemolytic activities of crabolin, a13-residues peptide from the venom of the european Hornet, Vespa crabro, and its analogs. J Peptide Res 1997;50:88–93. [16] Le Y, Gong W, Li B, Dunlop NM, Shen W, Su SB, Ye RD, Wang JM. Utilization of two seven-transmembrane, G-protein coupled receptors, formyl peptide receptor-like 1 and formyl peptide receptor, by the synthetic hexapeptide WKYMVM for human phagocyte activation. J Immunol 1999;163:6777–84. [17] Li ML, Lao RW, Qiu JW, Wang ZJ, Wu TM. Antimicrobial activity of synthetic all-D masoparan M. Int J Antimicrobial Agent 2000;13:203–8. [18] Meletiadis J, M´eis JGM, Mouton JW, Donnelly Verweij PE. Comparison of NCCLS and 3-(4,5-dimethyl-2-thiazyl)-2,5-diphenyl-2Htetrazolium bromide (MTT) methods of in vitro susceptibility testing of filamentous fungi and development of a new simplified method. J Clin Microbiol 2000;38:2949–54. [19] Mendes MA, Souza BM, Santos LD, Palma MS. Structural characterization of novel chemotactic and mastoparan peptides from the venom of the social wasp Agelaia pallipes pallipes by highperformance liquid chromatography/electropray ionization tandem mass spectrometry. Rap Commun Mass Spectrom 2004;18:636–42. [20] Mendes MA, de Souza BM, Marques MR, Palma MS. Structural and biological characterization of two novel peptides from the venom of the neotropical social wasp Agelaia pallipes pallipes. Toxicon 2004;44:67–74. [21] Mendes MA, de Souza BM, Palma MS. Structural and biological characterization of three novel mastoparan peptides from the venom of the neotropical social wasp Protopolybia exigua (Saussure). Toxicon 2005;45:101–6.

[22] Nagashima K, Masamichi K, Saito K, Yasuhara T, Tsukamoto T, Mori M, Fujino M, Nakajima T. The role of lysine at seventh position of wasp chemotactic peptides. Biochem Biophys Res Commun 1990;168:844–9. [23] Nakajima T. Biochemistry of vespid venoms. In: Tu AT, editor. Handbook of natural toxins, vol. 2. New York and Basel: Marcel Dekker; 1984. [24] Nakajima T, Uzu S, Wakamatsu K, Saito K, Miyazawa T, Yasuhara T, Tsukamoto Y, Fujino M. Amphiphilic peptides in wasp venom. Biopolymers 1986;25:115–21. [25] Nakajima T, Wakamatsu K, Mukai H. Mastoparan as a G protein activator. In: Rochat H, Martin-Eauclaire MF, editors. Animal toxins. Basel, Switzerland: Birkh¨auser Verlag; 2000. p. 388–98. [26] Ozaki Y, Matsumoto Y, Yatomi Y, Higashihara M, Kariya T, Kume S. Mastoparan, wasp venom activates platelets via pertussis toxinsensitive GTP-binding proteins. Biochem Biophys Res Commun 1990;170:779–85. [27] Perianin A, Snyderman R, Mastoparan. A wasp venom peptide, identifies two discrete mechanism for elevating cytosolic calcium and inositol trisphosphates in human polymorphonuclear leukocytes. J Immunol 1989;143:1669–73. [28] Ribeiro SP, Mendes MA, Santos LD, Souza BM, Marques MR, Azevedo Jr WF, Palma MS. Structural and functional characterization of N-terminally blocked peptides isolated from the venom of the social wasp Polybia paulista. Peptides 2004;25:2069–78. [29] Schwarz G, Reiter R. Negative cooperativity and aggregation in biphasic binding of mastoparan × peptide to membranes with acidic lipids. Biophys Chem 2001;90:269–77. [30] Sforza ML, Oyama Jr S, Canduri F, Lorenzi CCB, Pertinhez T, Konno K, Souza BM, Palma MS, Ruggiero-Neto J, Azevedo Jr WF, Spisni A. How C-terminal carboxyamidation alters the biological activity of peptides from the venom of the Eumenine solitary wasp. Biochemistry 2004;43:5608–17. [31] Singleton P. Bacteria in biology. In: Biotechnology and medicine. John Wiley & Sons Inc; 2004. [32] Song DL, Chang GD, Ho CL, Chang CH. Structural requirements of mastoparan for activation of membrane-bound guanylate cyclase. Eur J Pharmacol 1993;247:283–8. [33] Yasuhara T, Nakajima T, Fukuda K, Tsukamoto Y, Mori M, Kitada C, Fujino M. Structure and activity of chemotactic peptide from the venom sac of vespinae. In: Munekata E, editor. Peptide chemistry. Osaka: Protein Research Foundation; 1983. p. 185–90. [34] Yokokawa N, Komatsu M, Takeda T, Aizawa T, Yamada T, Mastoparan. Mastoparan, a wasp venom stimulates insulin release by pancreatic islets through pertussis toxin sensitive GTP-binding protein. Biochem Biophys Res Commun 1989;158:712–6.