Isolation and characterisation of a new antimicrobial peptide from the skin of Xenopus laevis

International Journal of Antimicrobial Agents 38 (2011) 510–515

Contents lists available at SciVerse ScienceDirect

International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag

Isolation and characterisation of a new antimicrobial peptide from the skin of Xenopus laevis Feng Hou a,b,1 , Jiping Li a,b,1 , Pengpeng Pan a,b , Jing Xu a,b , Linna Liu a,b , Wensen Liu a,b , Bocui Song a,b , Nan Li a,b , Jiayu Wan a,b , Hongwei Gao a,b,∗ a b

Institute of Military Veterinary Science, Academy of Military Medical Science, No. 666, Liu’ying West Road, Changchun, Jilin 130062, China Key Laboratory of Jilin Province for Zoonoses Prevention and Control, No. 666, Liu’ying West Road, Changchun, Jilin 130062, China

a r t i c l e

i n f o

Article history: Received 24 February 2011 Accepted 19 July 2011 Keywords: Antimicrobial peptides Xenopus laevis Skin PGLa-H

a b s t r a c t A new antimicrobial peptide (AMP) named PGLa-H has been isolated from the skin of the African clawed frog (Xenopus laevis) using gel filtration and reverse-phase high-performance liquid chromatography (RP-HPLC). Its amino acid sequence was determined as KIAKVALKAL by Edman degradation, with a molecular weight of 1053.727 Da as analysed by matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI-TOF/TOF-MS). No similar AMP was found by BLAST search. Purified PGLa-H demonstrated antimicrobial ability against the reference bacteria Escherichia coli ATCC 25922 [minimum inhibitory concentration (MIC) = 23.6 ␮g/mL], Staphylococcus aureus ATCC 25923 (MIC = 8.7 ␮g/mL) and Bacillus subtilis (MIC = 14.4 ␮g/mL) and was active against multidrug-resistant meticillin-resistant S. aureus (MRSA) (MIC = 67.8 ␮g/mL). The antimicrobial mechanism for this new peptide was further investigated by transmission electron microscopy. PGLa-H killed cells by destroying the cell membrane. © 2011 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.

1. Introduction Antimicrobial peptides (AMPs) have been isolated and characterised from all living organisms [1,2]. They are effector molecules that initiate the fight against bacterial infection by perturbing the phospholipid bilayer of the bacterial cell membrane, interfering with metabolism or targeting cytoplasmic components [3]. Their unique mechanisms of action offer new possibilities for developing drugs against resistant bacteria [4]. Owing to differences in the organisation of bacterial membranes and those of multicellular organisms, the majority of AMPs are toxic to bacteria but not to eukaryotes [5]. Whilst they often exhibit potencies comparable with those of conventional antibiotics, the physical nature of their action implies a different and faster killing process. Modern amphibians from all three extant orders produce noxious/toxic defensive skins secretions from specialised granular glands, which may be concentrated in parotid glands, as in Bufo toads and some salamanders, or distributed across the skin either generally, as in Xenopus, or in more specific but diffuse concentrations such as in the dorsolateral plicae of Rana frogs or the inguinal glands of many leptodactylid frogs [6,7]. Biochemical studies on

∗ Corresponding author. Fax: +86 431 8698 5951. E-mail address: [email protected] (H. Gao). 1 These two authors contributed equally to this paper.

amphibian skin secretions have revealed numerous biologically active substances, including alkaloid toxins, biogenic amines, proteins and bioactive peptides, of which the latter are the most abundant and structurally diverse [7]. Skin secretions of the African clawed frog (Xenopus laevis) contain high concentrations of a diverse array of biologically active components, including thyrotropin-releasing hormone, peptide glycine–leucine amide, and myotropic peptides caerulein, xenopsin and levitide (reviewed in [7]). These peptides, some of which are also produced in the gastric mucosa [8], are synthesised in granular glands in the skin and are released in a holocrine manner upon stress or injury as a result of contraction of myocytes surrounding the glands [3]. Development of new families of antibiotics that can overcome the resistance problem has become a very important task. Bacterial resistance to conventional antibiotics has become a major problem worldwide owing to their extensive use. Strains of bacteria already exist that are resistant to all available drugs. The killing mechanism of AMPs is different from that of conventional antibiotics, thus AMPs are considered as attractive substitute and/or additional drugs. This work describes the isolation and structural elucidation of a new AMP from a single sample of skin secretion from X. laevis. The antimicrobial activity of the peptide against a selection of Gram-positive and Gram-negative bacteria is also reported. Its antimicrobial mechanism was investigated by transmission electron microscopy (TEM).

0924-8579/$ – see front matter © 2011 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. doi:10.1016/j.ijantimicag.2011.07.012

F. Hou et al. / International Journal of Antimicrobial Agents 38 (2011) 510–515

511

Adult male and female X. laevis (n = 100; 8–10 cm in length) were purchased from Changchun Market of Aquatic Products (Changchun, China). Frog skin secretions were obtained by gentle transdermal electrical stimulation [9] and were collected by washing the dorsal region with physiological saline, after which the frogs were released. The collected solutions were immediately centrifuged (12 000 × g for 10 min) and the supernatant was further clarified by centrifugation (4000 × g for 30 min at 4 ◦ C) and then lyophilised.

1-hydroxybenzotriazole in N,N -dimethylformamide (DIC/HOBt in DMF) for 60–120 min. Cleavage and final deprotection were conducted with TFA:diisopropylsilane:ethanedithiol (90:5:5, v:v:v) for 90 min at room temperature. Peptides were precipitated with ethyl ether, solubilised in water and lyophilised. Purification was conducted by RP-HPLC on a C18 column (250 mm × 10 mm) with a 5 mL/min flow rate and the absorbance of the column effluent was detected at 215 nm. Elution occurred by using the following solutions: 0.1% (v/v) TFA aqueous solution (A) and 0.08% (v/v) TFA in acetonitrile (B). The percentage of B was varied from 0% to 40% over 40 min. Fractions containing the main product were pooled and lyophilised and the purity and mass of PGLa-H were confirmed by MALDI-TOF/TOF-MS.

2.2. Peptide purification

2.5. Antimicrobial assays

A lyophilised skin secretion sample of X. laevis (100 mg) was dissolved in 5 mL of 0.1 M phosphate buffer (pH 6.0) containing 5 mM ethylene diamine tetra-acetic acid (EDTA). The sample was applied to a Sephadex G-25 Superfine gel filtration column (2.6 cm × 100 cm) (Amersham Biosciences, Uppsala, Sweden) equilibrated with 0.1 M phosphate buffer (pH 6.0). Elution was performed with the same buffer. Absorbance of elute was monitored at 280, 254 and 215 nm. The antimicrobial activity of each fraction was determined as indicated below. Fractions displaying antimicrobial activity were collected, pooled (20 mL) and lyophilised. The protein peak containing antimicrobial activity was dissolved in 10 mL of 0.05% (v/v) aqueous trifluoroacetic acid (TFA), clarified by centrifugation and fractionated by preparative reversephase high-performance liquid chromatography (RP-HPLC) using a C8-300 column (10 mm × 250 mm) (ACE, Aberdeen, UK) equilibrated with 0.05% (v/v) aqueous TFA. The concentration of acetonitrile in the eluting solvent was raised to 100% (v/v) over 80 min using a linear gradient at a flow rate of 2 mL/min. Absorbance of the column effluent was monitored at  = 215 nm and fractions (4 mL) were collected at 2-min intervals. Fractions displaying antimicrobial activity were collected and freezedried. The dried samples were re-dissolved in 0.05% (v/v) aqueous TFA (500 ␮L) and were applied to an analytical C18-300 RP-HPLC column (5 mm × 250 mm) (ACE) equilibrated with 0.05% (v/v) aqueous TFA at a flow rate of 1 mL/min. The concentration of acetonitrile in the eluting solvent was raised to 100% (v/v) over 60 min using a linear gradient at a flow rate of 1 mL/min and the absorbance was monitored at 215 nm. Fractions were collected by peak and were freeze-dried for subsequent structural characterisation.

Antimicrobial activity of individual fractions was monitored by an inhibition zone assay. Briefly, plates were prepared by adding 7 mL of molten Mueller–Hinton agar cooled to 37 ◦ C to Petri dishes containing 2 ␮L of each of a log-phase culture [1 × 106 colonyforming units (CFU)/mL] of Staphylococcus aureus ATCC 25923, Escherichia coli ATCC 25922, Pseudomonas aeruginosa, meticillinresistant S. aureus (MRSA) and Bacillus subtilis. After the agar had solidified, sample wells were created by cutting the agar with a sterile cork borer and then aspirating out the agar plug. Aliquots (100 ␮L) of the appropriate HPLC fraction were dried down, reconstituted in sterile phosphate-buffered saline (PBS) (0.1 M, pH 7.3) and 5 ␮L aliquots were applied to each well and the plates were incubated overnight at 37 ◦ C. For determination of minimum inhibitory concentrations (MICs), PGLa-H was diluted in 0.1 M PBS and then 100 ␮L of each dilution was incubated with the appropriate 100 ␮L of inoculum (104 CFU/mL of S. aureus ATCC 25923, E. coli ATCC 25922, P. aeruginosa, MRSA and B. subtilis) in 96-well microtitre cell culture plates for 18 h at 37 ◦ C in a humidified incubator. MICs were measured as previously described [11] and were taken as the lowest concentration of peptide at which no visible growth was observed. Serial dilutions of benzylpenicillin potassium (96 U/mL, 58.7 ␮g/mL) and gentamicin (96 ␮g/mL) were used as control antibiotics for Gram-positive and Gram-negative bacteria, respectively.

2. Materials and methods 2.1. Tissue collection and extraction

2.3. Structural characterisation

2.6. Haemolytic assay The haemolysis assay was undertaken using rabbit red blood cells in liquid medium as described previously [12]. Serial dilutions of the peptide were used. Following incubation at 37 ◦ C for 30 min, cells were removed by centrifugation and the absorbance in the supernatant was measured at 595 nm. Maximum haemolysis was determined by adding 1% Triton X-100 to a sample of cells. 2.7. Transmission electron microscopy

The primary structure of the peptide was determined by automated Edman degradation using a ABI Procise® model 491 sequencer (Applied Biosystems, Foster City, CA). Mass spectrometric analysis and tandem mass spectrometry (MS/MS) were performed on a matrix-assisted laser desorption/ionisation timeof-flight (MALDI-TOF/TOF) mass spectrometer (Bruker, Bremen, Germany). Three-dimensional (3D) models of PGLa-H and PYLa were constructed using CPHmodels (ExPASy Proteomics Server, Swiss Institute of Bioinformatics). The 3D maps were observed with RASWIN.

Exponential-phase bacteria were treated with PGLa-H (100 ␮g/mL) for 2 h at 37 ◦ C. This concentration was used to observe an effect on a greater percentage of cells. The bacteria were centrifuged at 400 × g for 10 min and microscopy was performed with a JEM-1200EXII microscope (JEOL, Tokyo, Japan) under standard operating conditions following negative staining.

2.4. Peptide synthesis, purification and characterisation

3.1. Purification of antimicrobial peptides

PGLa-H was synthesised by the solid-phase approach on a Rink amide resin (0.61 mmol/g) using the Fmoc/t-butyl strategy [10]. Couplings were performed with N,N 0-diisopropylcarbodiimide/

The supernatant fluid of X. laevis skin secretions was divided into three peaks by Sephadex G-25 gel filtration (Fig. 1A), with the antimicrobial activity occurring in the last peak. The active peak

3. Results

512

F. Hou et al. / International Journal of Antimicrobial Agents 38 (2011) 510–515

Fig. 1. Fractionation of Xenopus laevis skin secretion. (A) Sephadex G-25 gel filtration: X. laevis skin secretion was applied to a Sephadex G-25 Superfine gel filtration column (2.6 cm × 100 cm) (Amersham Biosciences, Uppsala, Sweden) equilibrated with 0.1 M phosphate buffer (pH 6.0). Elution was performed with the same buffer. The fraction with PGLa-H activity is indicated by an arrow (A). The active fraction from Sephadex G-25 gel filtration was further purified on a C8-300 reverse-phase high-performance liquid chromatography (RP-HPLC) column (10 mm × 250 mm) (ACE, Aberdeen, UK) equilibrated with 0.05% (v/v) trifluoroacetic acid (TFA)/water. Elution was performed at a flow rate of 2 mL/min. The fraction with PGLa-H activity is indicated by an arrow (B). The active fraction from C8 RP-HPLC was further purified on a C18-300 RP-HPLC column (5 mm × 250 mm) (ACE) equilibrated with 0.05% (v/v) TFA/water. Elution was performed at a flow rate of 1 mL/min. The fraction with PGLa-H activity is indicated by an arrow (C).

was collected and applied to a preparative C8 HPLC column. Forty peaks were obtained from this separation (Fig. 1B). All of the peaks eluted from the preparative HPLC were collected and tested for antibacterial activity. The active fraction was located in peak 28. Peak 28 was further purified by a C18 RP-HPLC column and nine peaks were obtained from this separation (Fig. 1C). The peak with antibacterial activity (marked by arrow) was collected and applied to MALDI-TOF/TOF-MS analysis.

3.2. Structural characterisation Purified AMP (marked by an arrow in Fig. 1C), named PGLaH, was subjected to amino acid sequence analysis by automated Edman degradation. It is composed of 10 amino acids with an amino acid sequence of KIAKVALKAL (Table 1). By MALDI-TOF/TOFMS analysis, the observed molecular weight was 1053.727 Da (Fig. 2), which matched well with the theoretical molecular

F. Hou et al. / International Journal of Antimicrobial Agents 38 (2011) 510–515

513

Table 1 Primary structural comparison of the new antimicrobial peptide (AMP) from Xenopus laevis (PGLa-H) with orthologous sequences in AMP databases established by BLAST analysis.

Other AMP

Sequence comparison with PGLa-Ha KLAKLAKKLAKLAK

DP1 (synthetic)

KI AKVAL

KAL

GMASKAGAIAGKIAKVALKAL

PYLa

KIAKVALKA

Homology (%) 50 47.61

L

Eumenine mastoparan-AF MP-VB1 Temporin-1Vb

INLLKIAKGI IKSL K I A K V A LK A L INMKASAAVAKKLL K

IAKVALKAL

FLSIIAKV

LGSLF

42.85 42.85 42.85

K I A K V A L KA L a

Conserved residues are highlighted in grey.

weight (1053.454 Da). By GenBank and AMP database BLAST searches, PGLa-H was deemed to be a new AMP. Analysis using the ExPASy MW/pI tool (http://www.expasy.ch/tools/pi tool.html) showed that it had a predicted isoelectric point (pI) of 10.30. 3.3. Antimicrobial activity PGLa-H exhibited antimicrobial activity against the tested strains (Table 2). Of the tested strains, S. aureus ATCC 25923 was the most sensitive to PGLa-H, with a MIC of 8.7 ␮g/mL (Table 2). Interestingly, PGLa-H exhibited antimicrobial activity against E. coli ATCC 25922 (MIC = 23.6 ␮g/mL). The antibiotic activity of PGLa-H was proven to be lethal for the sensitive strains as they were not capable of resuming growth on agar plates after 6 h of treatment with concentrations above the corresponding MICs. PGLa-H had no effect on P. aeruginosa in the current experiments.

Table 2 Antimicrobial activities of PGLa-H. Microorganism

Staphylococcus aureus ATCC 25923 Escherichia coli ATCC 25922 Pseudomonas aeruginosa MRSA Bacillus subtilis

MIC (␮g/mL)a BP

GEN

PGLa-H

4.1b –

– 3.9b

8.7 23.6 N/D 67.8 14.4

MIC, minimum inhibitory concentration; BP, benzylpenicillin potassium; GEN, gentamicin; N/D, not detected; MRSA, meticillin-resistant Staphylococcus aureus. a Minimum peptide concentration required for total inhibition of cell growth in liquid medium. These concentrations represent the mean value of three independent experiments performed in duplicate. b The control antimicrobials used against S. aureus (Gram-positive) and E. coli (Gram-negative) were BP and GEN, respectively.

3.5. Transmission electron microscopy 3.4. Haemolytic activity Some AMPs exhibit haemolytic activities [13]. Rabbit red blood cells were used to check for haemolytic capability. PGLa-H had little haemolytic activity (3.15%) on red blood cells even with peptide concentrations up to 100 ␮g/mL. PGLa-H showed slight haemolytic activity (7.95%) at 200 ␮g/mL.

To study the possible mechanisms of action of the AMP on S. aureus ATCC 25923 and E. coli ATCC 25922, TEM was performed on negative staining of bacteria that had been treated with PGLa-H for 2 h. PGLa-H had direct bacteria killing capability by various mechanisms as illustrated in Fig. 3. Control cells with no peptide treatment exhibited a normal shape and smooth surfaces (Fig. 3A and D). In contrast, the cellular shape and surface of the bacteria treated with

Fig. 2. Matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI-TOF/TOF-MS) of antibacterial peptide PGLa-H.

514

F. Hou et al. / International Journal of Antimicrobial Agents 38 (2011) 510–515

Fig. 3. Ultrastructure of (A–C) Escherichia coli ATCC 25922 and (D–F) Staphylococcus aureus 25923 treated with the tested samples (100 ␮g/mL): (A) E. coli control (no peptide); (B and C) peptide-treated E. coli; (D) S. aureus control (no peptide); (E and F) peptide-treated S. aureus.

Fig. 4. Schematic representation of antimicrobial peptides displayed by RASWIN: (A and C) PYLa and (B and D) PGLa-H. PGLa-H and PYLa both formed an ␣-helix to the C segment structure.

F. Hou et al. / International Journal of Antimicrobial Agents 38 (2011) 510–515

AMPs for 2 h exhibited bores on the cell surface (Fig. 3B). PGLa-H induces damage characterised by rough surfaces containing extracellular debris and outer membranous blebs (Fig. 3C), thickened cell walls and dissolved cytoplasmic material (Fig. 3E and F). The results indicated that PGLa-H could kill cells by destroying the cell membrane. 4. Discussion Amphibians, the first group of organisms forming a connecting link between land and water, are forced to adopt and survive in a variety of conditions laden with pathogens and predators. Therefore, they are endowed with an excellent chemical defence system composed of pharmacological and antimicrobial peptides [14]. AMPs distributed in amphibian skin may perform defensive roles alone or synergistically with other compounds. The diversity of AMPs and their precursors may extend their spectrum of biological activities. To date, no two amphibian species have yet been found that have the same complement of peptide antibiotics [15]. It has been predicted that there may be as many as 100 000 different natural peptides produced by the dermatous glands of more than 5000 amphibian species. Thus, the amphibian skin is a potential valuable source for the discovery of new peptide antibiotics. Furthermore, Conlon et al. [16] suggested that the profiles of the AMPs in frog skin may constitute a characteristic ‘fingerprint’ that is conserved within a species. The study described here has led to the isolation of a new AMP from the skin secretion of X. laevis. Its amino acid sequence was compared for homology with sequences in GenBank using the BLAST suite of programmes, and complete homology with a known peptide was not detected; it therefore represents a new AMP. As shown in Table 1, a comparison of its primary structure reveals that it represents orthologues of previously described dermal peptides (PYLa) isolated from the tetraploid frog X. laevis. Evolutionary pressure to conserve the primary structures of the peptides has not been uniform. It is a cationic antibiotic peptide because of its pI of 10.30, and according to the results of sequencing and 3D prediction PGLa-H was part of the C segment of PYLa. Both had a similar C segment structure, forming an ␣-helix (Fig. 4). In this study, PGLa-H exhibited significant antimicrobial activity both against Gram-positive and Gram-negative reference bacteria. Moreover, it also had antimicrobial activity towards MRSA. PGLa-H had little haemolytic activity on red blood cells. A key event occurring after membrane binding is the process of peptide structural or conformational phase transition, most well documented for ␣-helical AMPs. Numerous studies using various biophysical methodologies show that many AMPs are disordered in aqueous environments, exhibiting extended or random coil conformations in this setting. However, many such peptides rapidly assume a highly structured amphipathic ␣-helical conformation upon interaction with phospholipid bilayers or in

515

membrane-mimetic solvents. Interestingly, a number of peptides require a negatively charged bilayer to undergo this transition. The facts that electrostatic forces are active over relatively long molecular distances and that lysine and arginine interactions with phosphate groups in lipid bilayers are particularly strong likely contribute to the initial attraction and membrane-targeting step of many AMPs. PGLa-H is a cationic peptide that has been isolated from the skin of X. laevis. This peptide is constituted by 10 amino acids and it produces an ␣-helical structure that appears to be responsible for the formation of transmembrane pores, resulting in an alteration of cell membrane permeability that leads to cell death. Funding: This work was supported by grants from the Institute of Military Veterinary Science, Academy of Military Medical Science (Changchun, China). Competing interests: None declared. Ethical approval: Not required. References [1] Hancock REW, Sahl H-G. Antimicrobial and host-defense peptides as new antiinfective therapeutic strategies. Nat Biotechnol 2006;24:1551–7. [2] Tossi A, Sandri L, Giangaspero A. Amphipathic, ␣-helical antimicrobial peptides. Biopolymers 2000;55:4–30. [3] Simmaco M, Mignogna G, Barra D. Antimicrobial peptides from amphibian skin: what do they tell us? Biopolymers 1998;47:435–50. [4] Toke O. Antimicrobial peptides: new candidates in the fight against bacterial infection. Biopolymers 2005;80:717–35. [5] Yeaman MR, Yount NY. Mechanisms of antimicrobial peptide action and resistance. Pharmacol Rev 2003;55:27–55. [6] Erspamer V. Bioactive secretions of the integument. In: Heatwole H, Barthalmus GT, editors. Amphibian biology, vol. 1. The integument. Chipping Norton, UK: Surrey Beatty & Sons; 1994. p. 179–350. [7] Lazarus LH, Attila M. The toad, ugly and venomous, wears yet a precious jewel in his skin. Prog Neurobiol 1993;41:473–507. [8] Moore KS, Bevins CL, Brasseur MM, Tomassini N, Turner K, Eck H, et al. Antimicrobial peptides in the stomach of Xenopus laevis. J Biol Chem 1991;266:19851–7. [9] Tyler MJ, Stone DJ, Bowie JH. A novel method for the release and collection of dermal, glandular secretions from the skin of frogs. J Pharmacol Toxicol Methods 1992;28:199–200. [10] Chan WC, White PD. Fmoc solid phase peptide synthesis: a practical approach. Oxford, UK: Oxford University Press; 2000. [11] Barchiesi F, Colombo AL, McGough DA, Rinaldi MG. Comparative study of broth microdilution and macrodilution techniques for in vitro antifungal susceptibility testing of yeasts by using the National Committee for Clinical Laboratory Standards’ proposed standards. J Clin Microbiol 1994;32:2494–500. [12] Bignami GS. A rapid and sensitive hemolysis neutralization assay for palytoxin. Toxicon 1993;31:817–20. [13] Lai R, Zheng YT, Shen JH, Liu GJ, Liu H, Lee WH, et al. Antimicrobial peptides from skin secretions of Chinese red belly toad Bombina maxima. Peptides 2002;23:427–35. [14] Boman HG. Antibacterial peptides: key components needed in immunity. Cell 1991;65:205–7. [15] Vanhoye D, Bruston F, Nicolas P, Amiche M. Antimicrobial peptides from hylid and ranin frogs originated from a 150-million-year-old ancestral precursor with a conserved signal peptide but a hypermutable antimicrobial domain. Eur J Biochem 2003;270:2068–81. [16] Conlon JM, Kolodziejek J, Nowotny N. Antimicrobial peptides from ranid frogs: taxonomic and phylogenetic markers and a potential source of new therapeutic agents. Biochim Biophys Acta 2004;1696:1–14.