International Immunopharmacology 6 (2006) 1748–1753www.elsevier.com/locate/intimp

Preliminary Report

Cytotoxicity and antigenicity of antimicrobial synthesized peptidesderived from the beetle Allomyrina dichotoma defensin in mice

Yukari Koyama a,c, Maki Motobu a, Kenji Hikosaka a,d, Manabu Yamada a,Kikuyasu Nakamura a, Hisako Saido-Sakanaka b, Ai Asaoka b, Minoru Yamakawa b,d,

Takashi Isobe a, Kameo Shimura a, Chung-Boo Kang e, Hideki Hayashidani f,Yutaka Nakai c, Yoshikazu Hirota f,⁎

a National Institute of Animal Health, Tsukuba, Ibaraki, Japanb National Institute of Agrobiological Sciences, Tsukuba, Ibaraki, Japan

c Faculty of Agriculture, Tohoku Univrsity, Sendai, Miyagi, Japand Institute of Agriculture and Forestry, University of Tsukuba, Tsukuba, Ibaraki, Japan

e College of Veterinary Medicine, Gyeongsang National University, Jinju, Koreaf Laboratory of Veterinary Hygiene, Department of Veterinary Medicine, Faculty of Agriculture, Tokyo, University of Agriculture and Technology,

3-5-8, Saiwai-cho, Fuchu, Tokyo 183-8509, Japan

Received 23 December 2005; received in revised form 27 February 2006; accepted 20 June 2006

Abstract

Synthetic peptides, peptides A (Arg-Leu-Tyr-Leu-Arg-Ile-Gly-Arg-Arg-NH2) and B (Arg-Leu-Arg-Leu-Arg-Ile-Gly-Arg-Arg-NH2), derived from the beetle Allomyrina dichotoma defensin, have antimicrobial activities. Immunotoxicological effect of thesepeptides was evaluated by cytotoxicity of mouse peritoneal macrophages. In addition, antigenicity of these peptides was studied byevaluating antibody responses in mice immunized with these peptides. The toxicity of peptide A toward mouse peritoneal cells was lessthan that of polymyxin B, whenmorphologically evaluated in a cytotoxicity test. Almost all of mice injected intraperitoneally (i.p.) witheither peptideA or B at 50–150mg/kg survived, whereas all mice injected i.p. with polymyxinB at the doses ofmore than 25mg/kg diedwithin 24 h. Interestingly, almost all of mice injected intravenously with these peptides at the doses of 10 and 25 mg/kg also survived.Furthermore, mice immunized with these peptides conjugated with keyhole limpet hemocyanin (KLH) showed little or negligible anti-peptideA or B antibody production, although anti-KLH antibodywas significantly produced. The results indicated that peptides A andBwere less cytotoxic than polymyxin B and also had poor antigenicity to produce specific antibody in mice.© 2006 Elsevier B.V. All rights reserved.

Keywords: Defensin; Antimicrobial peptides; Cytotoxicity; Polymyxin B; Specific antibody generation

⁎ Corresponding author. Tel.: +81 42 367 5776; fax: +81 42 367 5776.E-mail address: hirotayh@cc.tuat.ac.jp (Y. Hirota).

1567-5769/$ - see front matter © 2006 Elsevier B.V. All rights reserved.doi:10.1016/j.intimp.2006.06.007

1. Introduction

The emergence of antibiotic-resistant pathogenicbacteria increases the demands on the development ofnovel antibacterial molecules. Antimicrobial peptideshave been found in a wide array of species from insects

Table 1Comparison of survival rate in mice injected with peptide A, peptide Band polymyxin B

Group Injectionof drug

Dose(mg/kg)

Injectionroute

No. of alive mice/No. of total mice

Survivalrate (%)

1 Untreatedcontrol

0 i.p. 3/3 100

2 Untreatedcontrol

0 i.v. 3/3 100

3 PolymyxinB

6.25 i.p. 3/3 100

4 PolymyxinB

12.5 i.p. 3/3 100

5 PolymyxinB

25 i.p. 0/3 0

6 PolymyxinB

50 i.p. 0/3 0

7 Peptide A 50 i.p. 3/3 1008 Peptide A 100 i.p. 3/3 1009 Peptide A 150 i.p. 3/3 10010 Peptide B 50 i.p. 3/3 10011 Peptide B 100 i.p. 3/3 10012 Peptide B 150 i.p. 3/3 10013 Peptide A 10 i.v. 3/3 10014 Peptide A 25 i.v. 3/3 10015 Peptide A 50 i.v. 0/3 016 Peptide B 10 i.v. 3/3 10017 Peptide B 25 i.v. 3/3 10018 Peptide B 50 i.v. 0/3 0

Mice (BALB/c, female, 5weeks old)were injected intraperitoneally(i.p.) orintravenously (i.v.) with various agents or saline. Survival ratesweredetermined at 2 weeks after injection.

1749Y. Koyama et al. / International Immunopharmacology 6 (2006) 1748–1753

to mammals [1–3]. They are evolutionarily ancientcomponents that can kill invading microorganisms bydisrupting their membranes via a peptide–lipid interac-tion, which is a different mechanism with antibiotics [4].Some peptides are shown to be effective againstantibiotic-resistant bacteria [5–7].

We demonstrated that synthetic cationic antimicro-bial peptides, based on the amino-acid sequence of thebeetle Allomyrina dichotoma defensin, were effectiveagainst both Gram-positive and Gram-negative bacteriaincluding antibiotic-resistant pathogenic bacteria [8].Especially, both of Arg-Leu-Tyr-Leu-Arg-Ile-Gly-Arg-Arg-NH2 (peptide A) and Arg-Leu-Arg-Leu-Arg-Ile-Gly-Arg-Arg-NH2 (peptide B) indicated strong antimi-crobial activity and that peptide A had also a protectiveeffect on lipopolysaccharide (LPS)-induced mortality inan endotoxin shock mouse model [9] with less hemolyticand cytotoxic activities [8,10]. These results suggest thatour synthesized peptides have their potential astherapeutic agents against bacterial infections. It is ofgreat importance to evaluate immunotoxicological safe-ty and antigenicity of these peptides for clinicalpractice. Therefore, the purpose of the present study isto evaluate cytotoxicity of these synthesized peptides inmice macrophages and antigenicity by determiningantibody titers in mice immunized with these peptides.

2. Materials and methods

2.1. Peptides

The two oligopeptides used in the present study, Arg-Leu-Tyr-Leu-Arg-Ile-Gly-Arg-Arg-NH2 (peptide A) and Arg-Leu-Arg-Leu-Arg-Ile-Gly-Arg-Arg-NH2 (peptide B), were designedand synthesized using the methods described in the previousstudy [8–10]. Polymyxin B (Wako Pure Chemical, Osaka,Japan) was used as a control in a cytotoxicity test. The finalpreparation was filtered through a sterile 0.22 μm filter(Millipore Japan, Tokyo, Japan) before intravenous injection.

2.2. Animals

BALB/c female mice, 5 weeks of age, were purchased fromJapan SLC (Hamamatsu, Japan). All mice were kept in an air-conditioned room and fed standard laboratory food pellets andwater ad libitum. This study was done according to the Guide-lines for Animal Experiments of National Institute of AnimalHealth, Japan.

2.3. Cytotoxicity assays

BALB/c mice were injected intraperitoneally (i.p.) with2 ml of 4% Brewer thioglycolate medium (Difco Laborato-ries, Detroit, MI, USA). Peritoneal cells were collected at

3 days after injection, and then washed with phosphatebuffered saline (PBS). The cells in RPMI1640 medium(Sigma, St. Louis, MO, USA) supplemented with 100 μg/mlstreptomycin (Meiji Seika Co., Ltd., Tokyo, Japan) and100 U/ml penicillin (Banyu Pharmaceutical Co., Ltd., Tokyo,Japan) were seeded at 2×105 cells/well into 96-well cultureplates. After incubation at 37 °C in 5% CO2 for 2 h, non-adherent cells were removed by washing twice with PBS andthe remaining adherent cells were collected [11]. Cytologicalexamination of adherent cells on a Giemsa-stained smearshowed that almost all of them were morphologically mac-rophages. Peritoneal mouse macrophages cultured withpeptide A (500 μg/ml) and polymyxin B (500 μg/ml) for24 h were stained with Giemsa. Morphological figures of thecultured cells were observed by a microscope.

2.4. Survival analysis

BALB/c mice were divided into 18 groups, as shown inTable 1. Group 1 was a negative control group injected i.p.with saline. Group 2 was a negative control group injectedintravenously (i.v.) with saline. Groups 3–6 were injected i.p.with polymyxin B (Wako Pure Chemical, Osaka, Japan) at thedoses of 6.25, 12.5, 25 and 50 mg/kg, respectively. Groups 7–12 were injected i.p. with peptides A and B at the doses of 50,

Fig. 1. Morphological evaluation of cytotoxicity in mouse peritoneal macrophages cultured with peptide A and polymyxin B. Mouse peritonealmacrophages were incubated for 24 h at 37 °C in the absence (control) or presence of peptide A (Pep A, 500 μg/ml) and polymyxin B (Polymyxin B,500 μg/ml). The smears were air-dried, fixed with methanol and then stained with Giemsa (×200).

1750 Y. Koyama et al. / International Immunopharmacology 6 (2006) 1748–1753

100 and 150 mg/kg, respectively. Groups 13–18 were injectedi.v. with peptides A and B at the doses of 10, 25 and 50 mg/kg,respectively.

2.5. Determination of antibody titers

Peptides conjugated with keyhole limpet hemocyanin(KLH) were generated by Imject™ carboxyl reactive antibodyproduction kit according to manufacturer's instruction (Pierce,Rockford, IL, USA). Briefly, 1 mg of each peptide was con-jugated to 1 mg of KLH using 1-ethyl-3-(3-dimethylamino-propyl)-carbodiimide hydrochloride by allowing the mixtureto stand at room temperature for 2 h. The peptide–KLH con-jugate was purified by D-Salt™ dextran desalting columns.Mice were immunized six times subcutaneously (s.c.) with themixture of peptide–KLH (10 or 50 μg /mouse) and complete(first injection) or incomplete (second and later injection)

Fig. 2. Detection of specific antibody production against each peptides A andof peptide A (●) or B (○) at 2-week interval of 13 weeks (n=5). Sera colleELISA plates coated with peptide A, peptide B or KLH.

Freund's adjuvant at 2-week interval of 13 weeks. Sera werecollected at 7 days after the sixth immunization.

The 96-well microtiter plates (Sumilon, Sumitomo BakeliteCo., Ltd., Tokyo, Japan) were coated with peptide A, peptide B(20 μg/ml) or KLH (10 μg/ml) in 50 mM sodium bicarbonate(pH 9.6) at 4 °C overnight. The wells were washed with PBScontaining 0.05% Tween 20 (PBST), and then blocked withPBS containing 1% (w/v) bovine serum albumin (BSA) at37 °C for 2 h. Serum samples diluted with PBST containing0.5% BSA were added, and then incubated for 1 h at 37 °C.Following washing with PBST, horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG antibody (diluted 2000-fold inPBSTcontaining 0.5% BSA) was incubated in the wells for 1 hat room temperature. The substrate solution containing 0.4 mg/ml of o-phenylenediamine and 0.02% H2O2 in 0.1 M citratephosphate buffer (pH 5.0) was added, and then incubated untilsufficient color had developed. The reaction was stopped by

B (A) or KLH (B). BALB/c mice were immunized six times with 10 μgcted at 7 days after the sixth immunization were determined by using

1751Y. Koyama et al. / International Immunopharmacology 6 (2006) 1748–1753

adding 2 N H2SO4, and the absorbance at 492 nm wasmeasured by a microtiter plate reader.

3. Results

3.1. Comparison of cytotoxic activity of synthesized peptidesand polymyxin B in mouse peritoneal macrophages

The results of cytotoxicity of mouse peritoneal macro-phages cultured in the presence of peptide A (500 μg/ml) andpolymyxin B (500 μg/ml) are shown in Fig. 1. Morphologicalfigures of macrophages cultured with peptide A (500 μg/ml)remained almost intact, but those of macrophages culturedwith polymyxin B were completely destroyed.

3.2. Comparison of toxic activity of peptides and polymyxin Bin mice

Almost all of mice died within 24 h after injected i.p. withpolymyxin B at the doses of 25 and 50 mg/kg (groups 5 and 6),as shown in Table 1. In contrast, all mice treated with peptidesA and B at the doses of 50, 100 and 150 mg/kg survived for2 weeks until they were slaughtered (groups 7–12). Similarly,all mice treated i.v. with peptides A and B at the doses of 10and 25 mg/kg survived for 2 weeks (groups 13, 14, 16 and 17),however, 50 mg/kg group of mice died (groups 15 and 18).

3.3. Antibody production in mice immunized with peptide–KLH

Specific antibody titers in mice immunized with peptide A–or B–KLH conjugate are shown in Figs. 2 and 3, when

Fig. 3. Detection of specific antibody production against peptide B (A) or KLwith 50 μg of peptide B without adjuvant (●), or subcutaneously (s.c.) with13 weeks (n=5). Sera collected at 7 days after the sixth immunization were dKLH.

determined at 13 weeks later. Almost no specific antibodyagainst peptideA or Bwas produced, although significantly hightiters of anti-KLH antibody were observed. Furthermore,repeated injections of peptide B without KLH conjugation orpeptide B–KLH at a higher concentration (total dose, 300 μg)also did not produce anti-peptide B antibody (Fig. 3). Similarly,repeated injection of peptide A (total dose, 300 μg) did not resultin the production of anti-peptide A antibody (data not shown).

4. Discussion

Polymyxin B is well known as an agent that neu-tralizes LPS. It is, however, very limited because of itsrelatively high toxicity and also difficult to i.v. admin-ister polymyxin B because of its toxic effects on thekidneys and central nervous system [12]. This drug wasalso reported to be cytotoxic to K562 cells in culture[13]. In this study, we demonstrated that the toxicity ofpeptide A toward mouse peritoneal cells was less thanthat of polymyxin B, when morphologically evaluatedby microscopy as shown in Fig. 1 and release of lacticacid dehydrogenase (data not shown).

We demonstrated that the survival rates in mice withendotoxin shock were improved by using this peptide A atthe doses of 100 and 150 mg/kg [9]. Although an i.p.injection of peptide A at a dose of 50 mg/kg resulted insurvival as shown in group 7 in Table 1, mice of group15 injected i.v. at the same dose of peptide A died. Ourpreliminary study demonstrated that the concentration of

H (B). BALB/c mice were immunized six times intraperitoneally (i.p.)50 μg of peptide B (□) or peptide B-KLH (▴) at 2-week interval ofetermined by using ELISA plates coated with peptide A, peptide B or

1752 Y. Koyama et al. / International Immunopharmacology 6 (2006) 1748–1753

polymyxin B to agglutinate mice plasma was more than325 μg/ml, when 50 μl of mice plasma was mixed with50μl of two-fold serial diluted polymyxin B in wells of U-bottom microtiter plates and then incubated at 37 °C for30min. On the other hand, that of peptide A to agglutinatemice plasma was more than 2.5 mg/ml, and any agglu-tination reaction was not observed at the concentration ofless than 1.25mg/ml. Interestingly, the cardiac arrest in themice treated with polymyxin B was more frequent thanthat in the mice treated with peptide A. Therefore, it mightbe considered that a cause of death of mice injected i.v.with peptides was due to cardiac arrest associated with thehaemagglutination.

Vancomycin, one of peptide antibiotics, must be ad-ministered in a dilute solution slowly, over at least 60min,due to the high incidence of pain and thrombophlebitis[14]. Furthermore, the result in the present study showedthat an i.v. injection with peptide A at a dose of 10 mg/kgimmediately after challenge with LPS and D-galactos-amine resulted in a significantly improved survival rate,suggesting a possible clinical application of this synthe-sized peptides.

In general, low molecular weight antigens with lowantigenicity require coupling to a protein carrier mole-cule containing many epitopes that stimulate T-helpercells which help B cell responses, in order to elicit aspecific anti-peptide antibody response [15]. The mole-cular weights of peptides A and B are approximately1200 Da, which supposed to be too small to generatespecific antibody without an immunogenic protein car-rier. In the present study, anti-peptide A or B antibodywas not produced significantly even though the peptidesconjugated with KLH were immunized. Since it washypothesized that the localization and orientation of thepeptide to the carrier protein were influenced by thecoupling method [16], the synthesized peptides conju-gated with KLH by an alternative coupling method, m-maleimidobenzoyl-N-hydroxysuccinimide ester, werealso examined in the same manner. However, no specificantibody generation was observed (data not shown).Therefore, these results suggest that frequently repeatedinjections of peptide A or B do not induce specific anti-peptide antibody due to their poor antigenicity.

The previous studies demonstrated that peptides A andB had (1) broad spectral anti-bacterial activity, (2) nohemolytic and cytotoxic activity, (3) no growth inhibitionof murine fibroblast cells [8,10], (4) protective effects onLPS-induced endotoxin shock [9] and (5) protectiveeffects against Escherichia coli and methicillin-resistantStaphylococcus aureus infection in mice [10,17,18]. It isof great importance to make an appropriate choice of drugwith toxicopharmacologically less side and reverse ef-

fects. In general, repeated or frequent treatment withantibiotics is involved in chemotheraphy of intractablebacterial infections. Therefore, it is essential to fulfill therequirements for securing safety in chemotherapy withantibiotics. Repeated immunization of mice with peptideA (total dose of 300 μg/mice) conjugated KLH resulted inlittle antibody response to peptide A, suggesting highsafety in the aspect of immunopharmacologically medi-ated cytotoxicity.

Taken together, synthesized antimicrobial peptides de-rived from A. dichotoma defensin have protective effectsagainst bacterial infection and low possibility of side andreverse effects accompanied with their cytotoxicity andantigenicity, suggesting that these peptides would be oneof candidates for therapeutic use. Further studies arewarranted to apply these peptides in clinical treatments ofinfectious diseases.

Acknowledgement

This work was supported by a Grant-in-Aid (Promo-tion of Basic Research Activities for Innovative Bios-ciences) from the Bio-oriented Technology ResearchAdvancement Institution, Tokyo, Japan.

References

[1] Cociancich S, Bulet P, Hetru C, Hoffmann JA. The inducibleantibacterial peptides of insects. Parasitol Today 1994;10: 132–9.

[2] Hancock RE, Lehrer R. Cationic peptides: a new source ofantibiotics. Trends Biotechnol 1998;16: 82–8.

[3] Lehrer RI, Ganz T, Selsted ME. Defensins: Endogenous anti-biotic peptides of animal cells. Cell 1991;64: 229–30.

[4] Boman HG. Peptide antibiotics and their role in innate immunity.Annu Rev Immunol 1995;13: 61–92.

[5] Miyanoshita A, Hara S, Sugiyama M, Asaoka A, Taniai K,Yukuhiro F, et al. Isolation and characterization of a newmember ofthe insect defensin family from a beetle, Allomyrina dichotoma.Biochem Biophys Res Commun 1996;220: 526–31.

[6] Moon HJ, Lee SY, Kurata S, Natori S, Lee BL. Purification andmolecular cloning of cDNA for an inducible antibacterial proteinfrom larvae of the coleopteran, Tenebrio molitor. J Biochem(Tokyo) 1994;116: 53–8.

[7] Yamada K, Natori S. Purification, sequence and antibacterialactivity of two novel sapecin homologues from Sarcophaga em-bryonic cells: similarity of sapecin B to charybdotoxin. Biochem J1993;291: 275–9.

[8] Saido-Sakanaka H, Ishibashi J, Momotani E, Amano F,Yamakawa M. In vitro and in vivo activity of antimicrobialpeptides synthesized based on the insect defensin. Peptides 2004;25:19–27.

[9] Koyama Y, Motobu M, Hikosaka K, Yamada M, Nakamura K,Saido-Sakanaka H, et al. Protective effects of antimicrobialpeptides derived from the beetle Allomyrina dichotoma defensinon endotoxic shock in mice. Int Immunopharmacol 2006;6:234–40.

1753Y. Koyama et al. / International Immunopharmacology 6 (2006) 1748–1753

[10] Saido-Sakanaka H, Ishibashi J, Sagisaka A, Momotani E,Yamakawa M. Synthesis and characterization of bactericidaloligopeptides designed on the basis of an insect anti-bacterialpeptide. Biochem J 1999;338: 29–33.

[11] Motobu M, Amer S, Yamada M, Nakamura K, Saido-SakanakaH, Asaoka A, et al. Effects of antimicrobial peptides derived fromthe beetle Allomyrina dichotoma defensin on mouse peritonealmacrophages stimulated with lipopolysaccharide. J Vet Med Sci2004;66: 319–22.

[12] Tsubery H, Ofek I, Cohen S, Fridkin M. Structure–functionstudies of polymyxin B nonapeptide: implications to sensitizationof gram-negative bacteria. J Med Chem 2000;43: 3085–92.

[13] Duwe AK, Rupar CA, Horsman GB, Vas SI. In vitro cytotoxicityand antibiotic activity of polymyxin B nonapeptide. AntimicrobAgents Chemother 1986;30: 340–1.

[14] Matzke GR, Zhanel GG, Guay DR. Clinical pharmacokinetics ofvancomycin. Clin Pharmacokinet 1986;11: 257–82.

[15] De Weck AL. Low molecular weight antigen. In: Sela M, editor.The antigens, vol. 2. New York: Academic Press; 1974. chapter 3.

[16] Deen C, Claassen E, Gerritse K, Zegers ND, Boersma WJ. Anovel carbodiimide coupling method for synthetic peptidesenhanced anti-peptide antibody responses. J Immunol Methods1990;129: 119–25.

[17] Yamada M, Nakamura K, Saido-Sakanaka H, Asaoka A,Yamakawa M, Sameshima T, et al. Effect of modifiedoligopeptides from the beetle Allomyrina dichotoma on Escher-ichia coli infection in mice. J Vet Med Sci 2004;66: 137–42.

[18] YamadaM, Nakamura K, Saido-Sakanaka H, Asaoka A, YamakawaM, Yamamoto Y, et al. Therapeutic effect of modified oligopeptidesfrom the beetle Allomyrina dichotoma on methicillin-resistant Sta-phylococcus aureus (MRSA) infection in mice. J Vet Med Sci2005;67: 1005–11.