NOTE

Neutralization of bacterial endotoxins by frogantimicrobial peptidesErmin Schadich, Drusilla Mason and Anthony L. Cole

School of Biological Sciences, University of Canterbury, Private Bag 4800, Christchurch, New Zealand

ABSTRACTThe ability of skin antimicrobial peptides of the southern bell frog, Litoria raniformis, to neutralize invitro the endotoxin, proinflammatory lipopolysaccharide (LPS) complex, from two different gram‐

negative bacterial pathogens, human pathogen Escherichia coli (0111:B4) and frog pathogen Klebsiellapneumoniae, was investigated. The LPS neutralization activity of the natural mixture of skin anti-microbial peptides was measured using chromogenic Limulus amebocyte lysate assays. These skinantimicrobial peptides neutralized the LPSs from both pathogens at physiologically relevantconcentrations (IC50 < 100 µg/mL) showing their potential for non‐specific LPS neutralizationin vivo in the skin of infected frogs and for development of anti‐endotoxin agents.

Key words antimicrobial peptides, bacterial lipopolysaccharide, inflammation.

In various vertebrates, neutralization of the endotoxin,LPS complex, of the outer membranes of gram‐negativebacterial pathogens is important for protection againstexcessive inflammation during bacterial infections of avariety of tissues (1, 2). The most significant LPSneutralization mechanism in human blood is the acutephase protein named LBP (3), whereas related antimi-crobial peptides operate in other tissues (4). Both LBPand antimicrobial peptides interfere with the proin-flammatory effects of LPSs on macrophages by bindingto its lipid component and thereby precluding itsinteraction with TLR4 and associated pathways ofproduction of the proinflammatory and chemotacticcytokines interleukin‐1, interleukin‐6 and tumor necro-sis factor (3, 4). They are essential for regulation ofsubsequent neutrophil responses and activation of thecomplement lytic system. Any pathogenic process thatresults in insufficient amounts of these factors isassociated with dysregulation and overrated responsesleading to intensive tissue inflammation and damage andendotoxic shock (2).

Although LPS induces inflammatory responses in ecto-thermic vertebrates like frogs, it has little toxicity in theseanimals (5, 6). Surprisingly, although the sequences ofgenes orthologous to the human LPB have been found ingenomicDNA in two frog species,Xenopus laevis and Siluranatropicalis (7), its protein product has not yet been identifiedeither in blood or liver. This is intriguing as it raises thequestion of how frogs protect themselves from LPS.

One possible LPS neutralization mechanism in frogsis antimicrobial peptides from skin granular glands,stomach and intestinal tissues (8–10). These peptideshave positively charged amino acid sequences that arerequired for interaction with negatively charged compo-nents of microbial membranes and lysis of microbialcells (11). A previous study by Schadich showed thatantimicrobial peptides from skin granular glands ofdifferent frog species have activity against differenthuman and frog bacterial pathogens, an activity whichstrongly correlates with resistance to bacterial disease(12). Moreover, researchers have also demonstrated theability of skin antimicrobial peptides to bind to LPS from

CorrespondenceErmin Schadich, School of Biological Sciences, University of Canterbury, Private Bag 4800, Christchurch 8140, New Zealand. Tel: 64 3 364 2500; fax: 64

3 364 2590; email: ermins@gmail.com

Received 28 September 2012; revised 11 November 2012; accepted 25 November 2012.

List of Abbreviations: E. coli, Escherichia coli; K. pneumoniae, Klebsiella pneumoniae; IC50, values of peptides peptide concentration at which 50%of

LPS is neutralized; LAL, Limulus amebocyte lysate; LBP, lipopolysaccharide‐binding protein; LPS, lipopolysaccharide; L. raniformis, Litoria raniformis ;TLR4, Toll‐like receptor 4.

Microbiol Immunol 2013; 57: 159–161doi: 10.1111/1348-0421.12012

© 2012 The Societies and Wiley Publishing Asia Pty Ltd 159

different human bacterial pathogens (13, 14), and thissuggests their possible role in the LPS neutralizationmechanism.Skin antimicrobial peptides could therefore provide

protection from the toxic effects of endotoxins in frogskin. This study aimed to determine whether skinantimicrobial peptides of the New Zealand introducedspecies, the southern bell frog (L. raniformis) can neutralizeLPS from different bacterial pathogens in vitro. We testedskin antimicrobial peptides from L. raniformis for theirability to neutralize LPS from the human pathogen E. coli(0111:B4) and the frog pathogen K. pneumoniae.We collected a natural mixture of skin antimicrobial

peptides from adult L. raniformis by using norepinephrineinjections and partially purified it by using C18‐Sep‐Pakcartridges (Waters Corporation, Milford, MA, USA) asdescribed by Schadich (12). We confirmed its content ofspecies‐specific aurein peptides by liquid chromatogra-phymass spectrometry analyses (12). Polymixin B sulfate,a reference control peptide with known ability toneutralize endotoxins from gram‐negative bacteria andendotoxin‐free water were purchased from SigmaChemical (St. Louis, MO, USA). We generated peptidedigests to provide a negative control for studies of theactivity of the skin peptides. We incubated the peptidemixtures (1 mg/mL) with pronase E, a protease mixturethat degrades peptides completely (Sigma Chemical), at aconcentration of 0.5 mg/mL in ammonium phosphatebuffer (pH 7.0) at 37°C for 20 hrs. After digestion, weinactivated the protease by heating it at 90°C for 10 min.Using a modified phenol–water technique as previ-

ously described (15), we isolated LPS from overnightcolonies of K. pneumoniae collected from the wild browntree frog, Litoria ewingii in Oxford forest, Zealand, andgrown on blood‐agar plates. We purchased the LPS of E.coli (0111:B4) from Cambrex Bio Science (Walkersville,MD, USA). We tested all solutions to ensure they wereendotoxin free by measuring the concentration of LPSusing chromogenic LAL assays (QCL‐1000 kit, CambrexBio Science). We sterilized all pyrogenic‐free consum-ables by heating them for 3 hrs at 180°C.We assessed neutralization of LPS by skin antimicrobial

peptides by measuring their free concentrations usingLAL assays after incubating them with skin peptides asdescribed by Ried et al. (16). We incubated the peptidesdissolved in endotoxin‐free water at different concen-trations of peptide (0–300 mg/mL) with 150 pg/mL ofbacterial LPS in 50 mL reactions at 37°C for 30 min. Theblank controls included the same concentration withoutthe LPS. Subsequently, we incubated 50 mL of the LALreagent containing pro‐enzyme at 37°C for 10 min in 96‐well microtiter plates. Next, we added 100 mL of the LALsubstrate to each sample and incubated them at 37°C for

an additional 6 min. We stopped the reactions using100 mL of 25% acetic acid and read the absorbance ofeach reaction at 405 nm using a microplate reader. Weused the absorbance values for curve analyses. Weperformed three assays for both tested peptides andcontrols and tested three replicate reactions for eachpeptide concentration. We automatically adjusted allcurves required for estimation of IC50 values of peptides(peptide concentration at which 50% of LPS is neutral-ized) by nonlinear regression using Graph Pad Prism 4.

The peptide mixture from L. raniformis neutralized theLPS of the standard reference strain E. coli (0111:B4) andthe isolate K. pneumoniae in a concentration dose‐dependent manner comparable to the reference controlpolymyxin B (Fig. 1). Their IC50 values for neutralizationof LPS of E. coli (0111:B4) and K. pneumoniae were below

Fig. 1. Neutralization of (a) Escherichia coli (0111:B4) and (b)Klebsiella pneumoniae lipopolysaccharide (LPS) by skin anti-microbial peptides of Litoria raniformis frogs and polymyxin B.

E. Schadich et al.

160 © 2012 The Societies and Wiley Publishing Asia Pty Ltd

100 µg/mL (Table 1). The peptide digests did notneutralize any of two bacterial LPSs.Skin antimicrobial peptides of L. raniformis neutralized

the LPS from bacterial pathogen K. pneumoniae in vitro atphysiologically relevant concentrations (IC50 < 100 mg/mL; Table 1), suggesting that they may be an LPSneutralization mechanism in infected frogs. This activityis not restricted to frog pathogens since the LPS fromhuman pathogen E. coli (0111:B4) was also neutralized(Fig. 1, Table 1). Such non‐specific in vitro activity couldbe an effective, broad and rapid mechanism forneutralization of LPS from different bacterial pathogensin vivo in the skin of infected frogs.The peptide mixture of L. raniformis was not as active as

polymixin B in neutralizing LPS; this may have been dueto dilution of the mixture by inactive peptides. Thus, onedirection of future studies should be also to analyze theeffects of single isolated peptides in order to show theirpotential in LPS neutralization.

ACKNOWLEDGMENTS

We thank Andrew Bagshaw, University of Otago foruseful comments on this manuscript. This study wassupported by a Royal Society of New Zealand MarsdenGrant (M1069).

DISCLOSURE

All authors have no conflict of interest.

REFERENCES1. Chilton P.M., Embry C.A., Mitchell T.C. (2012) Effects of

differences in lipid A structure on TLR4 pro‐inflammatorysignaling and inflammasome activation. Front Immunol 3: 1–7.

2. Karima R., Matsumoto S., Higashi H., Matsushima K. (1999)The molecular pathogenesis of endotoxic shock and organfailure. Mol Med Today 5: 123–33.

3. Shumann R.R., Leong S.R., Flaggs G.W., Gray, P.W., Wright S.D.,Mathison J.C., Tobias P.S., Ulevitch R.J. (1990) Structure andfunction of lipopolysaccharide binding protein. Science 249:14,293–31.

4. Levy O., Ooi C.E., Elsbach P., Doerfler M.E., Lehrer R.I., Weiss J.(1995) Antibacterial proteins of granulocytes differ in interactionwith endotoxin. Comparison of bactericidal/permeability‐increasing protein, p15s, and defensins. J Immunol 154: 5403–10.

5. Bicego K.C., Steiner A.A., Antunes‐Rodrigues J., Branco L.G.(2002) Indomethacin impairs LPS‐induced behavioral fever intoads. J Appl Physiol 93: 512–16.

6. Bugbee T.M., Ruben L.N., Beard M.E., Zettergren L.D. (1983)Antibody production by different sites and cyclophosphamide‐induced immunosuppression of the TNP‐LPS response in thegrass frog, Rana pipiens. Dev Comp Immunol 7: 569–74.

7. Klein S.L., Strausberg R.L., Wagner L., Pontius J., Clifton S.W.,Richardson P. (2002) Genetic and genomic tools for Xenopusresearch: The NIH Xenopus initiative. Dev Dyn 225: 384–91.

8. Rollins‐Smith L.A., Reinert L.K., O'Leary C.J., Houston L.E.,Woodhams D.C. (2005) Antimicrobial peptide defenses inamphibian skin. Integr Comp Biol 45: 137–42.

9. Moore K.S., Bevins C.L., Brasseur M.M., Tomassini N., TurnerK., Eck H., Zasloff M. (1991) Antimicrobial peptides in thestomach of Xenopus laevis. J Biol Chem 266: 19851–57.

10. Reilly D.S., Tomassini N., Bevins C.L, Zasloff M. (1994) A Panethcell analogue in Xenopus small intestine expresses antimicrobialpeptide genes: conservation of an intestinal host‐defense system.J Histochem Cytochem 42: 697–704.

11. Conlon M.J. (2011) Structural diversity and species distributionof host‐defence peptides in frog skin secretions. Cell Mol Life Sci68: 2303–15.

12. Schadich E. (2009) Skin peptide activities against opportunisticbacterial pathogens of the African Clawed Frogs (Xenopus laevis)and three Litoria frogs. J Herpetol 43: 173–83.

13. Matera G., Cook J.A., Geisel J., Ashton S.H., Wise W.C., Focá A.,Berkowitz B.A., Halushka P.V. (1993) Effects of two magaininpeptides on eicosanoid release from rat peritoneal macrophages.Antimicrob Agents Chemother 37: 393–97.

14. Nan Y.H., Jeon Y.J., Park I.S., Shin S.Y. (2008) Antimicrobialpeptide P18 inhibits inflammatory responses by LPS but notby IFN‐gamma‐stimulated macrophages. Biotechnol Lett30: 1183–87.

15. Johnson K.G., Perry M.B. (1976) Improved techniques for thepreparation of bacterial lipopolysaccharides. Can J Microbiol 22:29–34.

16. Ried C., Wahl C., Miethke T., Wellnhofer G., Landgraf C.,Schneider‐Mergener J., Hoess A. (1996) High affinity endotoxin‐binding and neutralizing peptides based on the crystal structureof recombinant Limulus anti‐lipopolysaccharide factor. J BiolChem 271: 28120–27.

Table 1. Lipopolysaccharide neutralization activity of skin antimicro-

bial peptides of Litoria raniformis and polymixin B

LPS

IC50 of peptides

L. raniformis Polymixin B

E. coli (0111:B4) 64.5 � 1.1 2.0 � 1.2

K. pneumoniae 55.8 � 1.2 1.7 � 1.0

© 2012 The Societies and Wiley Publishing Asia Pty Ltd 161

LPS neutralization by frog peptides