Title: Dissection of Antibacterial and Toxic Activity of Melittin
Abstract: Melittin, a naturally occurring antimicrobial peptide, exhibits strong lytic activity against both eukaryotic and prokaryotic cells. Despite a tremendous amount of work done, very little is known about the amino acid sequence, which regulates its toxic activity. With the goal of understanding the basis of toxic activity and poor cell selectivity in melittin, a leucine zipper motif has been identified. To evaluate the possible structural and functional roles of this motif, melittin and its two analogs, after substituting the heptadic leucine by alanine, were synthesized and characterized. Functional studies indicated that alanine substitution in the leucine zipper motif resulted in a drastic reduction of the hemolytic activity of melittin. However, interestingly, both the designed analogs exhibited antibacterial activity comparable to melittin. Mutations caused a significant decrease in the membrane permeability of melittin in zwitterionic but not in negatively charged lipid vesicles. Although both the analogs exhibited similar secondary structures in the presence of negatively charged lipid vesicles as melittin, they failed to adopt a significant helical structure in the presence of zwitterionic lipid vesicles. Results suggest that the substitution of heptadic leucine by alanine impaired the assembly of melittin in an aqueous environment and its localization only in zwitterionic but not in negatively charged membrane. Altogether, the results suggest the identification of a structural element in melittin, which probably plays a prominent role in regulating its toxicity but not antibacterial activity. The results indicate that cell selectivity in some antimicrobial peptides can probably be introduced by modulating their assembly in an aqueous environment. Melittin, a naturally occurring antimicrobial peptide, exhibits strong lytic activity against both eukaryotic and prokaryotic cells. Despite a tremendous amount of work done, very little is known about the amino acid sequence, which regulates its toxic activity. With the goal of understanding the basis of toxic activity and poor cell selectivity in melittin, a leucine zipper motif has been identified. To evaluate the possible structural and functional roles of this motif, melittin and its two analogs, after substituting the heptadic leucine by alanine, were synthesized and characterized. Functional studies indicated that alanine substitution in the leucine zipper motif resulted in a drastic reduction of the hemolytic activity of melittin. However, interestingly, both the designed analogs exhibited antibacterial activity comparable to melittin. Mutations caused a significant decrease in the membrane permeability of melittin in zwitterionic but not in negatively charged lipid vesicles. Although both the analogs exhibited similar secondary structures in the presence of negatively charged lipid vesicles as melittin, they failed to adopt a significant helical structure in the presence of zwitterionic lipid vesicles. Results suggest that the substitution of heptadic leucine by alanine impaired the assembly of melittin in an aqueous environment and its localization only in zwitterionic but not in negatively charged membrane. Altogether, the results suggest the identification of a structural element in melittin, which probably plays a prominent role in regulating its toxicity but not antibacterial activity. The results indicate that cell selectivity in some antimicrobial peptides can probably be introduced by modulating their assembly in an aqueous environment. Melittin, the major component of European bee venom from Apis mellifera, is a well studied peptide, which is known for its strong cytolytic and antimicrobial activities (1Habermann E. Science. 1972; 177: 314-322Crossref PubMed Scopus (1313) Google Scholar, 2Dempsey C.E. Biochim. Biophys. Acta. 1990; 1031: 143-161Crossref PubMed Scopus (784) Google Scholar). It belongs to the family of cytolytic peptides whose members are believed to be the key components of defense and offense mechanisms of all organisms (3Zasloff M. Nature. 2002; 415: 389-395Crossref PubMed Scopus (6648) Google Scholar, 4Boman H.G. Annu. Rev. Immunol. 1995; 13: 61-92Crossref PubMed Scopus (1496) Google Scholar).A considerable number of structure-function studies have been carried out to understand the molecular mechanism of the hemolytic and antimicrobial activities of melittin. The replacement of the first 20 amino acids by another helix-forming sequence did not alter the antimicrobial and hemolytic activity of melittin (5DeGrado W.F. Kezdy F.J. Kaiser E.T. J. Am. Chem. Soc. 1981; 103: 679-681Crossref Scopus (146) Google Scholar). However, the deletion of Leu-6, Lys-7, Val-8, Leu-9, Leu-13, Leu-16, Ile-17, Trp-19, and Ile-20 dramatically reduced the hemolytic activity of melittin and to a lesser extent the antimicrobial activity also (6Blondelle S.E. Houghten R.A. Biochemistry. 1991; 30: 4671-4678Crossref PubMed Scopus (243) Google Scholar). The crucial roles of both termini of melittin in its functional properties have been demonstrated by the design of several analogs with deletion of N-(7Habermann E. Kowallek H. Hoppe-Seyler's Z. Physiol. Chem. 1970; 351: 884-890Crossref PubMed Scopus (73) Google Scholar, 8Gevod V.S. Birdi K.S. Biophys. J. 1984; 45: 1079-1083Abstract Full Text PDF PubMed Scopus (50) Google Scholar) and C-terminal (9Schroder E. Lubke K. Lehmann M. Beetz I. Experientia (Basel). 1971; 27: 764-765Crossref PubMed Scopus (69) Google Scholar) sequences.Several efforts have been made to design analogs of melittin with decreased hemolytic but similar antimicrobial activity. For example, the diastereoisomers of melittin having d-amino acid analogs (10Oren Z. Shai Y. Biochemistry. 1997; 36: 1826-1835Crossref PubMed Scopus (373) Google Scholar) at a few positions, a hybrid peptide consisting of cecropin A and melittin sequences (11Boman H.G. Wade D. Bowman I.A. Wahlin B. Merrifield R.B. FEBS Lett. 1989; 259: 103-106Crossref PubMed Scopus (257) Google Scholar), and an analog designed from the C-terminal 15-residue fragment (12Subbalakshmi C Nagaraj R Sitaram N. FEBS Lett. 1999; 448: 62-66Crossref PubMed Scopus (72) Google Scholar) exhibited similar antibacterial activity to melittin but much less hemolytic activity. The crystal structure of melittin has been solved, which indicated a tetrameric helix-bend-helix structure (13Terwilliger T.C. Eisenberg D. J. Biol. Chem. 1982; 257: 6016-6022Abstract Full Text PDF PubMed Google Scholar) with two helical segments, positioned between amino acid residues 1–10 and 13–26 and the bend region within residues 11–12.Biophysical properties of melittin such as self-association and phospholipid membrane-interaction have been studied extensively to understand the mode of action of melittin (14Beschiaschvili G. Seelig J. Biochemistry. 1990; 29: 52-58Crossref PubMed Scopus (294) Google Scholar, 15Perez-Paya E. Houghten R.A. Blondelle S.E. J. Biol. Chem. 1995; 270: 1048-1056Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar, 16DeGrado W.F. Musso G.F. Lieber M. Kaiser E.T. Kezdy F.J. Biophys. J. 1982; 37: 329-338Abstract Full Text PDF PubMed Scopus (199) Google Scholar, 17Juvvadi P. Vunnam S. Merrifield R.B. J. Am. Chem. Soc. 1996; 118: 8990-8997Crossref Scopus (104) Google Scholar). Although in water melittin exists as monomer with mostly random coil conformation, an increase in peptide concentration or the addition of salt results in the transformation of monomer to tetramer with a pronounced helical structure (18Talbot J.C. Dufourcq J. de Bony J. Faucon J.F. Lussan C. FEBS Letters. 1979; 102: 191-193Crossref PubMed Scopus (170) Google Scholar, 19Tatham A.S. Hider R.C. Drake A.F. Biochem. J. 1983; 211: 683-686Crossref PubMed Scopus (31) Google Scholar, 20Brown L.R. Lauterwein J. Wuthrich K. Biochim. Biophys. Acta. 1980; 622: 231-244Crossref PubMed Scopus (144) Google Scholar). Melittin has been used as a model for studying the general features of lipid interaction of membrane proteins. Membrane-interaction studies suggest that melittin forms transmembrane pores in zwitterionic lipid bilayers by a barrel-stave mechanism (21Vogel H. Jahnig F. Biophys. J. 1986; 50: 573-582Abstract Full Text PDF PubMed Scopus (261) Google Scholar, 22Rex S. Schwarz G. Biochemistry. 1998; 37: 2336-2345Crossref PubMed Scopus (122) Google Scholar, 23Ladokhin A.S. Selested M.E. White S.H. Biophys. J. 1997; 72: 1762-1764Abstract Full Text PDF PubMed Scopus (189) Google Scholar, 24Papo N. Shai Y. Biochemistry. 2003; 42: 458-466Crossref PubMed Scopus (211) Google Scholar) and in negatively charged lipid vesicles it acts like a detergent through a carpet mechanism (24Papo N. Shai Y. Biochemistry. 2003; 42: 458-466Crossref PubMed Scopus (211) Google Scholar, 25Ladokhin A.S. White S.H. Biochim. Biophys. Acta. 2001; 1514: 253-260Crossref PubMed Scopus (218) Google Scholar).Despite all of the work done on melittin, it is not yet clear (i) why melittin has very low cell selectivity, i.e. why it exerts both high antibacterial and toxic activities, and (ii) is the same amino acid sequence responsible for both the hemolytic and antibacterial activities. Toward the identification of important structural and functional elements in melittin and to gain insight about the regulation of cell selectivity in an antimicrobial peptide in general, we have found a leucine zipper motif, located within the residues 6–20, which has not been reported earlier to the best of our knowledge. Because this motif is known to play important roles in the assembly of DNA-binding proteins (26Lansschulz W.H. Johnson P.F. McKnight S.L. Science. 1988; 240: 1759-1764Crossref PubMed Scopus (2521) Google Scholar, 27Kouzarides T. Ziff E. Nature. 1988; 336: 646-651Crossref PubMed Scopus (551) Google Scholar) or membrane-associated viral fusion proteins (28Dubay J.W. Roberts S.J. Broody D. Hunter E. J. Virol. 1992; 66: 4748-4756Crossref PubMed Google Scholar, 29Shugars D.C. Wild C.T. Greenwell T.K. Matthews T.J. J. Virol. 1996; 70: 2982-2991Crossref PubMed Google Scholar, 30Ghosh J.K. Ovadia M. Shai Y. Biochemistry. 1997; 36: 15451-15462Crossref PubMed Scopus (48) Google Scholar), it was of interest to look into the possible roles of this motif in the structural integrity and functional activity of melittin. For this purpose, two analogs of melittin were designed by selectively replacing heptadic leucine with single and double alanine and then synthesized. Hemolytic activity of single (MM-1) and double alanine mutants (MM-2) against human red blood cells was significantly less than that of melittin. Interestingly, MM-1 and MM-2 exhibited antibacterial activity comparable to melittin. Both of the designed analogs displayed contrasting membrane permeability in zwitterionic and negatively charged lipid vesicles. The results have been discussed in terms of the role of this motif in determining the hemolytic and antimicrobial activity of melittin.EXPERIMENTAL PROCEDURESMaterials—Rink amide 4-methylbenzhydrylamine resin (loading capacity, 0.63 mmol/g) and all the N-α Fmoc 1The abbreviations used are: Fmoc, N-(9-fluorenyl)methoxycarbonyl; PC, phosphatidylcholine; PG, phosphatidylglycerol; PE, phosphatidylethanolamine; Chol, cholesterol; HPLC, high performance liquid chromatography; PBS, phosphate-buffered saline; NBD, 4-fluoro-7-nitrobenz-2-oxa-1,3-diazole; RBC, red blood cell; hRBC, human RBC. and side-chain protected amino acids were purchased from Novabiochem, Switzerland. Coupling reagents for peptide synthesis such as 1-hydroxybenzotriazole, N,N′-diisopropylcarbodiimide, 1,1,3,3-tetramethyluronium tetrafluoroborate, and N,N′-diisopropylethylamine were purchased from Sigma. Dichloromethane, N,N′-dimethylformamide, and piperidine were of standard grades and procured from reputed local companies. Acetonitrile (HPLC grade) was procured from Merck, India, and trifluoroacetic acid was purchased from Sigma. Egg phosphatidylcholine (PC), egg phosphatidylglycerol (PG), and dimyristoyl phosphatidylethanolamine (PE) were procured from Northern Lipids Inc., and cholesterol (Chol) was purchased from Sigma. 3,3′-Dipropylthiadicarbocyanine iodide and 4-fluoro-7-nitrobenz-2-oxa-1,3-diazole (NBD)-fluoride were purchased from Molecular Probes (Eugene, OR). The rest of the reagents were of analytical grade and procured locally; buffers were prepared in milli Q water.Peptide Synthesis, Fluorescent Labeling, and Purification—All of the peptides were synthesized manually on solid phase. Stepwise solid phase syntheses were carried out on rink amide MBHA resin (0.15 mmol) utilizing the standard Fmoc chemistry, employing N,N′-diisopropylcarbodiimide/1-hydroxybenzotriazole or 1,1,3,3-tetramethyluronium tetrafluoroborate/1-hydroxybenzotriazole/N,N′-diisopropylethylamine coupling procedure (30Ghosh J.K. Ovadia M. Shai Y. Biochemistry. 1997; 36: 15451-15462Crossref PubMed Scopus (48) Google Scholar, 31Fields G.B. Noble R.L. Int. J. Pept. Protein Res. 1990; 35: 161-214Crossref PubMed Scopus (2313) Google Scholar, 32Yadav S.P. Kundu B. Ghosh J.K. J. Biol. Chem. 2003; 278: 51023-51034Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar) with a checking of the deprotection of the α-amino group and the coupling of amino acids by usual procedures (33Kaiser E. Colescott R.L. Bossinger C.D. Cook P.I. Anal. Biochem. 1970; 34: 595-598Crossref PubMed Scopus (3496) Google Scholar, 34Vojkovsky T. Pept. Res. 1995; 8: 236-237PubMed Google Scholar). After the synthesis was over, each peptide was cleaved from the resin by a standard method as reported earlier (32Yadav S.P. Kundu B. Ghosh J.K. J. Biol. Chem. 2003; 278: 51023-51034Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar) and precipitated with dry ether. Labeling at the N terminus of peptides with a fluorescent probe was achieved by a standard procedure (30Ghosh J.K. Ovadia M. Shai Y. Biochemistry. 1997; 36: 15451-15462Crossref PubMed Scopus (48) Google Scholar, 32Yadav S.P. Kundu B. Ghosh J.K. J. Biol. Chem. 2003; 278: 51023-51034Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar, 35Rapaport D. Shai Y. J. Biol. Chem. 1992; 267: 6502-6509Abstract Full Text PDF PubMed Google Scholar, 36Ghosh J.K. Shai Y. J. Biol. Chem. 1998; 273: 7252-7259Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar), and after sufficient labeling, the labeled peptides were cleaved from the resins and precipitated. All the peptides were purified by reverse phase HPLC on an analytical Vydac C18 column using a linear gradient of 0–80% acetonitrile for 45 min with a flow rate of 0.6 ml/min. Both acetonitrile and water contained 0.05% trifluoroacetic acid. The purified peptides were ∼95% homogeneous as shown by HPLC. Experimental molecular mass of the peptides, detected by electrospray-mass spectrometry analysis, corresponded to the desired values within 0.5 Da.Preparation of Small Unilamellar Vesicles—Small unilamellar vesicles were prepared by a standard procedure (32Yadav S.P. Kundu B. Ghosh J.K. J. Biol. Chem. 2003; 278: 51023-51034Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar, 35Rapaport D. Shai Y. J. Biol. Chem. 1992; 267: 6502-6509Abstract Full Text PDF PubMed Google Scholar, 36Ghosh J.K. Shai Y. J. Biol. Chem. 1998; 273: 7252-7259Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar) with required amounts of either of the PC/cholesterol (8:1 w/w), PC/PG/cholesterol (4:4:1 w/w), or PE/PG (7:3 w/w). Dry lipids were dissolved in CHCl3/MeOH (2:1 v/v) in small glass vial. Solvents were evaporated under a stream of nitrogen resulting the formation of a thin film on the wall of a glass vessel. The thin film was resuspended in a buffer at a concentration of 8.2 mg/ml by vortex mixing. The lipid dispersions were then sonicated in a bath-type sonicator (Laboratory Supplies Company, New York) for 10–20 min until it became transparent. The lipid concentration was determined by phosphorus estimation (37Bartlett G.R. J. Biol. Chem. 1959; 234: 466-468Abstract Full Text PDF PubMed Google Scholar).Circular Dichroism Experiments—The circular dichroism (CD) spectra of peptides were recorded in phosphate-buffered saline (PBS, pH 7.4) and in the presence of phospholipid vesicles by utilizing a Jasco J-810 spectropolarimeter, calibrated routinely with 10-camphor sulfonic acid. The samples were scanned at room temperature (∼30 °C) in a capped quartz cuvette of 0.20-cm path length in the wavelength range of 250–195 nm. The fractional helicities were calculated with the help of mean residue ellipticity values at 222 nm by the following equation (38Greenfield N. Fasman G.D. Biochemistry. 1969; 8: 4108-4116Crossref PubMed Scopus (3302) Google Scholar, 39Wu C.S.C. Ikeda K. Yang J.T. Biochemistry. 1981; 20: 566-570Crossref PubMed Scopus (251) Google Scholar),Fh=[θ]222−[θ]2220[θ]222100−[θ]2220(Eq. 1) as already reported (30Ghosh J.K. Ovadia M. Shai Y. Biochemistry. 1997; 36: 15451-15462Crossref PubMed Scopus (48) Google Scholar, 32Yadav S.P. Kundu B. Ghosh J.K. J. Biol. Chem. 2003; 278: 51023-51034Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar), where [θ]222 was the experimentally observed mean residue ellipticity at 222 nm. The values for [θ]100222 and [θ]0222 that correspond to 100 and 0% helix contents were considered to have mean residue ellipticity values of –32,000 and –2000, respectively, at 222 nm (39Wu C.S.C. Ikeda K. Yang J.T. Biochemistry. 1981; 20: 566-570Crossref PubMed Scopus (251) Google Scholar).Membrane Permeability Assay, Dissipation of Diffusion Potential from the Small Unilamellar Lipid Vesicles—The ability of melittin and its analogs to destabilize the phospholipid bilayer was detected by the assay of peptide induced dissipation of diffusion potential across the lipid vesicles by employing a standard procedure (40Sims P.J. Waggoner A.S. Wang C.H. Hoffmann J.R. Biochemistry. 1974; 13: 3315-3330Crossref PubMed Scopus (759) Google Scholar, 41Loew L.M. Rosenberg I. Bridge M. Gitler C. Biochemistry. 1983; 22: 837-844Crossref PubMed Scopus (63) Google Scholar) as reported before (30Ghosh J.K. Ovadia M. Shai Y. Biochemistry. 1997; 36: 15451-15462Crossref PubMed Scopus (48) Google Scholar, 32Yadav S.P. Kundu B. Ghosh J.K. J. Biol. Chem. 2003; 278: 51023-51034Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar, 41Loew L.M. Rosenberg I. Bridge M. Gitler C. Biochemistry. 1983; 22: 837-844Crossref PubMed Scopus (63) Google Scholar). Lipid vesicles, prepared in K+ buffer (50 mm K2SO4, 25 mm HEPES sulfate, pH 6.8), were mixed with isotonic (K+-free) Na+-buffer followed by the addition of the potential sensitive dye 3,3′-dipropylthiadicarbocyanine iodide. After the addition of valinomycin to the vesicle suspension when fluorescence of the dye exhibited a steady level, the respective peptide was added. The peptide-induced dissipation of diffusion potential, as detected by an increase in fluorescence (at 670 nm with excitation at 620 nm), was measured in terms of percentage of fluorescence recovery (Ft), defined by Ft = [(It – I0)/(If – I0)] × 100%, where It is fluorescence after the addition of a peptide at time t, I0 is fluorescence after the addition of valinomycin, and If is the total fluorescence observed before the addition of valinomycin.Enzymatic Cleavage Experiments—To detect the location of the peptides in their membrane-bound states, enzymatic cleavage experiments were performed with their NBD-labeled versions as reported earlier (30Ghosh J.K. Ovadia M. Shai Y. Biochemistry. 1997; 36: 15451-15462Crossref PubMed Scopus (48) Google Scholar, 32Yadav S.P. Kundu B. Ghosh J.K. J. Biol. Chem. 2003; 278: 51023-51034Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar). In brief, lipid vesicles made of either PC/Chol, PC/PG/Chol, or PE/PG were first added to the NBD-labeled peptide. When all the peptide was bound to lipid as detected by the saturation of the fluorescence level, proteinase K (final concentration, 12.0 μg/ml) was added. In this experiment fluorescence of NBD-labeled peptide was recorded at 527 nm with respect to time (s) with excitation wavelength set at 467 nm.Hemolytic Activity Assay—The hemolytic activity of melittin and its analogs was assayed by a standard procedure (10Oren Z. Shai Y. Biochemistry. 1997; 36: 1826-1835Crossref PubMed Scopus (373) Google Scholar, 12Subbalakshmi C Nagaraj R Sitaram N. FEBS Lett. 1999; 448: 62-66Crossref PubMed Scopus (72) Google Scholar). In brief, fresh human red blood cells (hRBCs) that were collected in the presence of an anti-coagulant from a healthy volunteer were washed three times in PBS. Peptides, dissolved in water, were added to the suspension of red blood cells (6% final in v/v) in PBS to the final volume of 200 μl and incubated at 37 °C for 35 min. The samples were then centrifuged for 10 min at 2000 rpm, and the release of hemoglobin was monitored by measuring the absorbance (Asample) of the supernatant at 540 nm. For negative and positive controls hRBCs in PBS (Ablank) and in 0.2% (final concentration v/v) Triton X-100 (ATriton) were used, respectively. The percentage of hemolysis was calculated according to the equation, Percentage of hemolysis = [(Asample – Ablank)/(ATriton – Ablank)] × 100.Assay of Antibacterial Activity of the Peptides—The antibacterial activity of the peptides was assayed in LB medium under aerobic conditions (10Oren Z. Shai Y. Biochemistry. 1997; 36: 1826-1835Crossref PubMed Scopus (373) Google Scholar, 12Subbalakshmi C Nagaraj R Sitaram N. FEBS Lett. 1999; 448: 62-66Crossref PubMed Scopus (72) Google Scholar, 42Leippe M. Andra J. Muller-Eberhard H.J. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 2602-2606Crossref PubMed Scopus (96) Google Scholar). Different concentrations of each of the peptides, dissolved in water, were added in duplicate to 100 μl (final volume) of medium containing the inocula of the test organism (∼106 colony forming units) in the midlogarithmic phase of growth. The inhibition of growth of microorganisms was assessed by measuring the absorbance at 492 nm after an incubation of 18–20 h at 37 °C. The antibacterial activity of the peptides was expressed as the minimal inhibitory concentration, the concentration at which 100% inhibition of microbial growth was observed after 18–20 h of incubation. The microorganisms used were Gram-positive bacteria, Bacillus subtilis and Staphylococcus aureus, and Gram-negative bacterium, Escherichia coli.RESULTSDesign of Analogs of Melittin—Melittin has an interesting sequence with 26 amino acids including five cationic residues. A leucine zipper motif, with every seventh amino acid as leucine/isoleucine (leucine at position 6 and 13 and isoleucine at position 20) and also leucine in two “d” positions, was identified (Fig. 1). To look into the possible role of this important structural motif, two analogs of melittin were designed. In one of which leucine at position 13 (MM-1) was replaced with alanine and in the other (MM-2) leucine at positions 6 and 13 were replaced with two alanines (Fig. 1A). Alanine was chosen for its helix propensity and similar hydrophobicity to leucine so that the amphipathic character of melittin was retained in MM-1 and MM-2.Mutation in the Leucine Zipper Motif Drastically Reduced the Hemolytic Activity of Melittin—To look into the role of the identified leucine zipper motif in the functional activities of melittin, hemolytic activity of melittin and its designed analogs were assayed. The substitution of leucine with alanine resulted in a drastic reduction of hemolytic activity of melittin against hRBCs (Fig. 2). MM-1 exhibited 10–20% hemolytic activity of melittin, whereas MM-2 showed ∼1–2% activity. The results indicated that the substitution of heptadic amino acids by alanine markedly inhibited the hemolytic activity of melittin probably suggesting a role of this motif in maintaining the lytic activity of the peptides toward the RBCs.Fig. 2Effect of mutation in the leucine zipper motif of melittin on its hemolytic activity. Dose-dependent hemolytic activity of melittin (a), MM-1 (b), and MM-2 (c) against human red blood cells.View Large Image Figure ViewerDownload Hi-res image Download (PPT)The Designed Analogs Exhibited Antibacterial Activity Comparable to Melittin—Melittin and its analogs were tested for growth-inhibiting activity in liquid cultures of two Gram-positive and one Gram-negative bacteria. Tetracycline was used as a positive control. In contrast to the hemolytic activity, the designed analogs of melittin (MM-1 and MM-2) exhibited very similar antibacterial activities to that of melittin (Table I). Thus the results clearly indicated that substitution of leucine by alanine in the heptadic positions did not affect the antibacterial activity of melittin unlike its hemolytic activity.Table IAntibacterial activity of melittin and its analogs against Gram-positive and Gram-negative bacteriaPeptide/antibiotic nameMinimal inhibitory concentrationE. coli DH5αB. subtilisS. aureusμmMelittin3.9 ± 0.62.0 ± 0.23.6 ± 0.5MM-14.5 ± 0.72.4 ± 0.34.3 ± 0.40MM-24.2 ± 0.71.8 ± 0.23.6 ± 0.6Tetracycline1.2 ± 0.20.2 ± 0.14.0 ± 0.2 Open table in a new tab Contrasting Difference in Membrane Permeability of the Designed Analogs of Melittin in Zwitterionic and Negatively Charged Membranes—The ability of melittin to destabilize the phospholipid bilayer is believed to be associated with its mechanism of action. Therefore, to understand the basis of the decreased hemolytic activity and almost unaltered antibacterial activity of the designed melittin analogs, the membrane permeability of melittin, MM-1, and MM-2 were examined by determining their efficacy to dissipate the diffusion potential across the phospholipid vesicles with different lipid composition. As shown in Fig. 3A, the membrane permeability (expressed as the percentage of fluorescence recovery) of MM-1 was significantly lower than melittin in zwitterionic PC/Chol vesicles, which decreased further with MM-2. However, interestingly, both MM-1 and MM-2 exhibited very similar membrane permeability to that of melittin in negatively charged PC/PG/Chol lipid vesicles (Fig. 3B). Experimental profiles for each of the peptides in both kinds of membranes are shown in the insets of Fig. 3, A and B. Moreover, the designed analogs induced membrane permeation as efficiently as melittin (not shown) in another kind of negatively charged membrane, namely PE/PG (7:3 w/w). The results indicated that although the substitution of leucine by alanine appreciably impaired the membrane permeability of melittin in the zwitterionic membrane, it had an almost negligible effect in negatively charged membrane.Fig. 3Membrane permeability induced by melittin and its analogs as plotted by percent fluorescence recovery induced by the individual peptides. A, plots of fluorescence recovery induced by different concentrations of individual peptides in zwitterionic PC/Chol lipid vesicles (68 μm, final concentration). Closed squares, melittin; closed circles, MM-1; closed triangles, MM-2. Inset depicts the representative experimental profiles for melittin and its analogs at 1.52 μm as marked. B, plots of fluorescence recovery induced by melittin, MM-1 and MM-2 at different concentrations (symbols are the same as in A) in negatively charged PC/PG/Chol lipid vesicles (68 μm, final concentration). Inset depicts the representative experimental profiles for melittin and its analogs as marked at 0.76 μm. Fluorescence recovery was measured at ∼300 s after the addition of peptides to lipid vesicles.View Large Image Figure ViewerDownload Hi-res image Download (PPT)The Designed Analogs Exhibited Similar Secondary Structures to Melittin in Negatively Charged Membranes but Not in Zwitterionic Membranes—Circular dichroism experiments were performed to determine the secondary structures of melittin and its analogs in aqueous environment (PBS, pH 7.4) and in the presence of zwitterionic and negatively charged phospholipid vesicles. The corresponding mean residue ellipticity values of the individual peptides at 222 nm were used to determine their helical contents in a particular environment. Although all three peptides exhibited low helical structure in PBS (Fig. 4A), the extent of helicity was slightly more in melittin (∼15%) than MM-1 (∼8.5%) and MM-2 (∼6.5%). However, in the presence of increasing amount of negatively charged PC/PG/Chol lipid vesicles significant amount of helicity was induced in all three peptides as shown by a representative CD spectrum for each of these peptides with a fixed lipid/peptide molar ratio (∼32) (Fig. 4B). Melittin (69%), MM-1 (67.5%), and MM-2 (68.5%) exhibited very similar extents of helical structures indicating that the substitution of leucine by alanine did not affect the secondary structure of melittin in the presence of negatively charged lipid vesicles.Fig. 4Determination of secondary structures of melittin (15.2 μm), MM-1 (14.9 μm), and MM-2 (14.5 μm) in PBS (A), in the presence of 482 μm PC/PG/Chol (B), and 362 μm PC/Chol (C). Solid line, melittin; dashed line, MM-1; dotted line, MM-2.View Large Image Figure ViewerDownload Hi-res image Download (PPT)However, very contrasting results were obtained in the presence of zwitterionic PC/Chol lipid vesicles (Fig. 4C). Although melittin adopted a significant amount of helical stru