Effects of Acyl versus Aminoacyl Conjugation on the Properties of Antimicrobial Peptides

Abstract

To investigate the importance of increased hydrophobicity at the amino end of antimicrobial peptides, a dermaseptin derivative was used as a template for a systematic acylation study. Through a gradual increase of the acyl moiety chain length, hydrophobicity was monitored and further modulated by acyl conversion to aminoacyl. The chain lengths of the acyl derivatives correlated with a gradual increase in the peptide's global hydrophobicity and stabilization of its helical structure. The effect on cytolytic properties, however, fluctuated for different cells. Whereas acylation gradually enhanced hemolysis of human red blood cells and antiprotozoan activity against Leishmania major, bacteria displayed a more complex behavior. The gram-positive organism Staphylococcus aureus was most sensitive to intermediate acyl chains, while longer acyls gradually led to a total loss of activity. All acyl derivatives were detrimental to activity against Escherichia coli, namely, but not solely, because of peptide aggregation. Although aminoacyl derivatives behaved essentially similarly to the nonaminated acyls, they displayed reduced hydrophobicity, and consequently, the long-chain acyls enhanced activity against all microorganisms (e.g., by up to 12-fold for the aminolauryl derivative) but were significantly less hemolytic than their acyl counterparts. Acylation also enhanced bactericidal kinetics and peptide resistance to plasma proteases. The similarities and differences upon acylation of MSI-78 and LL37 are presented and discussed. Overall, the data suggest an approach that can be used to enhance the potencies of acylated short antimicrobial peptides by preventing hydrophobic interactions that lead to self-assembly in solution and, thus, to inefficacy against cell wall-containing target cells.

Peptide-based antimicrobials represent a promising class of novel antimicrobial agents (2, 24, 30, 60). A large body of data indicates that antimicrobial peptides kill target cells by destabilizing the structure of cell membranes by a mechanism whose fine details remain to be fully understood (18, 25, 45). Antimicrobial peptides also display a variety of interesting properties: they may activate the microbicidal activities of leukocytes and monocytes/macrophages (1, 46, 47), suppress the production of inflammatory cytokines, offer protection from the cascade of events that lead to endotoxic shock (10, 16, 29), and display synergistic activity in the presence of other peptides (26, 35, 57) or conventional antibiotics (22). Antimicrobial peptide genes introduced into the genomes of plants endowed the plants with resistance to pathogens (40). These multifunctional peptides may be useful in food preservation (9, 58), as imaging probes for the detection of bacterial or fungal infection loci (56), and as linings for medical and surgical devices (4, 23). Clearly, the externally localized site of action and receptor-independent mechanism of peptide-based antimicrobials may significantly prevent drug resistance. However, this mechanism is also largely responsible for unselective activity over a wide range of cell types (12, 21, 31). Thus, a major challenge toward their safe use is the ability to endow antimicrobial peptides with increased specificity.

To achieve this goal, numerous strategies were adopted. Synthetic combinatorial libraries were used to rapidly obtain optimized classes of active compounds (7). Antimicrobial peptides were designed by using a sequence templates approach, which consists of the extraction of sequence patterns after comparison of large series of natural counterparts (52). Minimalist-approach methods were used to design antimicrobial peptides based on the requirement for amphipathic structures (6, 17, 49). Various sequence modification methods attempted to modify natural peptides by deleting, adding, or replacing one or more residues or assemble chimeric peptides from segments of different natural peptides. They have been extensively applied to the study of cecropins, magainins, and melittins (39, 43, 47) and of dermaseptins (4, 41, 48). Finally, acylation of antimicrobial peptides proved to be a useful technique for improving antimicrobial characteristics (3, 13, 32-34, 55). Moreover, naturally occurring, short peptides with antimicrobial activity characterized by a lipophilic acyl chain at the N terminus were discovered in various microorganisms (5, 42, 51, 53).

Tree frog dermaseptins (8, 11, 36) and their derivatives (14, 19, 37) exert rapid cytolytic activity against a variety of microorganisms. When the rates of the emergence of resistance were compared by propagating bacteria under selective antibiotic pressure, bacteria developed resistance to commercial antibiotics but not to the l or d isomers of the dermaseptin derivatives (38). Dermaseptin derivatives were also investigated for their effects on malaria parasite-infected red blood cells (RBCs) by taking advantage of their ability to translocate spontaneously across the plasma membrane of mammalian cells (15). Certain acylated derivatives displayed both increased specific antiparasitic efficiency and reduced hemolysis. Thus, a variety of reports suggest that antimicrobial activity could be enhanced through conjugation of an acyl moiety. However, in our experience, acylation was rather deleterious to various antimicrobial peptides. To explore this issue in depth, acyls ranging from acetyl to palmitoyl were conjugated to the amino terminus of peptide P, a 13-mer peptide, K4-S4(1-13), the shortest derivative that maintained potency in previous studies (19, 38). The rationale for selecting N-terminus acylation was based on observations that the carboxyl ends of dermaseptins are not major contributors to membrane lytic activity (1, 19, 35, 37), whereas the amino end is inserted within the membrane bilayer (20). The resulting lipopeptides were probed for their structural and biological properties, and these properties were compared to those of a second series of aminoacyl analogs. The effects of acyl conjugation were also investigated by using the frog derived 22-residue magainin analog MSI-78 and the 37-residue cathelicidin-derived human antimicrobial peptide LL37. The data presented indicate how potency and selectivity are affected by the nature of the acyl portion and further demonstrate the limits of this approach.

MATERIALS AND METHODS

Peptide synthesis.

The reference peptides were first synthesized by the solid-phase method by applying 9-fluorenylmethyloxy carbonyl (Fmoc) active ester chemistry on a fully automated Applied Biosystems model 433A peptide synthesizer. MBHA resin (4-methylbenzhydrylamine; Novabiochem, Darmstadt, Germany) was used to obtain an amidated peptide (dermaseptin and MSI derivatives), whereas the Wang resin was used to obtain the free carboxyl of LL37.

The various analogs were prepared by linking the N terminus of the reference peptides to one of the compounds listed in Table 1. Selective reaction with the amino-terminal group was ensured by selective removal of Fmoc from the N terminus of the Fmoc-protected resin-bound peptide by exposing the resin to a solution of 20% piperidine-N-methylpyrrolidone (NMP), whereas all other potentially reactive groups remained masked by orthogonal protecting groups. The deprotected resin-bound peptide was washed with NMP to remove the piperidine and suspended in dimethylformamide (DMF), to which a twofold molar excess of the relevant acid compound was added, followed by the addition of a threefold molar excess of 1-ethyl-3-(dimethylaminopropyl)carbodiimide. The reaction mixtures were sonicated (5 min) and then agitated for 24 h at room temperature. Each mixture was then centrifuged for 3 min (the supernatant was discarded), and the resin was washed four times with DMF and then three times with ether-dichloromethane (1:1). The residue was dried for 1 h at room temperature and then for 4 h at 50°C.

TABLE 1.

Primary structures of the peptides investigated and their properties

Peptide sequence	Designation	Ha	Helixb	LC50 (μM) for RBCc	MLC (μM) for L.m.d	MIC (μM)e
						S.a.	E.c.
ALWKTLLKKVLKACONH2	P	45	35	50	12.5	9 ± 3	4.5 ± 1.5
Acetyl-------------	C2-P	53	41	>100	18 ± 7	25	25
Propionyl-------------	C3-P	53.5	42	80	18 ± 7	12.5	25
Butyroyl-------------	C4-P	54	48	75	12.5	12.5	25
Pentanoyl-------------	C5-P	55	ND	60	6.25	3.12	12.5
Hexanoyl-------------	C6-P	57	49	15	6.25	2.2 ± 0.8	12.5
Octanoyl-------------	C8-P	60	43	4.5	6.25	1.5	6.25
Decanoyl-------------	C10-P	63	50	4.5	6.25	3.12	12.5
Lauryl-------------	C12-P	64	55	4.5	6.25	4.5 ± 1.5	>50
Myristoyl-------------	C14-P	66	65	4.5	3.12	>50	>50
Palmitoyl-------------	C16-P	68	65	9	6.25	>50	>50
Aminoacetyl-------------	NC2-P	46	38	75	25	38 ± 12	12.5
Aminobutyroyl-------------	NC4-P	47.5	49	>100	6.25	50	12.5
Aminoheptanoyl-------------	NC7-P	49	NDf	80	6.25	18 ± 7	4.5 ± 1.5
Aminolauryl-------------	NC12-P	52.5	59	18	4.5 ± 1.5	0.78	1.56
GIGKFLKKAKKFGKAFVKILKKCONH2	MSI-78	44	11	45	ND	9 ± 3	1.56
Lauryl----------------------	C12-MSI-78	55	19	6	ND	>50	>50
Aminolauryl----------------------	NC12-MSI-78	47	17	18	ND	3.12	3.12
LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES	LL37	58	30	8	ND	50	50
Lauryl-------------------------------------	C12-LL37	68	32	4	ND	>50	>50
Aminolauryl-------------------------------------	NC12-LL37	62	31	2	ND	25	12.5

^aH, hydrophobicity measure determined by reversed-phase HPLC (% acetonitrile-water eluent).

^bHelical content (%) calculated from circular dichroism measurements in 20% trifluoroethanol.

^cLC₅₀, peptide concentration that induced 50% lysis of RBCs after 3 h incubation. Values were extracted from dose-response curves and represent the means from two experiments. Standard deviations were <10%.

^dMLC, peptide concentration that induced 100% lysis of L. major (L.m.) after 3 h culture at room temperature.

^eMIC, peptide concentration that caused 100% growth inhibition of S. aureus (S.a.) and E. coli (E.c.) after 3 h culture at 37°C. Values represent the means ± standard deviations obtained from at least two experiments.

^fND, not determined.

Cleavage of the peptides (LL37 and derivatives excepted) from the resin was performed in a mixture of 2.5% water, 2.5% triethylsilane, and 95% trifluoroacetic acid (TFA); the mixture was stirred in an ice bath for 15 min and then at room temperature for 2 h. LL37 and derivatives were cleaved from the resin with a mixture of 2% tri-isopropylsilane, 5% phenol, 5% water, and 88% TFA. After filtration of the resin, the peptide-TFA filtrate was precipitated (by addition of the cleavage mixture drop by drop) in ice-cold diethyl ether, placed in a fume hood to dry at 60°C for 2 h, and then dissolved in 10% acetic acid and lyophilized. The crude peptide was purified to chromatographic homogeneity in the range of 95 to >99% by reverse-phase high-performance liquid chromatography (HPLC; LC-MS Alliance-ZQ; Waters). HPLC runs were performed on a C₄ column (Vydac) with a linear gradient of acetonitrile in water (1%/min); both solvents contained 0.1% TFA. The purified peptides were subjected to mass spectrometry analysis in order to confirm their compositions. The peptides were stocked as lyophilized powders at −20°C. Prior to testing of the peptides, fresh solutions were prepared in water, briefly vortexed, sonicated, centrifuged, and then diluted in the appropriate medium. Buffers were prepared with distilled water (mQ; Millipore). All other reagents were analytical grade.

CD.

Circular dichroism (CD) spectra (in millidegrees) were measured as described previously (28) with an Aviv model 202 CD spectrometer (Aviv Associates, Lakewood, N.J.) by using a 0.010-cm rectangular QS Hellma cuvette at 25°C (controlled by thermoelectric Peltier elements with an accuracy of 0.1°C). The CD spectra of the peptide samples (100 μM; determined with UV light by using standard curves of known concentrations for each peptide), which were dissolved in 20% trifluoroethanol-water (1:4, vol/vol), were scanned. Fractional helix contents were determined from [θ]₂₂₂ measurements by using −40,000 (1 to 2.5/n, where n is the number of amino acid residues in the peptides) and 0 deg × cm² × dmol⁻¹ as values for 100 and 0% helixes, respectively. CD data represent average values from three separate recordings.

Organizational studies.

Peptide self-assembly (aggregation) in solution was investigated by measurement of static light scattering, as described previously (28). The peptides were successively diluted in phosphate-buffered saline (PBS; 50 mM sodium phosphate, 150 mM NaCl, pH 7.3), and light scattering was evaluated by measuring the reflected light at an angle of 90°, with both the excitation and the emission held at 400 nm. The intensity of scattered light was plotted against the total peptide concentrations, and linear regression analysis was performed on the data at the concentration range close to the monomer-micelle transition zone. The static light-scattering signal is proportional to the number of aggregated molecules and their size. Therefore, the slope is indicative of the aggregation tendency and reveals the aggregation properties of the peptides, where a slope value above unity indicates the presence of the micellar form. The critical micelle concentration (CMC) was evaluated by extrapolating the curve to the intercept with the x axis.

Bioassays. (i) Bacteria.

MICs were determined by microdilution susceptibility testing (28). Antibacterial activity was assessed against two clinical isolates, Escherichia coli (U16318) and Staphylococcus aureus (B38302), representatives of gram-negative and gram-positive bacteria, respectively. Antibacterial assays were routinely performed in 2xty culture medium (16 g/liter trypton, 10 g/liter yeast extract, 5 g/liter NaCl, pH 7.4). Alternatively, assays were performed in Luria-Bertani (LB) medium (10 g/liter trypton, 5 g/liter yeast extract, 5 g/liter NaCl, pH 7.4). Inocula of 10⁶ bacteria/ml were used. The cell populations were estimated by measurement of the optical density at 620 nm, with reference to a calibration curve. A total of 100 μl of the bacterial suspension was added to 100 μl of culture medium containing no peptide or a peptide at various concentrations (serial twofold dilutions) in 96-well plates. Inhibition of proliferation was determined by measurement of the optical density (620 nm) after the incubation period at 37°C.

(ii) Activity against E. coli with enhanced membrane permeability.

To enhance the outer membrane permeability of E. coli, bacterial cultures were treated with EDTA according to the following procedure. E. coli was grown in 2xty culture medium until the exponential phase of growth. The cells (about 1 ml) were washed in saline by successive (twice) centrifugation (1 min, 20,000 × g) in order to remove the divalent cations that can be present in the growth medium. The washed cells were resuspended in Tris-EDTA in saline (25 mM Tris-HCl, 20 mM EDTA, pH 8.0) or Tris in saline (25 mM Tris-HCl, pH 8.0) (control experiment) to yield a bacterial population of 10⁶ cells/ml, as estimated by measurement of the optical density at 620 nm, with reference to a standard calibration curve. After an incubation period of 30 min under shaking at 37°C, 100 μl of the bacterial suspension was added to 100 μl of culture medium containing no peptide or various peptide concentrations (serial twofold dilutions) in 96-well plates. Inhibition of proliferation was assessed by measurement of the optical density measurement (620 nm) after 16 h of incubation at 37°C.

(iii) Kinetic studies.

For the kinetic studies (58), 100-μl stock solutions of peptides prepared in LB culture medium to yield a final concentration of four multiples of the MIC were added to Eppendorf tubes containing 100 μl of bacteria (S. aureus or E. coli) at the exponential phase of growth. After 0, 5, 30, and 60 min of exposure to the peptides at 37°C, the cultures were subjected to serial 10-fold dilution (up to 1/10,000) by adding 20 μl of sample to 180 μl of cold PBS, from which 50-μl aliquots were plated on TyE agar plates (15 g/liter of agar, 10 g/liter of tryptone, 5 g/liter of yeast extract, 8 g/liter of NaCl, pH 7.5) for determination of CFU counts after an additional overnight incubation at 37°C. Statistical data were obtained from at least two independent experiments performed in duplicate.

(iv) Hemolysis.

The peptides' membranolytic potentials against human RBCs in PBS were determined as described previously (28). Human blood was rinsed three times in PBS by centrifugation at 200 × g for 2 min and resuspended in PBS at 5% hematocrit. A 50-μl suspension containing 2.5 × 10⁸ RBCs was added to test tubes containing 200 μl of peptide solutions (serial twofold dilutions in PBS), PBS alone (for baseline values), or distilled water (for 100% hemolysis). After 3 h incubation at 37°C under agitation, the samples were centrifuged and the hemolytic activity was determined as a function of hemoglobin leakage by measuring the absorbance (405 nm) of 200 μl of the supernatants. Statistical data were obtained from two independent experiments performed in duplicate.

(v) Leishmania.

Activity against the promastigote form of a Leishmania major clinical isolate was assessed. To measure the inhibition of proliferation, a 100-μl suspension of promastigotes (1 × 10⁶ cells/ml of RPMI 1640 complemented with 20% fetal calf serum, 1% penicillin, and 1% streptomycin) was added to 100 μl culture medium in 96-well plates containing no peptide or the peptides at various concentrations (serial twofold dilutions). After the incubation period (3 h, 27°C), the number of viable (motile) cells was determined by counting aliquots from each culture on a cell counter under a microscope. Statistical data were obtained from two independent experiments performed in duplicate.

(vi) Susceptibility to plasma proteases.

Peptide sensitivity to enzymatic degradation was assessed by determining the antibacterial activity after exposure to human plasma. For this, 250 μl of peptide saline solution (0.9% NaCl) at a concentration of 100 × the MIC was preincubated with 50% (vol/vol) human plasma in culture medium at 37°C. After incubation periods of 3, 6, and 18 h, the peptide solutions were subjected to serial twofold dilution in LB medium in 96-well plates. Inhibition of E. coli and S. aureus growth was determined as described above for the antibacterial bioassay. In parallel, the antibacterial activity was determined in culture medium conditions in the absence of plasma (referred as 0 h of preincubation). Statistical data were obtained from at least two independent experiments performed in duplicate.

RESULTS

To investigate the effects of acyl conjugation to the amino terminus, peptides belonging to three well-characterized families of linear antimicrobial peptides were probed. Initially, a dermaseptin derivative was subjected to a systematic acylation study (from acetyl to palmitoyl), and selected representatives were converted to the aminoacyl form for comparison. Reference peptides and derivatives were investigated in terms of molecular organization in solution as well as in terms of cytolytic activities against two types of cells. The first cell type was bacteria (S. aureus and E. coli) to represent the increasingly limited access of the peptides to their stipulated target, the plasma membrane. The second cell type included L. major promastigotes and human RBCs to represent maximal access to cells with differences in their plasma membrane compositions. The effects of acylation on longer peptides (MSI-78 and LL37) were investigated subsequently. The primary structures of the peptides studied are shown in Table 1, along with their structural and biological properties.

Acyl-P derivatives.

As shown in Table 1, the hydrophobicity of the reference peptide, peptide P, increased with increasing chain length of the acyl moiety, while circular dichroism measurements indicated that hydrophobicity stabilized the peptide's alpha-helical content.

The effect of acylation on cytolytic properties, however, revealed a complex pattern. The length of the acyl chain displayed, in general, a biphasic behavior, whereby short-chain acyls (e.g., C₂-P and C₃-P) had rather reduced cytolytic activities against both RBCs and L. major, whereas long-chain acyls had enhanced activities. Antibacterial activity, though, revealed an even more complex pattern: short-chain acyls limited the peptide's growth-inhibitory activity for the gram-positive bacterium S. aureus, while longer acyls, up to C₈-P, gradually enhanced inhibition by up to sixfold. Beyond C₈-P, activity was gradually limited and eventually abolished; i.e., C₁₄-P and C₁₆-P did not display MICs up to the highest concentration tested (50 μM). As for E. coli, although the general pattern was somewhat similar to that observed with S. aureus, none of the acyl derivatives enhanced the activity of peptide P against the gram-negative bacteria.

Aminoacyl-P derivatives.

Addition of the amino group to the acyl derivatives resulted in a pattern of general reduced hydrophobicity compared with those of the nonaminated acyls. Thus, NC₁₂-P, for instance, was significantly less hydrophobic than C₁₂-P and was slightly more hydrophobic than the reference peptide, peptide P. Yet, addition of the amino group did not seem to greatly alter the peptide structures compared with those of the nonaminated acyl derivatives. Thus, NC₁₂-P, for instance, displayed a helical content that was significantly greater than that of peptide P and that was practically identical to that of C₁₂-P, despite their differences in hydrophobicity.

Nonetheless, the aminoacyl derivatives affected the cytolytic properties of peptide P quite unlike they did those of the corresponding acyl conjugates. Although antileishmanial activity was practically unaffected, hemolysis was considerably reduced compared with the hemolysis caused by the corresponding nonaminated acyl peptides. Inhibition of the growth of both types of bacteria by short-chain aminoacyls was still limited but was considerably enhanced by long-chain aminoacyls. Thus, NC₁₂-P, for instance, displayed 12-fold and 3-fold increased potencies against S. aureus and E. coli, respectively.

Peptide organization in solution.

To understand the molecular basis for the observed discrepancies, peptide P and selected derivatives (including C₁₆-P, C₁₂-P, and NC₁₂-P) were further investigated with respect to their molecular organization in solution by measuring their concentration dependence on light scattering in PBS. As shown in Fig. 1A, C ₁₆-P and C₁₂-P were found to undergo self-assembly at low concentrations (CMCs, 0.02 and 0.1 μM, respectively), whereas NC₁₂-P behaved similarly to peptide P (both are assumed to be monomeric).

FIG. 1.

Assembly and disassembly in solution and consequences on antibacterial activity. (A) Peptide self-assembly in PBS, as investigated by static light-scattering measurements. The intensity of scattered light is plotted against the total peptide concentrations, and linear regression analysis was performed on the data at the concentration range close to the monomer-micelle transition zone. CMC was evaluated by extrapolation of the curve to the intercept with the x axis. (B) Antibacterial activity against E. coli, assessed as described in Materials and Methods, with the exception of the presence of EDTA (20 mM). The asterisk indicates that the MIC was >50 μM. Values represent the means from two independent experiments performed in duplicate.

This result suggested that the inactivities of C₁₂-P and the other highly hydrophobic peptides against bacteria may arise from their aggregated states. To further explore this possibility, the activities of both derivatives against E. coli were compared in the presence and the absence of EDTA, which is known to induce defects in the bacterial external membrane. As shown in Fig. 1B, the MIC of C₁₂-P dropped to 1.5 μM (a >30-fold reduction) in the presence of EDTA, whereas the MIC of the aminated counterpart displayed little difference (the MIC was reduced by 2-fold) under the same conditions.

Susceptibility to plasma proteases.

To assess the effect of acyl conjugation on the interactions of the peptides with plasma components, namely, inactivating interactions, such as susceptibility to enzymatic cleavage, peptide P and its acylated derivatives were incubated in presence of 50% plasma at various concentrations (from 0.25 to 25 times the MIC) and for various time periods (0, 3, 6, and 18 h) and then tested for their abilities to inhibit the growth of E. coli. Figure 2 shows typical results obtained for NC₁₂-P compared to those obtained for peptide P. As shown in Fig. 2A, unmodified peptide P was rapidly inactivated, showing about 20% inhibition but no MIC at 25 MIC multiples after 3 h incubation. After 6 or 18 h, inhibition was reduced to 0% at all peptide concentrations. NC₁₂-P also displayed various degrees of inactivation, but at a significantly slower rate (Fig. 2B). Thus, 100% inhibition was obtained, although at 3-, 12.5-, and 25-fold the MIC after 3, 6, and 18 h incubation, respectively. Increased stabilities of comparable magnitudes were observed for the other acylated dermaseptin derivatives (data not shown).

FIG. 2.

Peptide inactivation by plasma components. (A and B) Peptide activity against E. coli after exposure to human plasma (50% at 37°C). After various incubation periods, the samples were subjected to serial twofold dilution in culture medium and mixed with bacteria, and inhibition was determined as described in Materials and Methods. Symbols: asterisks, control experiment without plasma; squares, circles, and triangles, incubation for 3, 6 and 18 h, respectively. (C and D) Antibacterial activities of l and d isomers, determined as described above for panels A and B, after 3 h of exposure to human plasma. Statistical data were obtained from two independent experiments performed in duplicate.

To distinguish between the possibilities that the loss of activity was due to enzymatic degradation versus other sources of peptide inactivation (such as peptide nonavailability due to nonspecific binding to plasma components), the experiment was repeated at a single time point (3 h incubation) by comparing both the activities of peptide P and NC₁₂-P as their respective l and d stereoisomers. As shown in Fig. 2C, nearly total inactivation was observed for the l isomer but not the d isomer of peptide P, which displayed 100% activity at the MIC, while both isomers of NC₁₂-P displayed 100% activity only at threefold the MIC (Fig. 2D). Since the d isomer is unlikely to be cleaved by human plasma proteases, this result indicated that part (between half and two-thirds) of the lipopeptide was inactivated nonenzymatically. Comparison by reversed-phase HPLC combined with mass analysis of the samples after 3 h incubation support these results, inasmuch as nearly 20% intact NC₁₂-P was recovered, whereas peptide P was undetectable (data not shown).

Acylated derivatives of nondermaseptin peptides.

The effects of acyl conjugation were also investigated with two nondermaseptin antimicrobial peptides: MSI-78 and LL37. Both peptides were conjugated to lauryl and aminolauryl moieties at their amino ends, as described for the dermaseptin derivatives, and the results are summarized in Table 1.

Lauryl-MSI-78 displayed increased hydrophobicity and stabilization of the peptide's alpha-helical structure (the helical content nearly doubled). Hemolytic activity was also enhanced, but the activities against both S. aureus and E. coli were inhibited. Replacement of the lauryl with aminolauryl reduced the hydrophobicity, while it maintained the increase in the helical structure. Hemolysis was still enhanced compared with that by MSI-78 but was reduced compared with that by lauryl-MSI-78, whereas inhibition of the growth of both types of bacteria was recovered (and was even enhanced threefold against S. aureus).

Acyl-conjugated LL37 displayed behavior similar to those of the dermaseptin and magainin derivatives with respect to increased hydrophobicity and its consequences on both hemolysis and antibacterial acitivity (Table 1), yet both acyls failed to increase the alpha-helix content (Fig. 3).

FIG. 3.

Effect of acylation on peptide structure. Circular dichroism spectra were measured for peptide samples (100 μM) that were dissolved in 20% (vol/vol) trifluoroethanol-water. Data represent average values from three separate recordings.

Kinetic studies.

The differences between C₁₂-P and NC₁₂-P were investigated in terms of their bactericidal kinetics; and the bactericidal kinetics were compared with those of MSI-78, LL37, and their aminolauryl derivatives at four times their respective MICs. The outcome is shown in Fig. 4. The dermaseptin derivative peptide P reduced the number of CFU by 6 log units within 30 min, whereas the activities of C₁₂-P and NC₁₂-P were indistinguishable; but both derivatives displayed faster kinetics than peptide P, achieving the same effect in 5 min or less.

FIG. 4.

Effect of acylation on bactericidal kinetics. Suspensions of S. aureus were added to culture medium containing zero or four multiples of the MIC. After exposure to the peptide, the cultures were subjected to serial 10-fold dilution and plated on TyE agar dishes for counting of the CFU after overnight incubation. The plotted values represent the mean ± standard deviations obtained from at least two independent experiments performed in duplicate.

Under the same conditions, MSI-78 reduced the number of CFU by 1 log unit within 120 min, whereas NC₁₂-MSI-78 managed to reduce the number of CFU by nearly 3 log units.

The activities of LL37 and NC₁₂-LL37 were similar to those of no treatment, although initially, the aminolauryl derivative displayed a slight activity that vanished at longer incubation times. Both peptides, however, were clearly aggregated at the concentration range tested (4× the MIC), as observed by the naked eye or upon measurement of the optical density at 620 nm (data not shown).

DISCUSSION

The systematic study around the 13-mer dermaseptin S4 derivative provided solid insight into the effects of acylation on the structural properties of the peptide and their relationships to cytolytic activity. The acyl moieties enhanced the overall hydrophobicity of reference peptide K4-S4(1-1) and stabilized its helical structure. Since the CD spectrum of the dermaseptin derivative correlated with the structure obtained from nuclear magnetic resonance imaging analysis (28), the CD method was used to monitor the relative structural changes caused by acylation. Given that a significant population of K4-S4(1-13) molecules is structured in amphipathic helices, the observed helical increase must point to the interactions of the acyl chain with the hydrophobic face of the peptide carrying the acyl or with that of a neighboring peptide molecule, in case of oligomerization. Either way, such stabilizing interactions presumably influence the electrostatic potential of the peptide surface (28) and thereby influence the interaction with membranes. Although it is possible that other mechanisms may explain the observed changes in activity, addition of various aminoacyl groups to various truncated (inactive) antimicrobial peptides did not lead to activity (unpublished data). However, if we assume that the outcome of peptide-membrane interactions depends on both charge and hydrophobicity, then full expression of either one of these forces obviously depends on adequate access to the plasma membrane. In other words, even in microbial cells, whose membranes are rich in negative charge, which promotes the binding affinities of cationic peptides (59), activity may still be hampered if access to large polymeric peptides is denied. Such polymerization can lead to inactivity against bacteria (but not against RBCs or Leishmania, for example), presumably because bacterial cell walls deny the access of large polymers to the cell membrane (19). These considerations may well explain the differential behaviors of the various derivatives investigated. Since access to the membranes of Leishmania and RBCs is not hindered, the effect of acylation is not complicated by this issue. Thus, given that the charge of peptides investigated is constant within each set (acyls and aminoacyl counterparts), increased hydrophobicity is expected to increase the interaction with the membrane and its disruption, as observed experimentally (data not shown). Indeed, acylation did enhance cytolytic activity against both L. major and RBCs. Addition of the amino group did not alter the structure of the corresponding acyl derivatives. It is therefore not surprising that antileishmanial activity was not altered; however, hemolysis was considerably reduced compared with that by the nonaminated acyl peptides. This discrepancy might be explained simply by differences in the phospholipid compositions of these cells. That is, unlike Leishmania (54), the external leaflet of the membrane of RBCs exposes a relatively poor negative charge (59) and the electrostatic component in this interaction is weak; hence, hemolytic activity depends acutely on the hydrophobic force. Therefore, reduced hydrophobicity systematically led to reduced hemolysis. The fact that the acyl derivatives of MSI-78 or LL37 displayed similar behaviors further supports this view, even though aminolauryl-LL37 was practically as hemolytic as lauryl-LL37. This is most probably due to the high hydrophobicity of LL37. Correspondingly, the complex antibacterial behaviors of the lipopeptides are probably dictated by the nature of the bacterial external barriers. The enhanced activities of some of the acyl peptides (such as C₆-P to C₁₂-P) against S. aureus indicated that the peptidoglycan-based cell wall was permeable to at least moderate-length acyl derivatives but not to longer ones. Moreover, the limited activity of the more hydrophilic NC₇-P (compared with that of C₆-P or C₈-P) indicates that S. aureus is more sensitive to hydrophobic antimicrobial peptides, as observed by other authors (27, 33). However, as hydrophobicity exceeds a certain threshold, it leads to peptide aggregation and to exclusion by the cell wall. It is thus clear from the data (namely, those for C₁₂-P) that hydrophobic peptides can display high potency as long as they are in (or close to) a monomeric state. Similar considerations may explain the behaviors of lipopeptides with the lipopolysaccharide-based, tightly packed, and highly hydrophilic external membrane of the gram-negative bacterium E. coli. We propose, therefore, that by introducing an amino group to the acyl moiety, hydrophobicity is reduced, aggregation is avoided, and peptide access to the plasma membrane becomes possible; hence, the activity is recovered. This hypothesis is supported by the results of experiments in which the activities of the peptides were compared before and after the introduction of defects in the external membrane (by using ion chelators), which artificially increased its permeability to aggregated C₁₂-P. The loss of antibacterial activity observed for C₁₂-MSI-78 and C₁₂-LL37 (especially at high concentrations) is most likely due to their high levels of aggregation. Likewise, the fact that aminoacyl conjugation to LL37 was least effective (compared with aminoacyl conjugation of peptide P and MSI-78) probably points to the small weight carried by the amino group relative to the weights of much larger molecules. It is also noteworthy that, contrary to the other cell types investigated, increased hydrophobicity was detrimental to the activity against E. coli, regardless of the aggregation issues.

Peptides are generally prone to rapid degradation by amino-, carboxy-, and endopeptidases, with plasma half-lives of about a few minutes, usually (44, 50). Whereas all-d isomers conserve biological activity and are insensitive to proteolysis, their accumulation could become problematic. In this respect, peptides that are degradable, but at a slower rate, may present an advantage. Our data showed that acyl derivatives have reduced susceptibilities to elimination in plasma. suggesting that they are less prone to degradation by plasma proteases due to the lipid moiety and the C-terminal amide, which protect the peptide from the action of aminopeptidases and carboxypeptidases, respectively.

In conclusion, the data presented here strongly indicate that both potency and selectivity can be affected by the nature of the acyl moiety and further demonstrate the limits of this approach as a potentiating strategy. Acylation of an antimicrobial peptide can have dramatic consequences on its structure, organization, and spectrum of cytolytic activity; and highly hydrophobic acyls were shown to have deleterious consequences on antibacterial properties. The data obtained with MSI-78 and LL37 support the view that acylation on short peptide sequences is most effective and that these results may hold true for at least many linear antimicrobial peptides.