Antimicrobial Activity of Short, Synthetic Cationic Lipopeptides

Authors


Corresponding author: Dr Brendan F. Gilmore, b.gilmore@qub.ac.uk

Abstract

The increasing emergence of multidrug-resistant micro-organisms presents one of the greatest challenges in the clinical management of infectious diseases. Therefore, novel antimicrobial agents are urgently required to address this issue. In this report, we describe the solid phase synthesis, characterization, microbiological and toxicological evaluation of a library of ultrashort cationic antimicrobial lipopeptides based on the previously described tetrapeptide amide H-Orn-Orn-Trp-Trp-NH2 conjugated with saturated fatty acids which have inherent antimicrobial activity. The microbiological activity of these ultrashort cationic lipopeptides, which exhibit excellent, broad-spectrum antimicrobial activity against a number of clinically important pathogenic bacteria and fungi, including multidrug resistant micro-organisms in both planktonic and sessile (biofilm) cultures is reported.

The increasing emergence of multidrug-resistant pathogenic micro-organisms has become one of the most pressing concerns in modern medicine. Incidences of nosocomial and community-acquired Staphylococcus aureus infections have risen dramatically in recent years (1), with almost 50% of hospital-acquired S. aureus infections exhibiting methicillin resistance and 30% of enterococci classified as vancomycin resistant (VRE) (2). Of equal concern is the dearth of new anti-infectives in development as a result of reduced research and development activity by major pharmaceutical companies since the mid-1980s (2,3). Furthermore, the extensive use of the limited classes of effective antibiotics currently available, in combination with the factors outlined earlier, constitutes a real and significant threat to the ‘antibiotic era’ (3) and to the effective management of bacterial infections. Therefore, there exists an urgent requirement for new antimicrobial agents with activity against pathogens that are resistant to the available armoury of antibiotics (3,4).

One class of compounds that has attracted increasing attention in the last two decades are the cationic antimicrobial peptides (AMPs). Antimicrobial peptides are short (typically ranging from 12 to 100 amino acid residues in length), generally exhibit rapid and efficient antimicrobial toxicity against a range of pathogens (5,6) and constitute critical effector molecules in the innate immune system of both prokaryotic and eukaryotic organisms (7). To date, over 1700 endogenous AMPs have been isolated with many more synthetic analogues reported in the literature (8). With such intensive industry directed towards studying this class of compounds, studies into their mechanism of action and structure activity relationship analyses (SAR) have yielded vital information relating to the structural features of effective AMPs, indicating that antimicrobial activity is governed primarily by charge and hydrophobicity (9) and that the initial target is the negatively charged bacterial/fungal cell membrane (10). These studies have also facilitated the design of ultrashort, highly active antimicrobial peptide scaffolds (11,12), which may be prepared via established facile, solid phase synthetic protocols at lower costs compared with their natural antimicrobial peptide counterparts (13). Therefore, ultrashort antimicrobial peptide scaffolds represent attractive lead compounds for the development of novel antimicrobial agents for multidrug resistant micro-organisms.

Antimicrobial lipopeptides are primarily bacterial compounds synthesized via non-ribosomal biosynthetic pathways and comprise a peptidyl portion conjugated to a fatty acid to form an acylated peptide (14). Lipopeptides may comprise an anionic (e.g., Daptomycin, surfactin) or cationic (e.g., polymixin B, colistin) peptide motif, which dictates the spectrum of activity (14). Fatty acids, fatty amines, alcohols and glyceryl esters have all been shown to exhibit varying degrees of antimicrobial activity (15,16), whilst acylation of peptide scaffolds has been demonstrated to significantly improve antimicrobial activity (14,17). Therefore, the combination of an optimized peptidyl scaffold and N-terminal acyl substituent (i.e. fatty acid) with inherent antimicrobial activity represents an approach to the development of potent antimicrobial agents, whereby spectrum of activity may be modulated via modification of the N-terminal substituent, whilst circumventing the commercial barriers associated with the manufacture of natural AMPs in high yields.

The work of Svendsen et al. defined the minimum antibacterial motif in cationic AMPs using lactoferrin derivatives, expressed in units of charge and bulk/lipophilicity in the peptide sequence (18). These studies have permitted the design of ultrashort AMPs (including dipeptides) bearing non-proteinogenic lipophilic/bulky amino acids (19). Recently, Makovitzki et al. reported the synthesis and evaluation of a number of ultrashort antibacterial and antifungal lipopeptides consisting of four l- and d-amino acids linked to fatty acids, which exhibited potent, broad-spectrum antimicrobial activity against a range of pathogenic plant and human pathogens (20,21). In addition, Bisht et al. reported the antimicrobial activity of a number of amino terminal modified tetrapeptides (12). In this report, we describe the solid phase synthesis, characterization, microbiological and toxicological evaluation of a library of ultrashort antimicrobial lipopeptides based on the Orn-Orn-Trp-Trp tetrapeptide motif previously described by Bisht et al. (12), conjugated with saturated fatty acids which have inherent antimicrobial activity (15,16). We report on the biological activity of these compounds, which exhibit excellent, broad-spectrum antimicrobial activity against a number of bacteria and fungi, including multidrug resistant micro-organisms in both planktonic and sessile (biofilm) cultures.

Materials and Methods

Materials

Rink amide 4-(2′,4′-dimethoxyphenyl-Fmoc-aminomethyl)-phenoxyacetamido-MHBA (MBHA) resin, all 9-fluorenylmethoxy carbonyl (Fmoc) l-amino acids (Fmoc-Orn(Boc)-OH and Fmoc-Trp(Boc)-OH) and 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU) were obtained from Merck Chemicals Ltd (Nottingham, UK). All fatty acids; hexanoic (caproic) acid, octanoic (caprylic) acid, decanoic (capric) acid, dodecanoic (lauric) acid, tetradecanoic (myristic) acid and hexadecanoic (palmitic) acid were obtained from Sigma-Aldrich (Dorset, UK). All other reagents/solvents were peptide synthesis grade.

Peptide synthesis

All peptides were synthesized using standard 9-fluorenylmethoxy carbonyl (Fmoc) solid phase protocols on Rink Amide MHBA resin, using a CEM Liberty (Buckingham, UK) microwave enhanced automated peptide synthesizer. Peptide elongation was effected using standard HBTU coupling chemistry in dimethylformamide (DMF) solvent with fourfold molar excess of diisopropyl ethylamine (DIEA) in N-methyl-2-pyrrolidone (NMP) and a threefold molar excess of each Fmoc-protected amino acid or free fatty acid. Fatty acids were coupled using standard (microwave enhanced) amino acid coupling conditions (18 W, 75 °C, 300 s) employed for all syntheses. All peptides were cleaved from the resin using 95% trifluoroacetic acid, 2.5% triisopropylsilane and 2.5% thioanisole (2 h, room temperature), following standard work-up (crude product was dissolved in ethyl acetate and subjected to a series of washes with 1 mM HCl (3 × 50 ml) and water (3 × 50 ml) and dried over anhydrous MgSO4), the identity of each lipopeptide was confirmed by electrospray mass spectroscopy (Thermo Finnigan LCQ Deca ion trap), as detailed in Table 1. Peptide purity was analysed by RP-HPLC using a Gemini C18, 250 × 4.6 mm column (Phenomonex, UK), a 2–60% acetonitrile gradient (30 min) in 0.05% TFA-water at a flow rate of 1 mL/min. All peptides/lipopeptides were found to have >90% purity.

Table 1.   Mass spectrometric analysis of peptides/lipopeptidesa
PeptideExpected massObserved mass (M + H)+
  1. aAll analyses were performed on a Thermo Finnigan LCQ Deca ion trap mass spectrometer (Thermo Fischer Scientific Inc.).

FMOC-OOWW-NH2839.41840.3
H-OOWW-NH2617.34618.4
C6-OOWW-NH2715.42716.4
C8-OOWW-NH2743.45744.4
C10-OOWW-NH2771.48772.3
C12-OOWW-NH2799.51800.5
C14-OOWW-NH2827.54828.4
C16-OOWW-NH2855.57856.3

Strains and growth conditions

The following strains were used in this study: Staphylococcus epidermidis ATCC 35984 (methicillin resistant, MRSE), Staphylococcus aureus ATCC 29213, Staphylococcus aureus (methicillin resistant, MRSA) ATCC 43300, Pseudomonas aeruginosa PA01, Escherichia coli NCTC 8196 and Candida tropicalis NCTC 7393. All microbial strains were stored at −70 °C in Microbank vials (Pro-Lab Diagnostics, Cheshire, UK) and subcultured in Müller Hinton broth (MHB) before testing.

Cell culture conditions

Human keratinocyte (HaCaT) cells, originally created by Dr. N. Fusenig, German Cancer Research Centre (Heidelberg, Germany) were purchased from Cell Line Service (Eppelheim, Germany). Cells were cultured in DMEM containing 4500 mg/L d-glucose (Invitrogen), supplemented with 2 mM l-glutamine, foetal bovine serum and 1% penicillin/streptomycin. Cells were grown at 37 °C and 5% CO2 with cells subcultured at 80–90% confluency. Cell monolayers were rinsed with phosphate buffered saline (PBS) and treated with Trypsin–EDTA (Invitrogen, Paisley, UK) to detach cells before resuspension in fresh medium.

Minimum inhibitory /cidal concentration determination

Broth microdilution tests were carried out according to NCCLS guidelines (22). A working solution of each peptide/lipopeptide was prepared in MHB and 0.22-μm sterile filtered. From this stock solution, serial twofold dilutions in MHB were carried out in 96-well microtitre plates (maximum concentration 1000 μg/mL). Micro-organisms under investigation were grown over 18–24 h at 37 °C in MHB, from which an inoculum was taken and adjusted to optical density 0.3 at 550 nm which is the equivalent to 1 × 108 CFU/mL. This suspension was further diluted to give a final inoculum density of 1 × 106 CFU/mL, as verified by total viable count. All controls and test concentrations were prepared as six replicates. The microtitre plates were then incubated for 24 h at 37 °C in a stationary incubator. Following determination of the MIC for each compound, the minimum bactericidal concentrations (MBC) were derived by transferring 20 μl of the suspension from the wells which displayed no signs of growth to MHA plates. The MHA plates were then incubated in a stationary incubator at 37 °C for 24 h and examined for 99.9% killing.

Biofilm susceptibility assay

Biofilm susceptibility assays were performed using the Calgary Biofilm Device [MBEC Assay for Physiology & Genetics (P&G), Innovotech Inc., Alberta, Canada], essentially according to the manufacturers instructions with some slight modifications as previously described (23). Positive and negative controls (6 replicates) were included in each challenge plate alongside serial twofold dilutions of the compound under test (6 replicates). After exposure of 24 h biofilms to challenge concentrations of peptides/lipopeptides, the recovery plate was incubated for a further 24 h at 37 °C and visually checked for turbidity, clear wells indicating biofilm eradication. The lowest concentration at which no observable growth was apparent after 24 h was designated the minimum biofilm eradication concentration (MBEC). For verification purposes, optical density in each well was recorded at 550 nm and compared with negative (no-growth) control. Recovery plates were incubated for a further 24 h at 37 °C to confirm MBECs.

Haemolysis assay

Fmoc-OOWW-NH2, H2N-OOWW-NH2 and the least (C6-OOWW-NH2) and most (C12-OOWW-NH2) potent lipopeptides were assayed spectrophotometrically for their ability to induce haemoglobin release from fresh equine erythrocytes according to the method of Shin et al. (24). Fresh defibrinated equine erythrocytes (Laboratory Supplies & Instruments Ltd, Antrim, UK), were washed three times with equal volumes of PBS. After centrifugation for 15 min at 900 × g, erythrocytes were resuspended in 4% (v/v) in PBS. Equal volumes (100 μl) of the erythrocyte suspension were added to each well of a 96-well microtitre plate. Erythrocytes were subsequently exposed to selected peptide/lipopeptide concentrations, incubated at 37 °C for 1 h and centrifuged at 1000 × g. Aliquots of the supernatant were transferred to a fresh 96-well microtitre plate, and haemoglobin release measured spectrophotometrically at 405 nm. As a positive control (100% haemolysis), erythrocytes were treated with 0.1% Triton X-100, whilst PBS (0% haemolysis) acted as a negative control. All peptide/lipopeptide concentrations are reported as the mean of six replicates. Percentage haemolysis was calculated as follows:

image

Cell survival analysis

Cell viability was assessed by means of a quantitative MTT assay, using a modification of the method of Mossman (25). Human keratinocyte (HaCaT) cells were cultured (until at least third passage) and inoculated into 96-well tissue culture treated microtitre plates at a concentration of 1 × 104 cells/well and incubated at 37 °C for 24 ± 1 h, until approximately 90% confluency. After this time, the medium was removed and replaced with fresh growth medium containing dilutions of H-OOWW-NH2 and C12-OOWW-NH2 at final concentrations of 10, 50, 100 and 200 μg/mL. Peptide/lipopeptide treatments were incubated 24 ± 1 h, after which 10 μg/mL 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) reagent (Sigma) was added to each well and cells incubated, as described, for a further 2 h. Culture medium was removed, and the resulting insoluble intracellular formazan crystals dissolved in 200 μl of DMSO. Absorption was measured at 570 nm in a Biotek EL808 spectrophotometer (Labtech International Inc., UK). Percentage cell viability was calculated relative to untreated control wells after subtraction of the blank value corresponding to DMSO dissolved untreated cells in the absence of MTT reagent.

Results and Discussion

The antimicrobial activity (MIC/MBC values) of the Fmoc-protected peptide (Fmoc-Orn-Orn-Trp-Trp-NH2), deprotected tetrapeptide amide (H-Orn-Orn-Trp-Trp-NH2) and N-acyl-Orn-Orn-Trp-Trp-NH2 (Cn-Orn-Orn-Trp-Trp-NH2, where = 6, 8, 10, 12, 14 or 16) lipopeptides against a range of pathogenic micro-organisms are detailed in Table 2. All peptides and lipopeptides exhibited excellent, broad-spectrum activity against the range of organisms tested. The antimicrobial activity is clearly influenced by the nature of the N-terminal substituent, because the Fmoc-protected peptide exhibited significantly higher antimicrobial activity compared with the deprotected tetrapeptide amide. The MIC values determined for the tetrapeptide amide against Gram-positive and Gram-negative organisms is in keeping with those originally determined by Bisht et al. (12) and comparable to acylated SC4 peptide derivatives described by Lockwood et al. (26). For each of the lipopeptides synthesized in this study, the antimicrobial activity was dependent on the number of carbon atoms in the N-acyl substituent (Figure 1), with tetrapeptides bearing N-terminal C12 (dodecyl) acyl substituents exhibiting optimal antimicrobial potency against all organisms tested. The lipopeptide C14-Orn-Orn-Trp-Trp-NH2 exhibited similar antimicrobial potency against Gram-positive organisms compared with the C12-modified peptide, as shown in Figure 2A. However, for both Gram negatives and fungi, N-acyl substituents of 14 and 16 carbon atoms resulted in decreased antimicrobial activity, as shown in Figure 2B.

Table 2.   Antimicrobial activity of N-terminally modified peptides/lipopeptides CnOOWW-NH2 (where = 6, 8, 10, 12, 14, 16). Minimum inhibitory concentrations (MIC) and minimum bactericidal/fungicidal concentrations (MBC/MFC) are given in μg/mL and are quoted as the mean of 6 replicates
  FmocNH2C6C8C10C12C14C16
S. epidermidis (MRSE) ATCC 35984MIC15.6312512531.257.811.951.957.81
MBC15.6325025031.257.811.951.9515.63
S. aureus ATCC 29213MIC3.9112562.515.633.910.951.957.81
MBC7.8125062.531.257.811.951.9515.63
S. aureus ATCC 43300 (MRSA)MIC7.8125012531.257.811.951.953.91
MBC7.8125050031.2515.633.913.9131.25
P. aeruginosa PA01MIC12525012531.257.811.9531.2531.25
MBC25050025012515.6315.6331.2562.5
E. coli NCTC 8196MIC62.550012562.515.637.8131.2562.5
MBC62.550025012515.6315.6331.25125
C. tropicalis NCTC 7393MIC15.612562.515.637.811.953.9115.63
MFC31.2525062.515.637.813.913.9115.63
Figure 1.

 Structure of the most potent antimicrobial lipopeptide synthesized in this study, C12-Orn-Orn-Trp-Trp-NH2.

Figure 2.

 (A) Minimum inhibitory concentrations for lipopeptides N-terminally modified peptides/lipopeptides CnOOWW-NH2 (where = 6, 8, 10, 12,14, 16) against Gram-positive organisms. (B) Minimum inhibitory concentrations for N-terminally modified peptides/lipopeptides, CnOOWW-NH2 (where = 6, 8, 10, 12, 14, 16), against Gram-negative organisms and a representative fungi, C. tropicalis.

This general trend is in keeping with earlier studies into the activity of saturated fatty acids, where lauric acid (dodecanoic acid) was found to possess the highest antimicrobial activity against both methicillin sensitive and methicillin-resistant S. aureus (16), and also for a range of antimicrobial N-alkylquinolinium bromide ionic liquids (27), where optimal activity was observed with the C12 substituent and activity decreased in compounds with higher homologues. Therefore (in addition to the hydrophobicity and charge of the tetrapeptide amide), hydrophobicity of the N-acyl substituent is a key determinant of antimicrobial activity for the lipopeptides described here. The N-terminal modification of the tetrapeptide amide increases the antimicrobial potency of both the tetrapeptide amide (12) and fatty acid (16), in keeping with previous observations (14,17). Indeed, antimicrobial activity of lauric acid (16) was increased by 200- and 400-fold against S. aureus and MRSA, respectively, when conjugated to the tetrapeptide amide.

The antibiofilm activity, or the determined MBEC, of each peptide/lipopeptide is shown in Table 3. For comparison, MIC values are shown alongside MBEC values in Table 3. The MBEC is the minimum concentration of an antimicrobial agent required to eradicate a microbial biofilm (28) and is determined for each antimicrobial peptide/lipopeptide. Interestingly, biofilms of Gram-positive organisms were susceptible to eradication by relatively low concentrations of lipopeptides (compared with corresponding MIC values), with antibiofilm activity increasing with length of the acyl substituent up to = C12/C14/C16, which were essentially equivalent in biofilm eradication activity. However, biofilms of P. aeruginosa did not exhibit susceptibility to any of the peptides/lipopeptides tested up to a concentration of 1000 μg/mL. E. coli biofilms were only eradicated by C12-Orn-Orn-Trp-Trp-NH2 at a concentration of 500 μg/mL (a 32-fold increase in MBEC compared with S. epidermidis and an eightfold increase compared with MRSA). C. tropicalis biofilms were eradicated by both C10- and C12-Orn-Orn-Trp-Trp-NH2 at concentrations of 250 μg/mL. The lack of sensitivity of Gram-negative bacterial biofilms to antimicrobial lipopeptides may be because of binding of the positively charged peptide moiety to negatively charged biofilm matrix polymeric material which may retard the penetration of these antimicrobial agents into the biofilm, because similar effects have been previously described with positively charged aminoglycosides and Pseudomonas biofilms (29–32) and the antiseptic chlorhexidine and oral biofilms (33). Furthermore, we have obtained similar results with previously reported lipopeptides C16-KKKK, C16-KAAK and C16-KGGK (19), with MBEC values against Pseudomonas aeruginosa PA01 in each case >1000 μg/mL (data not shown).

Table 3.   Antibiofilm activity of N-terminally modified peptides/lipopeptides CnOOWW-NH2 (where = 6, 8, 10, 12, 14, 16). Minimum inhibitory concentrations (MIC) and minimum biofilm eradication concentrations (MBECs) are given in μg/mL and are quoted as the mean of 6 replicates
  FmocNH2C6C8C10C12C14C16
S. epidermidis (MRSE) ATCC 35984MIC15.6312512531.257.811.951.957.81
MBEC63.5500>1000>100025015.6315.6315.63
S. aureus ATCC 29213MIC3.9112562.515.633.910.951.957.81
MBEC500500>1000>100050062.531.2562.5
S. aureus ATCC 43300 (MRSA)MIC7.8125012531.257.811.951.953.91
MBEC25050010001000100062.562.562.5
P. aeruginosa PA01MIC12525012531.257.811.9531.2531.25
MBEC>1000>1000>1000>1000>1000>1000>1000>1000
E. coli NCTC 8196MIC62.550012562.515.637.8131.2562.5
MBEC>1000>1000>1000>1000>1000500>1000>1000
C. tropicalis NCTC 7393MIC15.612562.515.637.811.953.9115.63
MBEC>1000>1000>1000>1000250250>1000>1000

To evaluate the potential of these lipopeptides to cause cellular toxicity, the ability of a range of lipopeptides (and the tetrapeptide amide) to cause lysis of equine erythrocytes was assayed by colorimetric haemolysis assay, the results of which are shown in Figure 3. Neither the tetrapeptide amide nor the lipopeptide, C6-Orn-Orn-Trp-Trp-NH2 exhibited significant haemolysis even at the highest concentration tested (200 μg/mL), whereas treatment of erythrocytes with C12-Orn-Orn-Trp-Trp-NH2 resulted in significant lysis at concentrations greater than 50 μg/mL (>98.7% ± 0.9%). However, at 10 μg/mL, C12-Orn-Orn-Trp-Trp-NH2 exhibited low haemolytic activity (9.8% ± 1.8%), indicating that this lipopeptide could be safely used at its MIC values for all organism tested in this study. In keeping with its microbiological toxicity profile, the lipopeptide C16-Orn-Orn-Trp-Trp-NH2 exhibited significantly lower haemolytic activity compared with the dodecyl derivative. These data indicate that conjugation of lipid moieties to peptide motifs has the effect of increasing biological toxicity generally and despite apparently increasing the spectrum of activity of the original antimicrobial peptide, may limit their utility when used at, for example, biofilm eradication concentrations. Similar data were obtained with the tetrapeptide amide and C12-Orn-Orn-Trp-Trp-NH2 in human keratinocyte cell viability (MTT) assays, as shown in Figure 4. Once again, the lipopeptide C12-Orn-Orn-Trp-Trp-NH2 exhibited significant cytotoxicity at concentrations greater than 62.5 μg/mL (MIC range 0.95–7.81 μg/mL), whereas human keratinocyte cells retained 71.6% ± 7.8% viability when treated with the tetrapeptide amide at a maximum concentration of 1000 μg/mL. Taking the haemolysis and cell viability data together, the likely mode of action of these lipopeptides is by a direct membrane disruption, in both prokaryotic and eukaryotic cells.

Figure 3.

 Haemolytic activity of the tetrapeptide amide and lipopeptides CnOOWW-NH2 (where = 6, 12 and 16) against equine erythrocytes. Each value is expressed as the mean of six replicates.

Figure 4.

 Cytotoxicity of the tetrapeptide amine and the most potent antimicrobial lipopeptide, C12OOWW-NH2 evaluated against human keratinocyte (HaCaT) cells. Each value is expressed as the mean of six replicates.

Conclusion

In summary, the modification of the tetrapeptide amide H-Orn-Orn-Trp-Trp-NH2 with N-acyl substituents increased antimicrobial potency against all organisms tested. The lipopeptides synthesized in this study exhibited excellent broad-spectrum antimicrobial toxicity, with N-acyl substituents of 12–14 carbon atoms in length exhibiting the highest degree of antimicrobial and antibiofilm activity. However, N-terminal modification with fatty acids also increased their general biological toxicity (haemolytic/cytotoxic activities), indicating that hydrophobicity is a key determinant of biological activity and that membrane disruption is primary mechanism of biological toxicity.

Ancillary