Dilipid ultrashort cationic lipopeptides as adjuvants for chloramphenicol and other conventional antibiotics against Gram‑negative bacteria

Ronald Domalaon1 · Marc Brizuela1 · Benjamin Eisner1 · Brandon Findlay1 · George G. Zhanel2 · Frank Schweizer1,2

Abstract

The necessity to develop therapeutic agents and strategies to abate the spread of antibiotic-resistant pathogens is prominent. Antimicrobial peptides (AMPs) provide scaffolds and inspiration for antibiotic development. As an AMP of shorter scaffold, eight dilipid ultrashort cationic lipopeptides (dUSCLs) were prepared consisting of only four amino acids and varying dilipids. Lipids were acylated at the peptide N-terminus and the ε-amine side chain of the N-terminal l-lysine. Compounds that possess aliphatic dilipids of ≥ 11 carbons-long showed significant hemolysis and therefore limited therapeutic application. Several non-hemolytic dUSCLs were identified to enhance the activity of chloramphenicol and other conventional antibiotics against Gram-negative bacteria. Compounds 2 and 6 have a short peptide sequence of KKKK and KKGK, respectively, and are both acylated with an aliphatic dilipid of nine carbons-long potentiated chloramphenicol against MDR clinical isolates of Pseudomonas aeruginosa, Acinetobacter baumannii and Enterobacteriaceae. Both dUSCLs showed comparable adjuvant potency in combination with chloramphenicol. However, dUSCL 2 synergized with a wider span of antibiotic classes against P. aeruginosa relative to dUSCL 6 that included rifampicin, trimethoprim, minocycline, fosfomycin, piperacillin, ciprofloxacin, levofloxacin, moxifloxacin, linezolid and vancomycin. Our data revealed that dUSCLs can indirectly disrupt active efflux of chloramphenicol in P. aeruginosa. This along with their membrane-permeabilizing properties may explain the dUSCLs synergistic combination with conventional antibiotics against Gram-negative bacteria.

Keywords Antimicrobial peptide · Dilipid ultrashort cationic lipopeptides · Combination therapy · Chloramphenicol ·

Introduction

Antimicrobial peptides (AMPs) present an attractive reservoir of potential therapeutic agents to treat infectious diseases (Domalaon et al. 2016). These naturally occurring AMPs are produced by animals and bacteria to defend themselves from invading pathogens. Most AMPs, especially those with inherent overall positive charge at physiological pH, eradicate bacteria via membrane permeabilization and cell lysis (Fjell et al. 2011; Juhaniewicz-Dębińska et al. 2018). However, protease instability limits their therapeutic usage. Moreover, the sheer size of typical AMPs (normally ≥ 10 amino acids-long) confer disadvantage on their production cost. Ultrashort cationic lipopeptides (USCLs) mimic the bioactivity of longer AMPs through a small peptide composed of ≤ 5 amino acids that is acylated at the N-terminus (Makovitzki et al. 2006), providing drug candidates with a relatively short scaffold to optimize. An overall amphiphilic character is crucial to the activity of AMPs and thus incorporated in the USCL design through the use of hydrophilic amino acids and hydrophobic lipids.
Several structure–activity relationship studies have been reported by our group and others that provide meaningful insights to guide the development of bioactive USCLs. For instance, the presence of at least two protonatable amine groups (perceived to be protonated at physiological pH and therefore bestow cationic charge) was found crucial to yield agents with antibacterial activity (Greber et al. 2017). Constraining the typically flexible amino acid side chain in a rigidified cyclic ring seemed to be detrimental for activity (Domalaon et al. 2014). Hydrophobic lipids of ≥ 14 carbons-long appeared necessary to eradicate pathogens (Findlay et al. 2012). However, substantial hemolysis of red blood cells occurred with lipids of ≥ 16 carbons-long (Findlay et al. 2012). In an attempt to adjust hydrophobicity, dilipid USCLs (dUSCLs) have been developed possessing two shorter instead of one longer lipid to modulate antibacterial activity relative to unwanted toxicity. So far, several membrane-acting dUSCLs have been reported to possess potent antibacterial activity against Gram-positive and Gram-negative bacteria with little hemolytic activity (Greber et al. 2014; Ahn et al. 2014).
Combination therapy of AMPs and conventional antibiotics has been identified as a viable strategy to eradicate antibiotic-resistant pathogens (Steenbergen et al. 2009; Jorge et al. 2017). In this approach, the antibacterial activity of AMPs may work in synergy with an antibiotic and thus achieve enhanced bacterial killing. However, it is also possible that AMPs with limited activity may still help maximize antibiotic potency by allowing enhanced membrane permeation of the antibiotic resulting in an increased intracellular accumulation. The naturally occurring 18-residue cationic peptide novicidin was reported to synergize with rifampicin, ceftriaxone and ceftazidime against antibiotic-resistant Enterobacteriaceae (Soren et al. 2015). Other naturally occurring AMPs such as nisin Z and pediocin PA-1 were also described to potentiate several classes of antibiotics, including chloramphenicol, against antibiotic-resistant Pseudomonas fluorescens (Naghmouchi et al. 2012). Synthetic AMPs that act on bacterial membranes were also reported in combination with antibiotics. For instance, the synthetic 18-residue leucine–lysine-rich peptide P5 was found to synergize with chloramphenicol against Gram-positive and Gramnegative bacteria (Park et al. 2006). Moreover, the synthetic 26-residue α-helical peptide PL5 was demonstrated to enhance the efficacy of levofloxacin in a Staphylococcus aureus wound infection mouse model (Feng et al. 2015). Since most AMPs act on bacterial membranes, synergy with antibiotics was likely due to membrane permeabilization and disorganization that lead to an increased intracellular antibiotic concentration.
Most reported AMPs that enhanced the activity of conventional antibiotics have been composed of ≥ 10 amino acids. However, we recently demonstrated that even a short proline-rich lipopeptide, consisting of only seven amino acids linked to a 12 carbons-long lipid, can synergize with rifampicin and minocycline against multidrug-resistant
Fig. 1 Chemical structure of dilipid ultrashort cationic lipopeptides (dUSCLs) in this study. Aliphatic lipids used ranged from seven (C7) to fourteen (C14) carbons-long. Peptide N-terminus and ε-amine side chain of l-lysine at position 1 were acylated with various lipids while peptide C-terminus was amidated
(MDR) Pseudomonas aeruginosa clinical isolates (Domalaon et al. 2018c). We wondered whether even shorter AMPs such as USCLs have the ability to potentiate conventional antibiotics against antibiotic-resistant bacteria. Herein, we report the development of new dUSCLs with varying lipid component (Fig. 1) and their antibacterial evaluation against Gram-positive and Gram-negative pathogens. More importantly, we demonstrate that dUSCLs can potentiate chloramphenicol and other conventional antibiotics against wild-type and MDR clinical isolates of Gram-negative bacteria.

Materials and methods

Peptide synthesis

All dUSCLs were prepared on a solid-support 4-methylbenzhydrylamine (MBHA) Rink amide resin following standard fluorenylmethyloxycarbonyl (Fmoc) protection strategy (Domalaon et al. 2014). All ε-amine side chain of l-lysine were protected with tert-butyloxycarbonyl (Boc) with the exception of l-lysine at amino acid position 1, which was masked with Fmoc. Peptide coupling was done using the coupling reagent O-(benzotriazol-1-yl)N,N,N′,N′-tetramethyluronium tetrafluoroborate (TBTU) (3 molar equivalent) and the weak base N-methylmorpholine (8 molar equivalent), which were reacted with the resin for 45 min per coupling steps. The dilipid was acylated at the peptide N-terminus and the ε-amine side chain of l-Lysine at position 1 via peptide coupling reaction. Once the solid-phase peptide synthesis was completed, dUSCLs were cleaved from the resin using an acidic solution of trifluoroacetic acid:water (95:5 v/v), which were reacted for 30 min followed by removal of solvent in vacuo. The dUSCLs were purified using reverse-phase flash chromatography with C18-functionalized silica gel (40–63 µm) obtained from Silicycle (USA), the solvents used were a gradient mixture of water and methanol (both spiked with 0.1% trifluoroacetic acid). All purified dUSCLs were in their trifluoroacetic acid salt form. Purity was measured using high-performance liquid chromatography (HPLC) on a Breeze HPLC Waters with 2998 PDA detector (1.2 nm resolution) coupled to Phenomenex Synergi Polar (50 × 2.0 mm) 4 µm reverse-phase column and were determined to be > 95%. Characterization of dUSCLs was achieved using nuclear magnetic resonance (NMR) and mass spectrometry (MS). One- (1H and 13C) and twodimensional NMR experiments were done on a Bruker AMX-500 (Germany). Electrospray ionization mass spectrometry (ESI-MS) experiments were done on a Varian 500-MS ion trap mass spectrometer (USA).

Antimicrobial susceptibility assay

Bacterial isolates used in this study were obtained from the American Type Culture Collection (ATCC), the Canadian National Intensive Care Unit (CAN-ICU) surveillance study (Zhanel et al. 2008) and the Canadian Ward (CANWARD) surveillance study (Hoban and Zhanel 2013).
Clinical isolates belonging to the CAN-ICU and CANWARD surveillance studies were recovered from patients suffering presumed infectious diseases entering or admitted in a participating medical center across Canada during the time of study. Efflux-deficient P. aeruginosa PAO200 and PAO750 strains were generously provided by Dr. Ayush Kumar (University of Manitoba). All pharmaceutical-grade antibiotics and reagents used were purchased either from Sigma-Aldrich or AK Scientific.
Microbroth dilution susceptibility testing was done according to the Clinical and Laboratory Standards Institute (CLSI) guidelines (The Clinical and Laboratory Standards Institute 2016) to evaluate the in vitro antibacterial activity of dUSCLs. Briefly, overnight-grown bacterial culture was diluted in saline to achieve a 0.5 McFarland turbidity. The resulting bacterial solution was then diluted 1:50 in Mueller–Hinton broth (MHB) for inoculation to a final concentration of 5 × 105 colony-forming units/mL. The assay was performed on a 96-well plates where the agents were twofold serially diluted in MHB and incubated with equal volumes of inoculum for 18 h at 37 °C. Antibacterial activity of the agents was determined by their minimum inhibitory concentration or the lowest concentration to inhibit visible bacterial growth in form of turbidity, which was inspected visually and further confirmed using EMax Plus microplate reader (Molecular Devices, USA) at a wavelength of 590 nm. Wells containing MHB with or without bacterial cells were used as positive or negative control, respectively.

Hemolytic assay

The ability of dUSCLs to lyse eukaryotic cells was quantified by the amount of hemoglobin released upon incubation with pig red blood cells, following published protocols (Findlay et al. 2012). Fresh pig blood that was drawn from the antecubital vein was centrifuged at 1000×g for 5 min at 4 °C, washed with phosphate-buffered saline (PBS) three times and re-suspended in the same buffer, consecutively, to prepare the working erythrocyte solution. The agents were then twofold serially diluted in PBS on 96-well plate and mixed with equal volumes of working erythrocyte solution. Post 1-h incubation at 37 °C, intact cells on the 96-well plate were pelleted by centrifugation at 1000×g for 5 min at 4 °C. The supernatant was then transferred to a new 96-well plate. The released hemoglobin was then measured via EMax Plus microplate reader (Molecular Devices, USA) at 570 nm wavelength. Red blood cells in PBS with or without 0.1% Triton X-100 was used as negative or positive control, respectively.

Checkerboard assay

The assay was performed on a 96-well plate as previously described (Domalaon et al. 2018a). The antibiotic was twofold serially diluted along the x axis, while the adjuvant was twofold serially diluted along the y axis to create a matrix in which each well contain a combination of both agents at different concentrations. Overnight-grown bacterial culture was diluted in saline to 0.5 McFarland turbidity, followed by 1:50 dilution in MHB and inoculation on each well to a final concentration of approximately 5 × 105 colony-forming units/mL. Wells comprising of only MHB with or without bacterial cells were used as positive or negative control, respectively. The plate was incubated for 18 h at 37 °C and inspected for visible turbidity, which was confirmed using EMax Plus microplate reader (Molecular Devices, USA) at a wavelength of 590 nm. Fractional inhibitory concentration (FIC) of antibiotic was calculated by dividing the MIC of antibiotic in the presence of adjuvant by the MIC of antibiotic alone. Likewise, the FIC of adjuvant was calculated via dividing the MIC of adjuvant in the presence of antibiotic by the MIC of adjuvant alone. The FIC index was obtained by the summation of both FIC values. The FIC index was interpreted as synergistic, indifferent or antagonistic for values of ≤ 0.5, 0.5 < x≤4 or > 4, respectively (Meletiadis et al. 2010).

Results and discussion

Dilipid ultrashort cationic lipopeptides (dUSCLs) acylated with two aliphatic lipids

Adequate lipid hydrophobicity is crucial for USCLs to effectively lyse bacterial cells (Findlay et al. 2012). This selective action is due to the cationic peptide portion of USCL that electrostatically interact with the anionic bacterial membranes over zwitterionic eukaryotic membranes. However, too much hydrophobic moment leads to non-specific membrane lysis of eukaryotic cells such as red blood cells resulting in toxicity. To work around this conundrum, dUSCLs are prepared by acylating two shorter instead of one longer lipid to maximize hydrophobic input for better antibacterial activity while minimizing the propensity for toxicity (Dawgul et al. 2017).
All the dUSCLs were synthesized via solid-phase peptide synthesis following an Fmoc strategy on an MBHA Rink amide resin. We incorporated our previously reported USCL tripeptide sequence of KKK and KGK (Findlay et al. 2012), where K indicate l-lysine and G indicate glycine, into the dUSCL design (Fig. 1) and added another K at amino acid position 1 (K1) as a point of attachment of the dilipid. The peptide N-terminus and ε-amine side chain of K1 were then acylated with various lipids (Fig. 1) consisting of seven ( C7) to fourteen (C14) carbons-long, affording us eight dUSCLs. As a result of using an MBHA-based resin, the C-terminus of dUSCLs was amidated.
The antibacterial activity of dUSCLs was evaluated against Gram-positive and Gram-negative bacteria (Table 1), most of which were MDR clinical isolates. Minimum inhibitory concentration (MIC) or the lowest concentration of the agent that can inhibit bacterial growth was assessed. Out of the eight dUSCLs, compound 3 displayed good broad spectrum activity (MIC of 8 µg/mL) against all Gram-positive bacteria and Escherichia coli isolates in the panel. Moderate activity (MIC of 32 µg/mL) was observed for dUSCL 3 against P. aeruginosa strains, while limited activity was found against the rest of Gram-negative bacteria in the panel. The peptide sequence KKKK appeared to be better as dUSCLs with KKGK sequence displaying relatively poor antibacterial activity (Table 1). For instance, dUSCL 7 displayed poor activity (MIC of 64 or 128 µg/mL) against Gram-positive bacteria and E. coli even though it was a direct counterpart of 3, both consisting of eleven carbons-long ( C11) dilipids but different peptide sequences (Fig. 1). This difference in activity may be attributed to the number of protonatable amine side chain groups (thus, possible cationic charges) in KKKK (+3 charges) relative to KKGK (+2 charges) sequences. It should be noted that the ε-amine side chain of K1 was acylated and therefore did not have a protonatable group. Compound 2 demonstrated moderate activity (MIC of 16–32 µg/mL) against Gram-positive bacteria only. The length of the dilipid was found to influence antibacterial activity. Hydrophobicity in dUSCLs must reach a certain threshold and lipid chains should contain nine ( C9) to eleven (C11) carbons to generate favorable antibacterial activity (Table 1).

Dilipids of longer length resulted in hemolysis

To address the concern that augmented hydrophobicity may lead to non-specific cell lysis, we evaluated the propensity of dUSCLs to lyse eukaryotic red blood cells by measuring the amount of heme release upon addition at 50, 100 and 500 µg/mL of peptide (Table 2). As expected, dUSCLs comprising of longer aliphatic lipids elicited high hemolysis. For instance, compounds 4 and 8 that both contain C 14 dilipid resulted in 42.10% and 30.50% hemolysis, respectively, at 100 µg/mL. The most potent dUSCL 3 appeared to be too hemolytic for therapeutic use, as 23.72% hemolysis was observed at 100 µg/mL. However, dUSCLs with shorter dilipid of C7 to C 9 did not produce appreciable hemolysis at 100 µg/mL. For instance, 100 µg/mL of dUSCL 2 and 6 that both consist of C9 dilipid resulted in only 2.07% and 3.78% hemolysis.

dUSCLs synergized with chloramphenicol against wild‑type P. aeruginosa

We then evaluated the dUSCLs as adjuvants in combination with conventional antibiotics against Gram-negative bacteria. Adjuvants are biomolecules used in combination therapy that enhance antibacterial efficacy of antibiotics by either increasing their intracellular accumulation (e.g., membrane permeabilizers and efflux-pump inhibitors), preventing their inactivation (enzyme inhibitors) or targeting other pathways that incapacitate bacteria to resist their effect (anti-virulence and two-component system inhibitors) (Domalaon et al. 2018b). A synergistic adjuvant–antibiotic combination is expected to exhibit enhanced bacterial killing relative to monotherapy. We decided to explore the effect of dUSCLs on the antibacterial activity of chloramphenicol against P. aeruginosa. Chloramphenicol is a bacteriostatic antibiotic that inhibits bacterial protein synthesis and is on the World Health Organization’s list of essential medicines (World Health Organization 2015). Resistance to chloramphenicol is commonly attributed to overexpression of efflux pumps and inactivating enzymes (such as chloramphenicol acetyltransferases) but as well as decrease in membrane permeability, especially in Gram-negative pathogens (Schwarz et al. 2004).
Several large amphiphilic AMPs have been reported to enhance the antibiotic efficacy of chloramphenicol against Gram-negative bacteria (Park et al. 2006; Naghmouchi et al. 2012). Therefore, we were curious whether dUSCLs can also exhibit this property. Interactions between dUSCLs and chloramphenicol were assessed via fractional inhibitory concentration (FIC) index, to which values of ≤ 0.5, 0.5 < x≤4 and > 4 were interpreted as synergistic, additive and antagonistic interactions, respectively. We identified five out of eight dUSCLs that synergized with chloramphenicol against wild-type P. aeruginosa PAO1 (Table 3). Compound 2 appeared to be the most promising adjuvant as it potentiated chloramphenicol the best (FIC index of 0.094) and that it induced low red blood cell hemolysis (Table 2). Furthermore, the MIC of chloramphenicol was reduced 32-fold (from 32 to 1 µg/mL) in the presence of only 8 µg/ mL (7 µM) of dUSCL 2. From here on, we set the adjuvant working concentration to 8 µg/mL (< 10 µM) as this concentration is typically used for adjuvants in combination studies (Domalaon et al. 2018c; Park et al. 2006).
Similar to the trend observed for antibacterial activity, there appeared to be a hydrophobic threshold needed for dUSCL to exhibit adjuvant property. Dilipid components of C9–C11 appeared to be optimal for dUSCLs to synergize with chloramphenicol. Interestingly, the peptide sequence seemed to affect the ability of dUSCLs to act as adjuvant. Compound 6, which was the counterpart of 2 as both consist of the same dilipid components, displayed synergism with chloramphenicol but at a relatively higher FIC index of 0.281. The only difference between dUSCL 2 (KKKK) and dUSCL 6 (KKGK) was their peptide sequences. Hereon, we assessed both dUSCLs for their ability to enhance conventional antibiotics in Gramnegative bacteria. Synergy between dUSCL and chloramphenicol was retained against MDR clinical isolates of P. aeruginosa
Synergism between dUSCLs and chloramphenicol was further assessed against five MDR clinical isolates of P. aeruginosa. Chloramphenicol alone had very limited activity (MIC of 128 to > 512 µg/mL) against the five MDR P. aeruginosa strains (Table 4). Similar to the results observed against the wild-type P. aeruginosa strain PAO1, both compounds 2 and 6 potentiated chloramphenicol against all MDR P. aeruginosa strains tested (Table 4). For instance, 8 µg/mL (7 µM) of both dUSCLs 2 and 6 reduced the MIC of chloramphenicol 128-fold (from 128 to 1 µg/mL) against MDR P. aeruginosa PA260-97103. No difference in the degree of chloramphenicol potentiation against tested MDR P. aeruginosa clinical isolates was observed for both dUSCLs 2 and 6, and therefore the slight difference in peptide sequence (KKKK vs KKGK) appeared not to play a role in their adjuvant property against clinical isolates.

dUSCLs can disrupt active efflux of chloramphenicol in P. aeruginosa

While amphiphilic AMPs and dUSCLs are known to permeabilize bacterial membranes (Naghmouchi et al. 2012; Ahn et al. 2014) that may result in synergism with conventional antibiotics, we wondered whether dUSCLs can also affect active efflux. This is especially important since the intracellular chloramphenicol concentration is significantly affected by efflux (Schwarz et al. 2004). In P. aeruginosa, chloramphenicol is a known substrate of MexAB-OprM, MexCD-OprJ and MexXY-OprM efflux systems (Masuda et al. 2000). To study the effect of efflux, we compared the potentiation of chloramphenicol by dUSLCs 2 and 6 against wild-type PAO1 and two efflux-deficient (PAO200 and PAO750) P. aeruginosa strains (Fig. 2). P. aeruginosa PAO200 lacked the MexAB-OprM efflux pump, while P. aeruginosa PAO750 lacked five clinically relevant pumps (MexAB-OprM, MexCD-OprJ, MexEF-OprN, MexJK and MexXY) and outer membrane protein OpmH. As expected, chloramphenicol was greatly affected by efflux as we observed a 32-fold decrease in MIC from wild-type PAO1 (MIC of 32 µg/mL) to both efflux-deficient PAO200 and PAO750 strains (MIC of 1 µg/mL against both strains) (see Online Resource Table S1). Interestingly, the synergism found against wild-type PAO1 was not observed against PAO200 nor PAO750 for the combination of dUSCLs and chloramphenicol (Fig. 2). It appeared that knocking out these efflux systems in P. aeruginosa negated the ability of dUSCLs to enhance the antibacterial activity of chloramphenicol. This strongly suggests that amphiphilic dUSCLs can disrupt active efflux in P. aeruginosa that may result in potentiation of chloramphenicol. While further biochemical studies are needed, these membrane-acting dUSCLs suggestively may block efflux indirectly by either (1) sequestering lipids surrounding the transmembrane protein of efflux pumps that result in conformational change and inactivation or (2) by affecting the proton motive force in the inner membrane required to energize these efflux systems. We also do not discredit the possibility of these dUSCLs to interact directly with efflux pumps by “clogging” their pores. Nonetheless, dUSCLs can synergize with a partner antibiotic by enhancing their intracellular accumulation through permeabilization of bacterial membranes and/or disruption of active efflux leading to increased intracellular concentration.

Antibacterial activity of chloramphenicol was enhanced by dUSCLs against other MDR Gram‑negative bacteria

Synergy between dUSCLs and chloramphenicol was then assessed in other Gram-negative bacteria including Acinetobacter baumannii (5), E. coli (4), Klebsiella pneumoniae (3) and Enterobacter cloacae (1). Both dUSCLs 2 and 6 potentiated chloramphenicol against four out of five A. baumannii strains (Table 5), including the wild-type ATCC 17978 strain and MDR strains AB027, AB031 and 110193. Only additive interactions were found for the combinations against A. baumannii LAC-4 (Table 5), which may be attributed to phenotypic differences between the tested clinical isolates. Both dUSCLs also reduced the MIC of chloramphenicol against Enterobacteriaceae. Compound 6 potentiated chloramphenicol against all strains, while compound 2 was able to do so for only three out of four tested E. coli strains (Table 6). Both dUSCLs synergized with only one out of three MDR K. pneumoniae strains tested while they both synergized with MDR E. cloacae 117029 (Table 6).
These data show that dUSCLs 2 and 6 can potentiate the antibacterial activity of chloramphenicol against a panel of Gram-negative pathogens. However, it appeared that phenotypic variations between isolates affected the observed activity of the combination, suggesting that other resistance  
mechanisms besides reduced permeability and efflux were likely to be operational such as the expression of chloramphenicol-inactivating enzymes.
To test whether the observed chloramphenicol potentiation was specific to dUSCLs, three clinically used cationic amphiphile/surfactant comparators (benzethonium chloride, benzalkonium chloride and cetrimonium bromide) were assessed in combination with chloramphenicol against wildtype P. aeruginosa PAO1, A. baumannii ATCC 17978 and E. coli ATCC 25922. As shown in Fig. 3, none of the comparators were able to potentiate chloramphenicol suggesting that this adjuvant property is specific to dUSCLs 2 and 6.

dUSCLs also potentiated other antibiotics against P. aeruginosa

Besides chloramphenicol, we explored whether dUSCLs 2 and 6 can potentiate other classes of antibiotics against P. aeruginosa. We studied a panel containing fifteen antibiotics that included aminoglycosides, β-lactams, fluoroquinolones, fosfomycin and other antibiotics that have activity only against Gram-positive bacteria (Fig. 4). Compound 2 potentiated ten out of 15 clinically used antibiotics against P. aeruginosa PAO1 (Fig. 4). Synergism was observed with antibiotics that were greatly affected by efflux (trimethoprim, minocycline, fosfomycin, piperacillin, ciprofloxacin, levofloxacin, moxifloxacin and linezolid) and those with restricted permeation across outer membrane (rifampicin and vancomycin). On the other hand, dUSCL 6 only potentiated seven out of fifteen antibiotics against P. aeruginosa PAO1 (Fig. 4), including rifampicin, minocycline, fosfomycin, piperacillin, levofloxacin, moxifloxacin and linezolid. Our results suggest that the antibiotic potentiation effects are sequence-dependent and that dUSCL 2 can enhance a wider range of antibiotic classes relative to dUSCL 6 against P. aeruginosa, presumably by increasing their intracellular accumulation.

Conclusion

Fig. 3 Fractional inhibitory concentration (FIC) indices of combinations consisting of chloramphenicol (CHL) and either dilipid ultrashort cationic lipopeptides (dUSCLs) or clinically used cationic amphiphiles against wild-type Gram-negative bacteria. Benzethonium chloride, benzalkonium chloride and cetrimonium bromide were used as comparators of commonly used antiseptics/ surfactants. All used cationic amphiphile comparators did not synergize with CHL against Gram-negative bacteria. Red dashed line denotes the cutoff FIC index of ≤ 0.5 for synergistic interaction (color figure online)
An attempt to develop dUSCLs possessing potent antibacterial activity was described. We found that dUSCLs consisting of ≥ 11 carbons-long aliphatic dilipids strongly lysed red blood cells at 100 µg/mL, which greatly limited their therapeutic potential. However, several nonhemolytic dUSCLs appeared to be promising adjuvants in combination with chloramphenicol and other conventional antibiotics against Gram-negative bacteria. The dUSCLs 2 and 6 were identified as lead adjuvant candidates, both consisting of C 9 dilipid with peptide sequence of KKKK and KKGK, respectively. Both dUSCLs enhanced the antibacterial activity of chloramphenicol against MDR clinical isolates of P. aeruginosa, A. baumannii and Enterobacteriaceae with similar degrees of potency. However, dUSCL 2 synergized with a wider range of antibiotic classes against P. aeruginosa relative to dUSCL 6. While dUSCLs, like AMPs, permeabilize bacterial membranes resulting in enhanced intracellular antibiotic accumulation, our data suggest that they are also able to indirectly disrupt active efflux of chloramphenicol in P. aeruginosa. Our study demonstrates that dUSCLs can aid conventional antibiotics, such as chloramphenicol, in combination against antibiotic-resistant Gram-negative bacteria.

References

Ahn M, Jacob B, Gunasekaran P et al (2014) Poly-lysine peptidomimetics having potent antimicrobial activity without hemolytic activity. Amino Acids 46:2259–2269. https ://doi.org/10.1007/ s0072 6-014-1778-z
Dawgul MA, Greber KE, Bartoszewska S et al (2017) In vitro evaluation of cytotoxicity and permeation study on lysine- and arginine-based lipopeptides with proven antimicrobial activity. Molecules. https ://doi.org/10.3390/molec ules2 21221 73
Domalaon R, Yang X, O’Neil J et al (2014) Structure-activity relationships Levofloxacin in ultrashort cationic lipopeptides: the effects of amino acid ring constraint on antibacterial activity. Amino Acids 46:2517–2530. https ://doi.org/10.1007/s0072 6-014-1806-z
Domalaon R, Zhanel GG, Schweizer F (2016) Short antimicrobial peptides and peptide scaffolds as promising antibacterial agents. Curr Top Med Chem 16:1217–1230. https ://doi. org/10.2174/15680 26615 66615 09151 12459
Domalaon R, Berry L, Tays Q et al (2018a) Development of dilipid polymyxins: investigation on the effect of hydrophobicity through its fatty acyl component. Bioorg Chem. https ://doi. org/10.1016/j.bioor g.2018.07.018
Domalaon R, Idowu T, Zhanel GG, Schweizer F (2018b) Antibiotic hybrids: the next generation of agents and adjuvants against Gram-negative pathogens? Clin Microbiol Rev. https ://doi.org/10.1128/cmr.00077 -17
Domalaon R, Sanchak Y, Koskei LC et al (2018c) Short proline-rich lipopeptide potentiates minocycline and rifampin against multidrug- and extensively drug-resistant Pseudomonas aeruginosa. Antimicrob Agents Chemother. https: //doi.org/10.1128/aac.02374 -17
Feng Q, Huang Y, Chen M et al (2015) Functional synergy of alpha-helical antimicrobial peptides and traditional antibiotics against Gram-negative and Gram-positive bacteria in vitro and in vivo. Eur J Clin Microbiol Infect Dis 34:197–204. https ://doi. org/10.1007/s1009 6-014-2219-3
Findlay B, Zhanel GG, Schweizer F (2012) Investigating the antimicrobial peptide “window of activity” using cationic lipopeptides with hydrocarbon and fluorinated tails. Int J Antimicrob Agents 40:36–42. https ://doi.org/10.1016/j.ijant imica g.2012.03.013
Fjell CD, Hiss JA, Hancock REW, Schneider G (2011) Designing antimicrobial peptides: form follows function. Nat Rev Drug Discov 11:37–51. https ://doi.org/10.1038/nrd35 91
Greber KE, Dawgul M, Kamysz W et al (2014) Biological and surface-active properties of double-chain cationic amino acid-based surfactants. Amino Acids 46:1893–1898. https: //doi.org/10.1007/ s0072 6-014-1744-9
Greber KE, Dawgul M, Kamysz W, Sawicki W (2017) Cationic net charge and counter ion type as antimicrobial activity determinant factors of short lipopeptides. Front Microbiol 8:123. https ://doi. org/10.3389/fmicb .2017.00123
Hoban DJ, Zhanel GG (2013) Introduction to the CANWARD study (2007-11). J Antimicrob Chemother 68(Suppl 1):i3–i5
Jorge P, Perez-Perez M, Perez Rodriguez G et al (2017) A network perspective on antimicrobial peptide combination therapies: the potential of colistin, polymyxin B and nisin. Int J Antimicrob Agents 49:668–676. https ://doi.org/10.1016/j.ijant imica g.2017.02.012
Juhaniewicz-Dębińska J, Tymecka D, Sęk S (2018) Lipopeptideinduced changes in permeability of solid supported bilayers composed of bacterial membrane lipids. J Electroanal Chem 812:227–234. https ://doi.org/10.1016/j.jelec hem.2017.12.065
Makovitzki A, Avrahami D, Shai Y (2006) Ultrashort antibacterial and antifungal lipopeptides. Proc Natl Acad Sci USA 103:15997–16002. https ://doi.org/10.1073/pnas.06061 29103
Masuda N, Sakagawa E, Ohya S et al (2000) Substrate specificities of MexAB-OprM, MexCD-OprJ, and MexXY-oprM efflux pumps in Pseudomonas aeruginosa. Antimicrob Agents Chemother 44:3322–3327
Meletiadis J, Pournaras S, Roilides E, Walsh TJ (2010) Defining fractional inhibitory concentration index cutoffs for additive interactions based on self-drug additive combinations, Monte Carlo simulation analysis, and in vitro-in vivo correlation data for antifungal drug combinations against Aspergillus fumi. Antimicrob Agents Chemother 54:602–609. https ://doi.org/10.1128/AAC.00999 -09
Naghmouchi K, Le Lay C, Baah J, Drider D (2012) Antibiotic and antimicrobial peptide combinations: synergistic inhibition of Pseudomonas fluorescens and antibiotic-resistant variants. Res Microbiol 163:101–108. https ://doi.org/10.1016/j.resmi c.2011.11.002
Park Y, Park SN, Park S-C et al (2006) Synergism of Leu-Lys rich antimicrobial peptides and chloramphenicol against bacterial cells. Biochim Biophys Acta 1764:24–32. https ://doi.org/10.1016/j. bbapa p.2005.10.019
Schwarz S, Kehrenberg C, Doublet B, Cloeckaert A (2004) Molecular basis of bacterial resistance to chloramphenicol and florfenicol. FEMS Microbiol Rev 28:519–542. https: //doi.org/10.1016/j.femsr e.2004.04.001
Soren O, Brinch KS, Patel D et al (2015) Antimicrobial peptide novicidin synergizes with rifampin, ceftriaxone, and ceftazidime against antibiotic-resistant Enterobacteriaceae in vitro. Antimicrob Agents Chemother 59:6233–6240. https ://doi.org/10.1128/AAC.01245 -15
Steenbergen JN, Mohr JF, Thorne GM (2009) Effects of daptomycin in combination with other antimicrobial agents: a review of in vitro and animal model studies. J Antimicrob Chemother 64:1130–1138. https ://doi.org/10.1093/jac/dkp34 6
The Clinical and Laboratory Standards Institute (2016) Performance standards for antimicrobial susceptibility testing CLSI supplement M100S, 26th edn. Clin. Lab. Stand Institute, Wayne, PA
World Health Organization (2015) The selection and use of essential medicines. Report of the WHO Expert Committee, 2015 (including the 19th WHO Model List of Essential Medicines and the 5th WHO Model List of Essential Medicines for Children)
Zhanel GG, DeCorby M, Laing N et al (2008) Antimicrobial-resistant pathogens in intensive care units in Canada: results of the Canadian National Intensive Care Unit (CAN-ICU) study, 20052006. Antimicrob Agents Chemother 52:1430–1437. https ://doi. org/10.1128/AAC.01538 -07