Antibacterial amphiphiles based on ε-polylysine: synthesis, mechanism of action, and cytotoxicity

Abstract

The development of antibacterial materials is recently more and more important and urgent due to the emergence of antibiotic-resistant bacteria. To address the problem, series of new amphiphiles based on ε-polylysine (ε-PL) were synthesized by alkylation reaction with alkyl bromide, and concurrently their antibacterial activity and cytotoxicity were evaluated. The amphiphiles with a large concentration of positive charges and lipid chain promoted their adsorption to bacterial membranes through electrostatic interaction and hydrophobic interaction, subsequently killed both Gram-positive (S. aureus and B. amyloliquefaciens) and Gram-negative (E. coli and P. aeruginosa) bacteria. Morphology observed using SEM shows that these derivatives could cause leakage of intracellular contents. Analysis of the antimicrobial mechanism displays that these derivatives against the bacteria started with disruption of the bacterial membrane, which caused the leakage of cytoplasm, and killed the bacteria. Among the amphiphiles, ε-PL-g-butyl2 presented the most effective antibacterial activity and its minimum inhibitory concentration as low as 3.9 μg mL⁻¹. Importantly, the effective antibacterial concentration of ε-PL-g-butyl2 displayed no cytotoxicity against human cells. This work not only highlights the great promise of using ε-PL-g-butyl2 as a highly effective antibacterial agent but also provides the important tool for understanding the interactions between the microorganisms and amphiphiles-based ε-PL.

---

Introduction

The development of antibacterial materials is recently more and more important and urgent due to the emergence of antibiotic-resistant bacteria.^1,2 It has been a scientific challenge to create new antibacterial agents that are not susceptible to the development of resistance mechanisms in pathogenic bacteria.³ To address these issues, efforts were made to develop new macromolecules with multiple potential targets or complex mechanism to obtain desirable biological and physicochemical properties.⁴ Since bacterial cell membranes contain negatively charged lipids in greater abundance than mammalian cells membranes, cationic and amphiphilic peptides preferentially bind to bacteria by electrostatic attraction, resulting in the selective targeting of bacteria over human cells.⁵ Based on this, cationic and amphiphilic polymers have been utilized as a platform to mimic the structural features and function of antimicrobial peptides.⁶ It is noted that ε-polylysine (ε-PL) is highly effective against a broad spectrum of food pathogens and spoilage organisms, and thus has great potential for utilization in food and beverage products.^7–9 Chemically, ε-PL is a homopolymer consisting of L-lysine monomers (typically between 25 and 35) linked together by isopeptide bond between ε-amino group and α-carboxyl group.⁸ Previous study found that conjugation of derivatives of benzoic acids to the α-amino group of ε-PL diminished its antibacterial activity,⁷ while hydrophobically modified ε-PL also retained the antimicrobial activity of ε-PL.^10–12 Hydrophobic lipid membrane provides a virtually impenetrable barrier to ionic and polar substances, while hydrophobic flexible chain which is compatible with the lipid bilayer of the bacteria can interact further with the lipid membrane and subsequently disrupt the cytoplasmic membrane.¹³ An antimicrobial agent that acts directly by disrupting the membrane of microbial cell appears to be a better solution to curbing the development of bacterial resistance, which is because microorganisms hardly change their lipid bilayer structure of membrane to develop resistance.^14,15 In this work, we introduce lipophilic groups onto ε-PL, and thus modified ε-PL have both hydrophobic and hydrophilic regions that enable solubility in an aqueous environment and allow the molecule to pass through lipid-rich bacterial membranes. Moreover, the positively charged residues associated with these molecules interact electrostatically with negatively charged components of the microbial cell wall, causing disruption of bacterial membrane integrity.^9,16

To verify the hypothesis, ε-PL was hydrophobically modified with alkyl bromide (ethyl bromide, butyl bromide, hexyl bromide, octyl bromide and dodceyl bromide), and systematically investigated the impacts of the degree of substitution and alkyl chain length on antimicrobial behavior of the materials. The antibacterial activities of these materials were investigated using Gram-positive (S. aureus and B. amyloliquefaciens) and Gram-negative (E. coli and P. aeruginosa) bacteria as model microorganisms. The results confirm that ε-PL-g-butyl2 was the most active compound and the minimum inhibitory concentration as low as 3.9 μg mL⁻¹. Increases of fluorescence polarization value show that membrane fluidity of E. coli decreased in the presence of antibacterial materials, and releases of the β-galactosidase further illuminated the disruption of cytoplasmic membrane, which is lethal for bacteria. The disruption of membrane integrity induced by ε-PL-g-butyl2 was also confirmed by acridine orange/ethidium bromide double staining experiment and bacterial morphological observation using SEM. Finally, to explore their potential application as antibacterial agents, cytotoxicity was assessed by MTT assay using mouse fibroblast cells (NIH 3T3 cells), and results display that cytotoxicity of material changed with the chain length of substituents. These characteristics show that the ε-PL-g-butyl2 antibacterial agent has potential food application.

Experimental section

Materials

ε-Polylysine (ε-PL, 7000 Da) was purchased from Shijiazhuang Hongda Biologic Technology Co., Ltd (Shijiazhuang, China) and used without further purification. Ethyl bromide, butyl bromide, hexyl bromide, octyl bromide and dodecyl bromide were chemical grades from the J&K China Chemical Ltd (Beijing, China). The fluorescent probe 1,6-diphenyl-1,3,5-hexatriene (DPH, 98% pure), acridine orange (AO), and ethidium bromide (EB) were purchased from Alfa Aesar (Ward Hill, MA, US). O-Nitrophenyl-β-D-galactopyranoside (ONPG) was purchased from Tianjin Heowns Biochem Technologies LLC. Dialysis membrane (MWCO: 3500 Da) was purchased from Tianjin Lianxing Biotechnology Co., Ltd (Tianjin, China). The rest of the chemicals were analytical reagents and used without further purification.

Synthesis of alkylated ε-PL

Alkylated ε-PL was synthesized according to the previous method with minor modifications.¹⁷ A typical procedure was described as follows: ε-PL (0.5 g) was dissolved in 10 mL of dimethyl sulfoxide (DMSO), and 4 mL of 1 mol L⁻¹ sodium hydroxide was added dropwise under stirring at 50 °C for 2 h, then a certain amount of alkyl bromide in the solution of 1,4-dioxane was added dropwise. After the mixture was continuously stirred for 24 h at 60 °C, the product was finally purified by dialysis against deionized water for 3 days and subsequently lyophilized with a Flexi-dry freeze-dry system (America, FTS SYSTEMS Ltd), and collected as white or light brown cotton wool-like product.

Characterizations

The FT-IR spectra of ε-PL and alkylated ε-PL were recorded on a Fourier Transform Infrared Spectrometer (FTS-6000, Bio-Rad Co.) with a KBr tablet containing the powder of the sample at a resolution of 8 cm⁻¹. Proton nuclear magnetic resonance (¹H NMR) spectrum was recorded at room temperature using a Varian Unity-plus 400 MHz NMR spectrometer. The solvent was heavy water and dimethyl-d₆ sulfoxide. The degree of substitution was defined as the percentage of the reacted α-amine group on the lysine monomer in the ε-PL molecule,¹² which was calculated through elemental analysis (Elemental Analyzer Vario EL III, Germany).

Preparation of alkylated ε-PL nanoparticles

Synthesized ε-PL-g-ethyl, ε-PL-g-butyl, ε-PL-g-hexyl and ε-PL-g-octyl were water soluble. As the length of hydrophobic chain further increased, ε-PL-g-dodceyl was not completely soluble. So its nanoparticles were prepared by the nanoprecipitation method.^18,19 In brief, 30 mg of copolymer was dissolved in 2 mL of DMSO, and then added dropwise to 20 mL of water under stirring. The resulting solution was incubated overnight at room temperature with continuous stirring, and then dialyzed against water for 3 days to remove the organic solvent. Finally, the product was dried using a Flexi-dry freeze-dry system.

Determination of the MIC

The test organisms were representative in this study: Gram-positive S. aureus ATCC 6538 and B. amyloliquefaciens ATCC 23842, Gram-negative E. coli ATCC 8739 and P. aeruginosa ATCC 9027. The MICs of alkylated ε-PL were determined by a modified 2-fold microtiter broth dilution.^20,21

The microorganism was grown overnight in Luria–Bertani (LB) broth at 37 °C, and then diluted with LB broth to approximately 2.0 × 10⁶ CFU mL⁻¹. Then the tested microorganisms were mixed with an equal volume of 2-fold diluted antimicrobial agent solution and incubated. Media inoculated with the bacteria but without antimicrobial compounds served as positive controls, and medium alone was taken as negative control. After the bacteria were statically cultured for 8 h at 37 °C, the visible growth of the bacterial cells was assessed by determining the optical density at 600 nm (OD₆₀₀) using a UV-vis spectrometer (Shimadzu UV-2550). The MIC of each tested compound was defined as the lowest concentration of the compound at which bacteria had no optically prominent growth. Each assay was carried out in triplicate.

Assay of inhibition zone

The assay of inhibition zone was performed with minor modifications of the previous process.²² The bacteria were incubated in LB broth at 37 °C overnight. The resultant bacterial suspension was diluted to approximate 1.0 × 10⁷ CFU mL⁻¹ with LB broth. Subsequently, 50 μL of the bacterial suspension was inoculated evenly on LB agar plates. Then, the sample disk containing the antimicrobial agent solution was gently placed at the center of the LB agar plate and incubated overnight at 37 °C. Each test was performed in duplicate.

Fluorescence polarization measurement

The DPH stock solution was prepared according to the previous report.²³ The fast-growing E. coli bacteria (5.0 × 10⁷ CFU mL⁻¹) were washed with phosphate buffer solution (PBS, 0.01 mol L⁻¹, pH 7.4) twice and resuspended in 1.0 mL PBS. The tested bacteria were firstly inoculated with a series of concentration of the material for 20 min, then 1.0 mL of 1.0 × 10⁻⁵ mol L⁻¹ DPH was added to each treated bacteria and incubated in the dark for 15 min to allow the probe incorporation into the cytoplasmic membrane. Finally, the bacterial cells were rinsed with PBS and resuspended in 1.0 mL of PBS for the fluorescence measurement.

The steady state fluorescence measurement was performed with a Shimadzu RF-5301PC spectrofluorometer using 1 cm path length quartz cuvettes. Excitation and emission wavelengths were set at 360 and 442 nm, respectively. Polarizers were set in either a vertical or a horizontal position. The slit widths for the excitation and emission beams were both 5 nm, respectively. Background intensities of samples were subtracted from each sample spectrum to cancel out any contribution due to the solvent Raman peak and other scattering artifacts. Fluorescence polarization measurement was performed using a Shimadzu polarization accessory. Polarization value was calculated from the following equation:²⁴

Polarization value = (I_VV − GI_VH)/(I_VV + GI_VH)

where I_VV was the fluorescence intensity with vertical excitation and vertical emission, I_VH was the fluorescence intensity with vertical excitation and horizontal emission, I_HV was the fluorescence intensity with horizontal excitation and vertical emission and I_HH was the fluorescence intensity with horizontal excitation and horizontal emission. The correlation factor G was defined as the ratio of I_HV to I_HH.

Assay of cytoplasmic membrane permeabilization

The cytoplasmic membrane permeabilization was assayed by determining the release of β-galactosidase from E. coli using ONPG as the substrate as reported.^25,26 Briefly, midlog-phase E. coli cells were harvested (5000 rpm for 5 min, 4 °C), and resuspended in PBS. After 10 μL of ONPG (25 mmol L⁻¹) was added to 1 mL of bacterial suspension, 1 mL of ε-PL-g-butyl2 solution (1 or 0.5 mg mL⁻¹) was added to the above mentioned suspension, respectively. The control assay was carried out without the addition of ε-PL-g-butyl2. The hydrolysis of ONPG was determined by monitoring the increase in OD₄₂₀ using a UV-vis spectrometer.

AO/EB double staining for fluorescence microscopy

The fluorescent dyes were prepared by mixing 10 mg of AO and 10 mg of EB in 10 mL of PBS. The bacterial suspension (1.5 mL, 1.0 × 10⁸ CFU mL⁻¹) was harvested by centrifuging (5000 rpm for 5 min) at 4 °C and washing with PBS for three times. The supernatant was discarded and the remaining bacteria were resuspended in 1.5 mL of PBS. Then, 100 μL of 1.0 mg mL⁻¹ ε-PL-g-butyl2 solution was added to the bacterial suspensions. After incubation for 2 h, the bacteria cells were stained with 100 μL of fluorescent dyes for 15 min. After rinsing with PBS, 10 μL of the sample was placed on a glass slide with a glass coverslip and observed under an inverted fluorescence microscope (Leica DMI 4000B). The control assay was performed without ε-PL-g-butyl2 treatment.

Bacterial morphological observation

The morphological changes of S. aureus and E. coli were investigated by scanning electron microscope (SEM). Overnight grown bacteria suspensions were harvested by centrifugation at 5000 rpm for 5 min at 4 °C, and then inoculated with 66.7 μg mL⁻¹ ε-PL-g-butyl2 dispersion at 37 °C for 2 h. The cells were collected and washed twice with PBS, followed by resuspended in 500 μL of PBS. One droplet of cell suspension was dropped on the freshly treated glass slide and fixed with a 2.5% glutaraldehyde solution overnight. Then, bacterial cells were dehydrated with sequential treatment of 30%, 50%, 70%, 90%, 95% and 100% ethanol for 10 min, subsequently lyophilized with a Flexi-dry freeze-dry system. Finally, the dried samples were sputter-coated with platinum, and imaged on the SEM (Shimadzu SS-550).

Cytotoxicity assay

The cytotoxicity of alkylated ε-PL was evaluated by MTT assay in NIH 3T3 cells. The cells were cultured in Dulbecco's Modified Eagles Medium (DMEM, Gibco) which was supplemented with streptomycin (100 μg mL⁻¹), penicillin (100 U per mL), 1% non-essential amino acid and 10% heat-inactivated fetal bovine calf serum, respectively. The cells were seeded into 96-well plates at 10 000 cells per well in 100 μL of growth medium. The plates were then returned to the incubator and the cells were incubated for 24 h. The antibacterial materials were diluted with DMEM to give a final range of concentrations from 25 to 200 μg mL⁻¹. Thereafter, the medium from each well was replaced with 100 μL of the sample and incubated for 2 days, control solution using culture medium by itself as a control. Thereafter, each well was treated with 10 μL of MTT and incubated for further 4 h. Then, the culture medium was removed and 100 μL of DMSO was added to each well to dissolve the internalized purple formazan crystals. The plate was placed in a humidified atmosphere of 5% CO₂/95% O₂ at 37 °C for 10 min, and thereafter for 15 min at 6 °C prior to determination of optical density using a microplate reader at 490 nm. Relative cell proliferation rate was determined as a percentage of the positive control; untreated cells were used as the positive control and their proliferation rate was set to 100%. NIH 3T3 cells were seeded in 24-well plates at 10 000 cells per well in 400 μL of growth medium and incubated at 37 °C for 1 day. Thereafter, the test group treated with the amphiphiles for 2 days and then observed under brightfield mode by fluorescence microscope.

Statistical analysis

All data are expressed as means ± standard deviation of three or five different experiments, and compared via Kruskal–Wallis one-way analysis of variance (ANOVA), where *p < 0.05, significant; **p < 0.01, very significant; ***p < 0.001, extremely significant.

Results and discussion

Synthesis and characterization of alkylated ε-PL

ε-PL reacted with alkyl bromide (ethyl bromide, butyl bromide, hexyl bromide, octyl bromide and dodceyl bromide) in DMSO through nucleophilic reaction (Scheme 1). By regulating the feed ratio of alkyl bromide to the amino group in ε-PL, a series of alkylated ε-PL with different degrees of substitution and alkyl chain lengths were obtained (Table 1). As shown in Table 1, the degree of substitution increased by increasing the concentration of alkyl bromide in a range. The degree of substitution did not increase when the concentration of alkyl bromide reached a certain values. It was possibly caused by the space steric hindrance of the polymer chain. Moreover, the degree of substitution did not increase as the increase of alkyl chain under the same concentration.

Scheme 1 Synthetic scheme of alkylated ε-PL.

Table 1 Preparation conditions of alkylated ε-PL

Sample	n(ε-PL) (mmol)	n(Alkyl bromide) (mmol)	Degree of substitution^a (%)	Yield^b (%)
a Degree of substitution was calculated based on the different C, H, N rations of modified ε-PL through elemental analysis.b Determined by gravimetry.
ε-PL-g-ethyl	0.071	1.3	27	69
ε-PL-g-butyl1	0.071	3.9	74	48
ε-PL-g-butyl2	0.071	2.0	23	67
ε-PL-g-butyl3	0.071	1.0	15	51
ε-PL-g-hexyl	0.071	2.0	25	60
ε-PL-g-octyl	0.071	2.0	28	59
ε-PL-g-dodecyl	0.071	2.0	24	53

---

The successful incorporation of the alkyl group into ε-PL backbone was confirmed by FT-IR and ¹H NMR assay. As shown in Fig. 1, a broad absorption band in a range of 3200 to 3300 cm⁻¹ was attributed to the N–H stretching. The peaks assigned to the amide I and II groups (δ_C=O: 1674 and δ_N–H: 1557 cm⁻¹) in the spectrum for ε-PL appeared in alkylated ε-PL. The two peaks at 2920 and 2852 cm⁻¹ designated as C–H stretching were naturally more prominent comparing with that of ε-PL, suggesting the presence of N-alkyl substitution. Also, the band at about 1461 cm⁻¹ resulted from C–H bending was intensified after alkylation. Moreover, the longer the alkyl chain, the greater those bands increased the range.

Fig. 1 FT-IR spectra of (a) ε-PL; (b) ε-PL-g-ethyl; (c) ε-PL-g-butyl2; (d) ε-PL-g-hexyl; (e) ε-PL-g-octyl and (f) ε-PL-g-dodecyl.

As shown in Fig. 2, the ¹H NMR peaks specific to ε-PL appeared between 1.2 and 4.0 ppm. Peaks at about 3.82 and 3.12 ppm corresponded to terminal methane and methylene groups, respectively.⁹ In the ε-PL-g-dodecyl, peak assignment was as follows: the peak at 0.82 ppm was attributed to the methyl proton (–NH[CH₂–(CH₂)₁₀–CH̲₃]), and 1.1–2.0 ppm was assigned to the methylene hydrogen (–NH[CH₂–(CH̲₂)₁₀–CH₃]) of the N-alkyl group. These results indicate that alkylated ε-PL was successfully synthesized.

Fig. 2 ¹H NMR spectra of ε-PL in D₂O and ε-PL-g-dodecyl in DMSO-d₆.

Antimicrobial activities

The antibacterial activity of alkylated ε-PL was characterized by the MIC and inhibitory zone against Gram-positive (S. aureus and B. amyloliquefaciens) and Gram-negative (E. coli and P. aeruginosa) bacteria. To investigate the influence of the structure factor on the antimicrobial property of the materials, we respectively study antimicrobial activity of ε-PL-g-butyl with different degrees of substitution carbon chain lengths. MIC can qualitatively characterize the antibacterial activity of antimicrobial agent. The MIC value of the alkylated ε-PL was shown in Table 2. The results show that ε-PL-g-butyl2 with a substitution degree of 23% was the most active compound and the minimum inhibitory concentration as low as 3.9 μg mL⁻¹. Moreover, ε-PL-g-butyl2 presented significant antibacterial activity against both Gram-positive (S. aureus and B. amyloliquefaciens) and Gram-negative (E. coli and P. aeruginosa) bacteria when the concentration was 15.6 μg mL⁻¹, and which was higher than that of ε-PL. The MIC of ε-PL-g-butyl1 and ε-PL-g-butyl3 is very close to that of ε-PL. This result suggests that basic groups in ε-PL molecule played vital role in determining antibiotic activity, but they could produce stronger synergetic antibacterial effect along with the hydrophobic interactions of introduced alkyl groups. In addition, the MIC of ε-PL-g-butyl depended on the type of the bacteria, as shown in the following order: P. aeruginosa > S. aureus ∼ E. coli > B. amyloliquefaciens. For ε-PL-g-butyl2, Gram-positive bacteria were more sensitive than Gram-negative bacteria. It is possibly because Gram-positive and Gram-negative bacteria differ in their membrane structure. The components of Gram-positive bacteria cell wall are peptidoglycan, wall associated protein, and teichoic acid. Among them, teichoic acid as a special ingredient in Gram-positive bacteria cell wall is charged negatively, providing rigidity to the cell wall by attracting cations such as Mg²⁺. It can be suggested that ε-PL-g-butyl2 could chelate Mg²⁺ and extract it from the original binding sites in teichoic acids, thus damaging the bacterial cell wall.²⁷

Table 2 MIC values of ε-PL and alkylated ε-PL^a

Sample	MIC (μg mL^−1)
	S. aureus	B. amyloliquefaciens	E. coli	P. aeruginosa
a Asterisks indicate significant differences between ε-PL-g-butyl2 and the other material in each column: **p < 0.01; ***p < 0.001.
ε-PL	31.3***	15.6**	31.3***	62.5***
ε-PL-g-ethyl	62.5***	31.3***	125.0***	250.0***
ε-PL-g-butyl1	62.5***	15.6***	62.5***	500.0***
ε-PL-g-butyl2	15.6	3.90	15.6	15.6
ε-PL-g-butyl3	62.5***	15.6**	62.5***	500.0***
ε-PL-g-hexyl	15.6	31.3***	15.6	62.5***
ε-PL-g-octyl	15.6	15.6***	15.6	31.3***
ε-PL-g-dodecyl	125.0***	62.5***	250.0***	250***

---

For alkylated ε-PL with different alkyl chain lengths, it is found that antibacterial activity of ε-PL-g-butyl2 was better than that of ε-PL-g-ethyl, but alkylated ε-PL became weak as the length of alkyl chain varied from butyl to dodecyl (Table 2). This phenomenon may be explained by the balance of hydrophilic/hydrophobic groups of these cationic amphiphiles. The hydrophobicity of ε-PL-g-alkyl increased with the increase of alkyl chain length, leading to stronger hydrophobic interactions with the bacterial cell membrane in addition to the electrostatic interactions.²⁸ However, the hydrophobic tail longer than butyl significantly reduces antibacterial activity because of the high hydrophobicity property of these polymers, thus limiting the accessibility of the active groups to bacteria.²⁹

The test of inhibition zone was also carried out, which qualitatively characterized the antibacterial property of antimicrobial agent. By comparing antibacterial performance of ε-PL and ε-PL-g-butyl2 against four kinds of bacteria (Fig. 3), the diameters of the zones of inhibition of ε-PL-g-butyl2 were larger than those of ε-PL. In general, larger diameters of the bacteria-free zone surrounding the disk suggest that bacteria are more sensitive to the antibacterial materials contained in the disk.³⁰ The result was in accordance with that of the MIC test, as shown in the following order: P. aeruginosa > S. aureus ∼ E. coli > B. amyloliquefaciens. Clearly, ε-PL-g-butyl2 was the most active compound against both Gram-positive and Gram-negative.

Fig. 3 Photograph images of the inhibition zone of ε-PL and ε-PL-g-butyl2.

Membrane fluidity assay

DPH as a fluidity probe that can localize within the hydrophobic core of the bacterial membrane,²⁰ through determining the polarization values, has been widely used to monitor the change of membrane fluidity.^31–33 Membrane polarization values were estimated as the averages of total cells. A high polarization value indicates low membranes fluidity,³⁰ indicating that the material had a high antibacterial activity. Fig. 4 showed the effect of the alkylated ε-PL on the polarization values of cytoplasmic membrane of E. coli. It can be seen that the polarization values of treated bacteria were higher than control group, indicating the decreased membrane fluidity of E. coli and bacterial death. We found that the polarization values of ε-PL-g-butyl2 were up to 0.269. Moreover, the membrane fluidity was influenced by the carbon chain length of alkylated ε-PL, suggesting that the grafted hydrophobic chain affected the membrane function. The alkyl chain of external ε-PL provides a lipophilic segment compatible with the lipid bilayer of the bacterial cytoplasmic membrane. It hinders the bacteria to modulate the membrane composition to maintain fluidity for cell growth and division, leading to the decrease of the membrane fluidity and disruption of the bacterial cytoplasmic membrane.^31,34 The effect of ε-PL-g-butyl2 with different concentrations on the membrane fluidity of E. coli was also studied (Fig. 5). With the increase of the concentration of ε-PL-g-butyl2, polarization value increased gradually. Thus, high polarization indicates the decrease of membrane fluidity and bacterial death.

Fig. 4 Effects of alkylated ε-PL (1.0 mg mL⁻¹) on membrane fluidity of E. coli. The results are expressed as the mean of three independent experiments and error bars show the standard deviation; *** indicates statistical significance p < 0.001, **p < 0.01.

Fig. 5 Effects of ε-PL-g-butyl2 with different concentrations (0.125, 0.25, 0.5 and 1.0 mg mL⁻¹) on membrane fluidity of E. coli.

Membrane permeabilization assay

It has been reported that β-D-galactosidase has special hydrolyzing activity for β-D-galactose glycoside bond in sugar, fat, and lactose. Especially, β-D-galactosidase from E. coli has particular substrate specificity for the hydrolysis of ONPG. If the cytoplasmic membrane of E. coli disrupts, β-D-galactosidase will leak out from the cells and subsequently catalyzes the hydrolysis of ONPG in solution to produce O-nitro phenol (ONP), which has a characteristic absorption at 420 nm wavelength.³⁵ Previous researches showed that the damage to the cell wall and cytoplasmic membrane was the loss of structural integrity and the ability of the membrane to act as a permeability barrier.^36–38 Hence, in the present work, to investigate whether permeable alteration of cytoplasmic membrane occurred in the presence of ε-PL-g-butyl2, the OD₄₂₀ of ONP was measured using a UV-vis spectroscopy.

As shown in Fig. 6, OD₄₂₀ of the control was nearly zero and did not vary with time. The fact displays that the membrane of E. coli without the treatment of ε-PL-g-butyl2 was intact and no β-D-galactosidase was released. However, for E. coli treated with ε-PL-g-butyl2, OD₄₂₀ increased with the increase of incubation time and the material concentration. The result indicates clearly that ε-PL-g-butyl2 could increase the permeabilization of cytoplasmic membrane, leading to leakage of the intracellular contents including β-D-galactosidase. Moreover, the permeabilization effect on the cytoplasmic membrane showed dose-dependent and time-dependent models. The result shows that the antimicrobial activity of ε-PL-g-butyl2 was based on a process of bactericidal by the irreversible disruption of the cytoplasmic membrane.

Fig. 6 Absorption of ONP over the time and concentration.

Fluorescence microscope

The LIVE/DEAD bacterial viability assay was performed by E. coli and S. aureus stained with AO/EB fluorescent dyes by fluorescence microscope. Generally, healthy cells have green nucleoid and uniform chromatin with an intact cell membrane, whereas the cells in necrosis or a late stage of apoptosis have red nucleoid, which is because that impermeable nucleic acids dye EB only transport through cell with a damaged cell membrane.³⁹ As demonstrated in Fig. 7A and C, only a fewer necrotic cells were observed in untreated groups. In contrast, the bacteria treated with ε-PL-g-butyl2 for 2 h were almost red (Fig. 7B and D), indicating that ε-PL-g-butyl2 can rapidly and efficiently kill the bacteria.

Fig. 7 Fluorescence microscopy images of AO/EB dual staining bacterial cells after incubation with ε-PL-g-butyl2 (66.7 μg mL⁻¹) for 2 h. (A) S. aureus untreated; (B) S. aureus treated with ε-PL-g-butyl2; (C) E. coli untreated and (D) E. coli treated with ε-PL-g-butyl2. Each image is a result of the superposition of an image taken for the green fluorescent dye and red fluorescent dye using the appropriate filters. The scale bar is 50 μm.

Bacterial morphological observation

The results above were further analyzed by observing the morphology of bacterial cells using SEM. As shown in Fig. 8A and C, untreated S. aureus and E. coli were typically round-shaped and rod-shaped, respectively, both with smooth and intact cell walls. After being treated (Fig. 8B and D), the shape and size of cells changed dramatically. Almost all cell membranes became wrinkled, and some cells even ruptured completely and lost their cellular integrity, causing leakage of a large number of intracellular contents. Therefore, it can be concluded that ε-PL-g-butyl2 could sanitize against the bacteria started with disruption of the bacterial membrane, resulting in cell death.

Fig. 8 SEM images of bacterial cells after incubation with ε-PL-g-butyl2 (66.7 μg mL⁻¹) for 2 h. (A) S. aureus untreated; (B) S. aureus treated with ε-PL-g-butyl2; (C) E. coli untreated and (D) E. coli treated with ε-PL-g-butyl2. The white box in B and D demonstrates a detail of the merged image. The scale bar is 1 μm.

Cell viability

The cytotoxicity of materials is an important consideration to human healthy living. It is well-known that the cytotoxicity of ε-PL arises from its α-primary amine groups, however, ε-PL exhibits the low cell viability due to its charge density as reported in previous studies.^40,41 Fig. 9 showed the cytotoxicity of the samples analyzed by the MTT method in NIH 3T3 cells. As shown in Fig. 9, when the dose increased to 100 μg mL⁻¹ and incubated for 2 days, the cell viability was still more than 80%, indicating that ε-PL-g-butyl2 displayed no cytotoxicity. It is found that the cytotoxicity increased gradually with an increase in the length of alkyl chain, which is due to increasing hydrophobic chain–lipid interaction.⁴² However, the cytotoxicity of ε-PL-g-dodecyl was lower than that of ε-PL-g-octyl. This result was consistent with their trend of antibacterial performance. Fig. 9A–D showed the morphologies of the NIH 3T3 cells after treated with different ε-PL-g-butyl2 dosages for 2 days. There are no obvious differences in cell morphologies between the control and tested groups. Thus, optimization of the substituent group of the polycations is necessary to avoid introducing cytotoxicity while improving the antibacterial activity of alkylated ε-PL.

Fig. 9 Relative viabilities of NIH 3T3 cells after exposed to antibacterial materials with different concentrations for 2 days. The results are expressed as the mean of five independent experiments and error bars show the standard deviation; *** indicates statistical significance p < 0.001. Cell morphologies after 2 days culture with different dosages of ε-PL-g-butyl2: 0 μg mL⁻¹ (A); 50 μg mL⁻¹ (B); 100 μg mL⁻¹ (C) and 200 μg mL⁻¹ (D).

Conclusions

In this work, alkylated ε-PL was synthesized using a simple one-pot nucleophilic reaction. The degree of substitution and carbon chain length of substitution both affected the antibacterial activity of alkylated ε-PL, among all modified ε-PLs, ε-PL-g-butyl2 with the optimal degree of substitution (23%) showed remarkably antibacterial activity even at very low concentration (3.9 μg mL⁻¹). For E. coli treated with antibacterial amphiphiles, the decrease of membrane fluidity and release of β-D-galactosidase indicated disruption of the normal physiological functions of cytoplasmic membrane. Moreover, disruptions of bacterial membrane integrity induced by ε-PL-g-butyl2 were also visibly observed by AO/BE double staining experiment and SEM. Therefore, based on the results of antibacterial activity and cytotoxicity, ε-PL-g-butyl2 is expected to become a potential novel powerful antibacterial agent for future medical application.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant No. 21174071, 21474055 and 81170773), the Natural Science Foundation of Tianjin, China (Grant No. 14JYBJC29400), the Specialized Research Fund for the Doctoral Program of Higher Education (Grant No. 20130031110014), and Program for Changjiang Scholars and Innovative Research Team in University of Ministry of Education of China (IRT1257).

References

1. X. Liu, H. Zhang, Z. Tian, A. Sen and H. R. Allcock, Polym. Chem., 2012, 3, 2082–2091.
2. T. M. Uhizi, S. G. Relier and V. C. Oma, J. Agric. Food Chem., 2009, 19, 8770–8775.
3. Z. Lu, X. Zhang, Y. Zhao, Y. Xue, T. Zhai, Z. Wu and C. Li, Polym. Chem., 2015, 6, 302–310.
4. F. Siedenbiedel and J. C. Tiller, Polymers, 2012, 4, 46–71.
5. K. Matsuzaki, Biochim. Biophys. Acta, 1999, 1462, 1–10.
6. Y. Oda, S. Kanaoka, T. Sato, S. Aoshima and K. Kuroda, Biomacromolecules, 2011, 12, 3581–3591.
7. S. Shima, H. Matsuoka, T. Iwamoto and H. Sakai, J. Antibiot., 1984, 37, 1449–1455.
8. K. Yamanaka, N. Kito, Y. Imokawa, C. Maruyama, T. Utagawa and Y. Hamano, Appl. Environ. Microbiol., 2010, 76, 5669–5675.
9. L.-A. B. Rawlinson, S. M. Ryan, G. Mantovani, J. A. Syrett, D. M. Haddleton and D. J. Brayden, Biomacromolecules, 2010, 11, 443–453.
10. H. Y. Ting, S. Ishizaki and M. Tanak, J. Tokyo Univ. Fish., 1997, 84, 25–30.
11. Y.-T. Ho, S. Ishizaki and M. Tanaka, Food Chem., 2000, 68, 449–455.
12. H. Yu, Y. Huang and Q. Huang, J. Agric. Food Chem., 2010, 58, 1290–1295.
13. M. Zasloff, Nature, 2002, 415, 389–395.
14. J. Hoque, P. Akkapeddi, V. Yarlagadda, D. S. S. M. Uppu, P. Kumar and J. Haldar, Langmuir, 2012, 28, 12225–12234.
15. S. B. Levy, Adv. Drug Delivery Rev., 2005, 57, 1446–1450.
16. J. A. Castillo, A. Pinazo, J. Carilla, M. R. Infante, M. A. Alsina, I. Haro and P. Clapés, Langmuir, 2004, 20, 3379–3387.
17. J.-S. Kim, A. Maruyama, T. Akaike and S. W. Kim, Pharm. Res., 1998, 15, 116–121.
18. H. Gao, Y. Ma, X. Lu, Y. Liang, B. Chen and J. Ma, Eur. Polym. J., 2011, 47, 1232–1239.
19. V. Michailova, I. Berlinova, P. Iliev, L. Ivanov, S. Titeva, G. Momekov and I. Dimitrov, Int. J. Pharm., 2010, 384, 154–164.
20. S. Kim and K.-B. Oh, J. Microbiol. Biotechnol., 2002, 12, 1006–1009.
21. M. Wu and R. E. W. Hancock, J. Biol. Chem., 1999, 274, 29–35.
22. M. A. Rojas-Graü, R. J. Avena-Bustillos, M. Friedman, P. R. Henika, O. Martín-Belloso and T. H. McHugh, J. Agric. Food Chem., 2006, 54, 9262–9267.
23. S. Xia, S. Xu and X. Zhang, J. Agric. Food Chem., 2006, 54, 6358–6366.
24. A. Chattopadhyay and S. Mukherjee, J. Phys. Chem. B, 1999, 103, 8180–8185.
25. Z. Zhou, D. Wei, Y. Guan, A. Zheng and J. Zhong, J. Appl. Microbiol., 2010, 108, 898–907.
26. G. Tupinambá, A. J. R. Da silva, C. S. Alviano, T. Souto-Padron, L. Seldin and D. S. Alviano, J. Appl. Microbiol., 2008, 105, 1044–1053.
27. L. Mei, Z. Lu, W. Zhang, Z. Wu, X. Zhang, Y. Wang, Y. Wang, Y. Luo, C. Li and Y. Jia, Biomaterials, 2013, 34, 10328–10337.
28. M. Dathe, M. Schümann, T. Wieprecht, A. Winkler, M. Beyermann, E. Krause, K. Matsuzaki, O. Murase and M. Bienert, Biochemistry, 1996, 35, 12612–12622.
29. T. Rubén, L. Daniel, L.-F. Fátima, L. G.-G. José and F.-G. Marta, Polym. Chem., 2015, 6, 3449–3459.
30. L. Mei, Z. Lu, X. Zhang, C. Li and Y. Jia, ACS Appl. Mater. Interfaces, 2014, 6, 15813–15821.
31. N. C. S. Mykytczuk, J. T. Trevors, L. G. Leduc and G. D. Ferroni, Prog. Biophys. Mol. Biol., 2007, 95, 60–82.
32. V. M. Ioffea, G. P. Gorbenko, T. Deligeorgiev, N. Gadjev and A. Vasilev, Biophys. Chem., 2007, 128, 75–86.
33. S. Chu-Ky, R. Tourdot-Marechal, P.-A. Marechal and J. Guzzo, Biochim. Biophys. Acta, 2005, 1717, 118–124.
34. D. M. Engelman, Science, 1996, 274, 1850–1851.
35. B. Gao, S. He, J. Guo and R. J. Wang, Appl. Polym. Sci., 2006, 100, 1531–1537.
36. K. Knobloch, A. Pauli, B. Iberl, H. Wedigand and N. Weis, J. Essent. Oil Res., 1989, 1, 119–128.
37. P. Claeson, P. Rådström, O. Sköld, Å. Nilsson and S. Höglund, Phytother. Res., 1992, 6, 94–98.
38. V. G. de Billerbeck, C. G. Roques, J.-M. Bessière, J.-L. Fonvieille and R. Dargent, Can. J. Microbiol., 2001, 47, 9–17.
39. H. Peng, L. Meng, L. Niu and Q. Lu, J. Phys. Chem. C, 2012, 116, 16294–16299.
40. K. Kadowaki, M. Matsusaki and M. Akashi, Langmuir, 2010, 26, 5670–5678.
41. K. Matsumura and S.-H. Hyon, Biomaterials, 2009, 30, 4842–4849.
42. M. F. Ilker, K. Nüsslein, G. N. Tew and E. B. Coughlin, J. Am. Chem. Soc., 2004, 126, 15870–15875.

---