Journal of Nanoparticle Research

, 16:2517

Antibacterial effects of chitosan–tripolyphosphate nanoparticles: impact of particle size molecular weight

Authors

  • Atif Sarwar
    • Centre for Drug Delivery Research, Faculty of PharmacyUniversiti Kebangsaan Malaysia
    • Centre for Drug Delivery Research, Faculty of PharmacyUniversiti Kebangsaan Malaysia
  • Noraziah Mohamad Zin
    • Novel Antibiotic Research Group, Faculty of Health SciencesUniversiti Kebangsaan Malaysia
Research Paper

DOI: 10.1007/s11051-014-2517-9

Cite this article as:
Sarwar, A., Katas, H. & Zin, N.M. J Nanopart Res (2014) 16: 2517. doi:10.1007/s11051-014-2517-9

Abstract

This study revealed not only the antibacterial potential of smaller chitosan–tripolyphosphate nanoparticles (CS–TPP NPs) over larger ones, but also the attempt has been made to demonstrate antibacterial mechanism of action of CS–TPP NPs on the bacterial cell membrane. Several aspects of low-molecular-weight (LMW) and high-molecular-weight (HMW) CS–TPP NPs were evaluated by their interactions with selected Gram-positive and Gram-negative bacteria. The interaction of CS–TPP NPs with synthetic phospholipid membranes was also evaluated using Fourier transform infrared spectroscopy. The permeabilities of the bacterial outer and inner membranes were evaluated by determining the uptake of a fluorescent probe, 1-N-phenylnaphthylamine, and the release of cytoplasmic β-galactosidase. The morphology of the bacteria treated with LMW and HMW CS–TPP NPs was investigated using transmission electron microscopy. Flow cytometric analysis was also performed for the quantification of dead and surviving bacteria. These studies indicated that the antibacterial effects of LMW CS–TPP NPs (196 and 394 nm) were superior to those HMW CS–TPP NPs (598 and 872 nm). These data indicated that the antibacterial activity of CS–TPP NPs was negatively correlated with particle size and molecular weight, and that CS–TPP NPs represent a promising antimicrobial adjunct.

Keywords

ChitosanNanoparticlesParticle sizeMolecular weightAntibacterial

Introduction

Chitosan (poly-(b-1/4)-2-amino-2-deoxy-d-glucose) is a natural polycationic carbohydrate biopolymer derived by the deacetylation of chitin. Chitin is found naturally in the exoskeleton of arthropods (e.g., crab, crawfish, shrimp, and lobster) (Rinaudo 2006). Because of innate biocompatibility, biodegradability, and lack of toxicity, chitosan, its derivatives and nanoparticles have received great attention in the pharmaceutical, food, agriculture, textile, and cosmetics industries. Moreover, bacteria are not known to develop resistance to chitosan (Didenko et al. 2005).

Chitosan has biological activities including anti-tumor, wound-healing, and antimicrobial activities, and is useful for tissue-engineering applications (Tang et al. 2010). Chitosan also exhibits broad-spectrum inhibition of both Gram-positive and Gram-negative bacteria. Its minimum inhibitory concentration (MIC) against Gram-negative bacteria varies from 100 to 10,000 µg/mL (Helander et al. 2001), whereas that against Gram-positive bacteria varies from 100 to 1,250 µg/mL (Bae et al. 2006; Jeon et al. 2001; Vishu Kumar et al. 2004; No et al. 2002). Silver-loaded chitosan nanoparticles showed the highest antibacterial activity against Escherichiacoli and Staphylococcusaureus with MIC of 3 and 6 µg/mL, respectively. Meanwhile, the MIC of copper-loaded chitosan nanoparticles was 9 and 21 µg/mL against Gram-negative and Gram-positive bacteria, respectively (Du et al. 2009). It was also reported that unloaded chitosan nanoparticles and loaded with Cu2+ inhibited the growth of various microorganisms and exhibited higher antibacterial activity than chitosan solution or doxycycline (positive control) with MIC values of less than 0.25 µg/mL (Qi et al. 2004).

The MIC of Chitosan and its nanoparticles varies considerably, depending on a number of factors including the type of chitosan, its degree of polymerization, its molecular weight, solvent used (Jia et al. 2001), temperature, medium composition, the bacterial strain examined, and the growth phase of the bacterial strain. Antibacterial activity is also inversely affected by pH, with higher activity observed at lower pH values (No et al. 2002; Chen and Cooper 2002; Chung et al. 2004).

The precise antibacterial mechanism of chitosan is still unclear and remains to be understood, although it may involve the interaction of cationic chitosan with the anionic cell surface, which would alter membrane permeability and increase leakage of cellular material from the cell (Liu et al. 2004; Kong et al. 2010). Another potential mechanism involves chitosan’s entry into the nucleus, interference with mRNA synthesis, and ultimately, inhibition of protein synthesis (Rabea et al. 2003). Chitosan also acts as a chelating agent, selectively binding trace transition metals; this activity could also inhibit microbial growth and toxin production (Cuero et al. 1991). The chelating activity of chitosan contributes to the breakdown of the cytoplasmic membrane barrier and the chelation of trace metal cations, thus interfering with factors that are necessary for microorganism growth (Helander et al. 2001; Vishu Kumar et al. 2004; Chung et al. 2004).

Many studies have shown the significance and advantages of nanoparticles (Jung et al. 2000). These can be produced using a variety of methods including ionotropic gelation, microemulsion, emulsification solvent diffusion, and polyelectrolyte complex method. The ionotropic gelation technique is particularly attractive because it is non-toxic, controllable, and convenient, and does not involve the use of organic solvents (Agnihotri et al. 2004). It is based on the ionic interactions between the positively charged primary amino groups of chitosan and the negatively charged groups of a polyanion, such as sodium tripolyphosphate (TPP). TPP is the most extensively used ionic cross-linking agent, because of its non-toxic and multivalent properties (Shu and Zhu 2002). Chitosan, its derivatives, and nanoparticles have a great potential to inhibit bacterial growth. Most previous investigations focused on chitosan and its derivatives, and involved numerous techniques including outer membrane permeability assay [N-phenylnaphthylamine (NPN) uptake]; inner membrane permeability assay (β-galactosidase activity); lipopolysaccharide (LPS) release, alkaline phosphatase (ALP), and glucose-6-phosphate dehydrogenase (G6PDH) assays; measurements of intracellular material leakage (OD260 nm, OD280 nm) and electrical conductivity; SDS-PAGE; DNA fragmentation analysis; scanning electron microscopy; transmission electron microscopy; and atomic force microscopy (Didenko et al. 2005; Helander et al. 2001; Chung et al. 2004; Choi et al. 2001; Vishu Kumar et al. 2007; Liu et al. 2004). However, the antibacterial properties of chitosan nanoparticles (CS–TPP NPs) still require characterization because the smaller size of CS–TPP NPs might lead to enhanced antibacterial effects. The present study focused on the relationship between the particle size and antibacterial activity of CS–TPP NPs fabricated using two different molecular weights and concentrations of chitosan. The particles were loaded with neither drug nor metal. Emphasis was also given to highlight the antibacterial mechanism of action.

Materials and methods

Chemical reagents

Low-molecular-weight (LMW) 70-kDa chitosan with 75–85 % deacetylation and high-molecular-weight (HMW) 375-kDa chitosan with ≥75 % deacetylation were purchased from Sigma-Aldrich (USA). Penta-sodium tripolyphosphate (TPP), Muller–Hinton (MH) agar, and broth were supplied by Merck (Germany). Egg phosphatidylcholine, N-phenylnaphthylamine (NPN), and O-nitrophenyl-h-d-galactoside (ONPG) were also obtained from Sigma-Aldrich (USA).The cell viability kit was purchased from BD (USA). All other chemicals and reagents were of the highest commercial grade available.

Preparation and physical characterization of CS–TPP NPs

CS–TPP NPs were prepared by ionic gelation method (Calvo et al. 1997) with some modification. Chitosan was dissolved in an aqueous solution of acetic acid (0.2 mg/mL) stirred overnight at room temperature. TPP was dissolved in ultrapure water at a concentration of 0.1 mg/mL. The chitosan and TPP solution were then passed through a syringe filter (pore size 0.45 μm, Millipore, USA) to remove residues of insoluble particles. At room temperature, 1 mL of TPP solution was added dropwise using a pipette to 3 mL of (2–4 mg/mL) LMW and HMW chitosan solutions under a constant magnetic stirring at 900 rpm for 45 min. CS–TPP NPs were washed and harvested by ultracentrifugation thrice (Beckman Ultracentrifuge, USA) at 35,000×g at 10 °C for 30 min. Supernatants were discarded and the nanoparticles pellets were resuspended in 0.02 % acetic acid solution, adjusted to pH 6 using 10 % NaOH solution to a final concentration of 1 mg/mL.

Measurement of mean particle size (Z-average), polydispersity index (PDI), and zeta potential was carried out by a Zetasizer Nano ZS-90 (Malvern, UK). The measurements were performed in triplicate at a temperature of 25 °C. Samples were appropriately diluted with distilled water prior to measurement.

Determination of MIC and minimum bactericidal concentration

The MIC was determined by the microtiter broth dilution method (Andrews 2001). Three strains of Gram-positive bacteria including Staphyloccocus aureus, Bacillus cereum, and Bacillus subtilis and three strains of Gram-negative bacteria including Escherichia coli, Acinetobacter schindleri, and Pseudomonas aeruginosa were investigated. In brief, inocula of bacteria were prepared by adjusting an overnight culture to 5 × 105 cells/mL in Muller–Hinton Broth (MHB). Inocula (100 µL) were mixed with 100 µL of serial twofold dilutions of L1 (196 nm), L2 (394 nm), H1 (598 nm), and H2 (872 nm) CS–TPP NPs in MH broth in a 96-well plate. After incubation at 37 °C for 24 h, the antibacterial activity of CS–TPP NPs was determined on the basis of turbidity, which was considered as an indicator of bacterial growth. The MIC was defined as the lowest concentration of CS–TPP NPs at which no bacterial growth was observed after incubation. Subsequently, a 10-µL sample from wells with no growth was spread on MH agar plates for determination of the minimum bactericidal concentration (MBC). The concentration with the highest dilution exhibiting no growth on agar plates after incubation at 37 °C for 48 h was identified as the MBC. Two independent experiments were performed in triplicate to determine the MIC/MBC values.

Determination of time-dependent killing efficacy by Alamar Blue assay

Eschrechiacoli and S. aureus were used as a target to investigate the possible antibacterial potential of CS–TPP NPs. Overnight cultures of E. coli and S. aureus were adjusted in MHB to obtain 5 × 105 cells/mL and aliquots of 100 µL were incubated with L1, L2, H1, and H2 (CS–TPP NPs) at their previously determined MICs in 96-well plates at 37 °C. From each reaction condition, a 10-µL sample of the bacterial culture was taken out after 2, 4, 6, and 8 h and added to 100 µL of MHB containing (10 % v/v) Alamar Blue in new 96-well plates. Each 96-well plate was then placed in a microplate reader (NanoQuant infinite M200 PRO, Tecan, Switzerland), and the optical density (OD) at 570 and 600 nm was recorded for each well. OD data from each well and the time point were determined by subtracting the OD of the reagent blank. Growth inhibition curves were constructed by plotting the log10 cells/mL versus time. The experiment was performed in duplicate on at least two different occasions.

Transmission electron microscopy (TEM)

Overnight cultures of E. coli and S. aureus were centrifuged, and the pellets were washed and suspended in 0.1 M sodium phosphate buffer (pH 7.4) to achieve an absorbance of 0.4 at 600 nm. The cell suspensions were mixed with L1, L2, H1, and H2 (CS–TPP NPs) at a ratio of 1:1 (v/v) and incubated at 37 °C for 2 h, followed by centrifugation at 10,000×g for 10 min at 4 °C. The resulting pellets were washed twice with 0.1 M PBS (pH 7.2) and fixed with 3.0 % (v/v) glutaraldehyde in 0.1 M PBS. The samples were post fixed with 1 % w/v OsO4 in 0.1 M PBS for 1 h at room temperature and followed by three times washing with 0.1 M PBS, dehydrating separately at 4 °C for 10 min in a graded series of ethanol solutions (30, 50, 70, 80, 90, 100 %, v/v) and embedding in Agar 100 Resin. Specimens were then cut into thin sections with a diamond knife on an Ultracut (Ultramicrotome LE1CA UC6, Austria) and double-stained with saturated uranyl acetate and lead citrate. The grids were examined under a TEM (FEI Tecnai, Biotwin, Netherland) at an operating voltage of 120 kV.

Cell membrane integrity

Cell membrane integrity of E. coli and S. aureus was evaluated by determining the release of cellular constituents at 260 nm (Chen and Cooper 2002). Cultured bacteria were harvested, washed, and resuspended in sterile 0.5 % NaCl solution. The cell suspension was adjusted to achieve an absorbance of 0.6 at a wavelength of 420 nm. L1, L2, H1, and H2 (CS–TPP NPs) were mixed with these bacterial suspensions at a ratio of 1:1 (v/v), and the absorbance at 260 nm was monitored using an UV-1800 spectrophotometer (Shimadzu, Japan).

Outer and inner membrane permeability assays

Outer membrane (OM) permeability was evaluated by performing the NPN assay while inner membrane permeability was determined by measuring the release of cytoplasmic β-galactosidase from E. coli into the culture medium using ONPG as the substrate (Ibrahim et al. 2000). E. coli cultures grown in MH broth were harvested by centrifugation at 10,000×g for 10 min at 4 °C, washed, and resuspended in 0.5 % NaCl solution. The final cell suspension was adjusted to obtain an absorbance of 1.0 at 420 nm. The different concentrations of L1, L2, H1, H2 (CS–TPP NPs), and control 0.02 % acetic acid solution preadjusted to pH 6 using 10 % NaOH solution; at a ratio of 1:1 (v/v) was mixed with the bacterial cell suspension separately for outer and inner membrane permeability assays. OM permeability was determined in the presence of 20 µL of 1 mM NPN and fluorescence was measured using F-4500 fluorescent spectrophotometer (Hitachi, Japan), with an excitation wavelength of 350 nm and an emission wavelength of 420 nm, while inner membrane permeability was measured in the presence of 150 µL of 30 mM ONPG. The production of o-nitrophenol over time was evaluated on the basis of the increase in absorbance at 420 nm using a UV-1800 spectrophotometer (Shimadzu, Japan). The control contained a bacterial cell suspension aliquot and 0.02 % acetic acid at a ratio of 1:1 (v/v) and 20 µL of 1 mM NPN in OM permeability assay and 150 µL of 30 mM ONPG in inner membrane permeability assay.

Interaction of CS–TPP NPs with phospholipid membranes

Phosphatidylcholine (PC) liposomes were prepared in phosphate-buffered saline (PBS) as previously reported (Ibrahim et al. 2000; Yang et al. 2002). PC was dissolved in a chloroform–methanol mixture at a ratio of 98:2 (v/v) by gentle stirring with a magnetic stirrer. The solvent was evaporated under a stream of nitrogen by a rotary evaporator, and a thin film was obtained on the wall of the glass vessel. Any residual solvent was removed by keeping the resultant PC in a vacuum oven at 40 °C for 48 h. The dried lipids were resuspended in PBS buffer (pH 6) to achieve a final concentration of 0.5 % (w/v), and then sonicated in an ice-water bath until the turbid milky solution became transparent. L1 and H1 were mixed separately with PC liposomes by gentle stirring for 1 h at a ratio of 1:1 (v/v) as previously described (Liu et al. 2004). The mixture was then centrifuged at 10,000×g for 10 min at 4 °C. The resultant precipitates were dried in a vacuum oven at 50 °C for 48 h, and finely divided before characterization. Fourier transform infrared (FTIR) spectra (4,000–400 cm−1) of PC, L1 and H1 (CS–TPP NPs), PC–L1 complex, and PC–H1 complex were taken on KBr pellets by Spectrum 100 FTIR (Perkin Elmer, USA).

SDS-PAGE analysis

Logarithmic phase E. coli cells (106 cells/mL) in MH broth were mixed with 100 µg/mL of L1 and H1 (CS–TPP NPs) at equivalent ratio and then incubated at room temperature for 3 h. Aliquots of 5 mL were withdrawn each hour, and then centrifuged at 10,000×g for 10 min at 4 °C. The resultant pellets were subjected to SDS-PAGE analysis as previously described (Li et al. 2002; Laemmli 1970) by using a 4 % stacking gel and a 10 % separating gel. Protein bands were stained with 0.1 % Coomassie brilliant blue G-250 and de-stained in 40 % (v/v) ethanol and 10 % (v/v) acetic acid. The apparatus Mini-PROTEAN Tetra electrophoresis system (Bio-Rad, USA) was connected to a (75 mA) constant electric current. A marker of known molecular weight (prestained protein marker: broad range, 7–175 kDa) was also loaded along with the samples.

Fluorescent staining and flow cytometric analysis of bacteria using cell viability kit

E.coli and S. aureus were incubated for 12 h with L1, L2, H1, and H2 (CS–TPP NPs) at their previously determined MICs at 37 °C. From each reaction condition, E. coli and S. aureus treated with L1, L2, H1, and H2 (CS–TPP NPs) were diluted to an approximate concentration range of 5 × 105 cells/mL in staining buffer (PBS, 1 mM EDTA, 0.2 % Pluronic™ F-68, 0.1 % sodium azide, maintained at pH 7.4) and filtered through 0.22-µm filter. 5.0 μL of propidium iodide (PI) and thiazole orange (TO) solution was added into 500 μL of cell suspension to a final staining concentrations of 420 nmol/L for TO and 43 μmoL/L for PI. Stained samples were placed in the dark at ambient temperature for 10–15 min prior to analysis by BD FACScan flow cytometer (USA). The threshold to detect the forward scatter channel (FSC) and side scatter channel (SSC) was set to the level of the lowest cell signals. The detected interference (particulate and electronic) was gated such that it was removed from the analyses.

Cytotoxicity study

The Chinese hamster lung fibroblast cells (V79 cell line) (ATCC, Manassas, USA) were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10 % FBS and 1 % penicillin–streptomycin at 37 °C in a humidified atmosphere with 5 % CO2. Cell density was 5 × 104 cells per well. After 24 h, cells were incubated with L1, L2, H1, and H2 CS–TPP NPs for 24 h. After that, growth media were removed, and the cells were washed twice with PBS prior to the addition of 100 µL of fresh growth media. A volume of 10 µL Alamar Blue reagent was added into all the wells and incubated for 48 h. Cell viability was determined using a microplate reader (NanoQuant infinite M200 PRO, Tecan, Switzerland) at 570 nm (A570) using the following formula (Raja et al. 2013):
$${\text{Cell viability}} = {\text{A57}}0{\text{ sample}}/{\text{A57}}0{\text{ control}} \times 100.$$

Statistical analysis

One-way analysis of variance (ANOVA) was performed for between- and within-group comparisons using SPSS 16, and differences were considered to be significant at a level of p < 0.05.

Results and discussions

Particle size, PDI, surface charge, and morphology of CS–TPP NPs

CS–TPP NPs were fabricated by complexation of polyanionic TPP and cationic chitosan by using ionic gelation. The particle size of L1 and L2 was remarkably smaller and their PDI and surface charge were lower than those of the H1 and H2. The particle size of L1 and L2 was remarkably smaller and their PDI and surface charge were lower than those of the H1 and H2.

The particle size of CS–TPP NPs increased with an increase in chitosan concentration. Particle size significantly increased from 196.88 ± 3.37 to 394.79 ± 4.03 nm when the LMW chitosan concentration increased from 2 to 4 mg/mL (Table 1, p < 0.05). A similar trend was observed for HMW chitosan, with the particle size increasing from 598.74 ± 9.07 to 872.45 ± 7.94 nm (p < 0.05). The surface charge of CS–TPP NPs also increased with an increase in chitosan concentration for both LMW and HMW CS–TPP NPs. HMW chitosan produced a slightly higher particle surface charge than that produced by LMW chitosan (p < 0.05), but no significant effect was observed on the antibacterial activity. The electrostatic interaction between TPP and chitosan was affected by their concentration. Therefore, moderate concentrations (2–4 mg/mL) of LMW and HMW chitosan and TPP (1 mg/mL) were optimized by turbidity test for the preparation of nanoparticles. Significant increases in particle size and zeta potential were observed with increase in concentration and molecular weight. Smaller particle size was predicted on the basis of the low molecular weight and lower concentration on the basis of the reduced viscosity, which led to increased chitosan solubility in acetic acid, compared to high molecular weight and higher concentration (Katas and Alpar 2006; Morris et al. 2011; Ing et al. 2012). The TEM images showed the morphological characterization of CS–TPP NPs (Fig. 1). LMW chitosan produced a better colloidal system with a narrow particle size distribution than HMW chitosan did. All the nanoparticles prepared in this study were considered stable, and this was attributed to the high surface charge values (>30 mV), which prevented nanoparticles aggregation (Fan et al. 2012; Mohanraj and Chen 2006).
Table 1

Particle size, PDI, surface charge, and MIC/MBC of L1, L2,H1, and H2 (CS–TPP NPs), n = 3

CS–TPP NPs

Conc (mg/mL)

Particle size (nm)

Polydispersity index (μ2/Γ2)

Zeta Potential (mV)

MIC/MBC (µg/mL)

Gram-positive bacteria

Gram-negative bacteria

S. aureus

B. cereus

B. subtilis

E. coli

A. schindleri

P. aeruginosa

L1 (LMW)

2

196.88 ± 3.37

0.40 ± 0.02

42.14 ± 0.15

6.25/12.5

3.13/6.25

1.56/3.13

12.5/25.0

12.5/25.0

3.13/6.25

L2 (LMW)

4

394.79 ± 4.03

0.45 ± 0.01

53.75 ± 0.89

12.5/25.0

6.25/6.25

3.13/6.25

12.5/50.0

12.5/50.0

3.13/12.5

H1 (HMW)

2

598.74 ± 9.07

0.92 ± 0.14

49.84 ± 0.26

12.5/25.0

6.25/25.0

6.25/25.0

25.0/50.0

12.5/50.0

12.5/25.0

H2 (HMW)

4

872.45 ± 7.94

0.89 ± 0.16

52.23 ± 0.55

12.5/25.0

12.5/50.0

6.25/25.0

25.0/50.0

12.5/50.0

12.5/25.0

https://static-content.springer.com/image/art%3A10.1007%2Fs11051-014-2517-9/MediaObjects/11051_2014_2517_Fig1_HTML.jpg
Fig. 1

Transmission electron micrographs of L1, L2, H1, and H2

MIC and MBC

CS–TPP NPs were prepared in different sizes ranging from 100 to 900 nm by using different concentrations (1, 2, 3, and 4 mg/mL) of LMW and HMW chitosan. Only four samples of those small, medium, large, and very large nanoparticles were selected which showed a significant difference in particle size, antibacterial activity and had prominent differentiation in the inner and OM permeability. The antibacterial activity of LMW CS–TPP NPs and HMW CS–TPP NPs was, therefore, assessed against six different strains by determining the MIC and MBC values (Table 1). The control showed that at a low concentration, acetic acid did not exhibit any additional adverse effect on bacterial growth, while gentamycin was used as a positive control. The results of MIC/MBC indicated the bacteriostatic effect of CS–TPP NPs on bacteria. L1 and L2 had better inhibitory effects than H1 and H2, notably the L1 with smaller particle size of 196 nm. Surface area increases with decrease in particle size. It might be the pivotal factor in the increased antibacterial activity of small nanoparticles. Besides, LI and L2 had a relatively narrow particle size distribution which also explains why their antibacterial activity was higher compared to HMW CS–TPP NPs. Heterogeneity of particle size might be the participating factor of the reduced antibacterial activity of HMW CS–TPP NPs.

The results also depicted that CS–TPP NPs exhibited higher antibacterial activity against Gram-positive bacteria than against Gram-negative bacteria. This was possibly because of differences in cell wall structure and composition. The higher negative charge on the cell surface of Gram-positive than that on the cell surface of Gram-negative leading to greater CS–TPP NPs adsorption and higher inhibitory activity (Chung et al. 2004). The OM of Gram-negative bacteria is composed of LPS, lipoprotein, and phospholipids, which serve as a penetration barrier against macromolecules and hydrophobic compounds. E. coli is relatively more resistant to hydrophobic antibiotics and exogenous agents (Kong et al. 2008). The cell wall of S. aureus comprises a thin layer of peptidoglycan and teichoic acid with plenty of pores, which leads to an easy entry of foreign molecules into the cell. It is difficult to directly compare these chitosan nanoparticles data with results obtained by other researchers regarding chitosan antibacterial activity because of the different experimental conditions, chitosan grades, and bacterial strains employed.

Time-dependent Alamar Blue killing assay

The growth inhibition rate of all CS–TPP NPs was estimated by Alamar Blue assay using a method reported by (Shiloh et al. 1997) and a modified version reported by Kudo and Yano (2003). A decrease in bacterial growth rate was observed in all the test samples, but the L1 and L2 showed higher killing rates for both E. coli and S. aureus (Fig. 2) (p < 0.05). All the nanoparticles inhibited bacterial growth more rapidly in the initial hours, after which the inhibition rate progressively decreased. It is a simple, rapid, economical, and suitable method that requires inexpensive instrumentation to determine bacterial growth. Alamar Blue converts from its non-fluorescent oxidized form to its fluorescent reduced form with the growth of bacteria in the environment. The rate of killing bacteria increased as the size of the CS–TPP NPs decreased.
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-014-2517-9/MediaObjects/11051_2014_2517_Fig2_HTML.gif
Fig. 2

Curves for time-killing effects of L1, L2, H1, and H2 on the growth of aE. coli and bS. aureus, n = 3

Transmission electron microscopy of bacteria

TEM showed no cell membrane changes in the control bacteria; the cell membranes were fairly intact, without any notable rupture on the surface. However, the effects were noticeable in E. coli and S. aureus treated with L1, L2, H1, and H2 for 2 h (Fig. 3). The evidence of bacterial membrane damage was strongly supported by the transmission electron micrographs. Despite the distinction between Gram-positive and Gram-negative bacteria cell membranes, significant interactions at the cell surface could be seen in the form of distinct disruption and compromised cell membranes of both E. coli and S. aureus bacterial cells. Marked degradation, irregular shape, pore formation on the cell surface, perforated hollow structures, and scratched cell membrane on one side were observed. These findings supported the suggestion that CS–TPP NPs breached the bacterial envelopes and affected the plasma membranes (Liu et al. 2004), leading to leakage of small intracellular substances, ultimately killing the bacterial cells.
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-014-2517-9/MediaObjects/11051_2014_2517_Fig3_HTML.jpg
Fig. 3

Transmission electron micrographs. a Untreated E. coli, b untreated S. aureus. a1a4 show E. coli and b1b4 show S. aureus treated with L1, L2, H1, and H2, respectively

Integrity of bacterial cell membranes

The amount of intracellular material released from E. coli and S. aureus treated with L1, L2, H1, and H2 is shown in (Fig. 4). The absorption at 260 nm increased for up to 80 min, and then decreased gradually. All CS–TPP NPs were capable of damaging the cell membranes of E. coli and S. aureus. However, more nucleic acid contents were released upon treatment with L1 and L2 than upon treatment with H1 and H2 (p < 0.05).
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-014-2517-9/MediaObjects/11051_2014_2517_Fig4_HTML.gif
Fig. 4

The ratio of release of cell contents, identified by absorption at 260 nm from aE. coli and bS. aureus treated with L1, L2, H1, and H2, n = 3

Outer and inner membrane permeability assays

The interaction of CS–TPP NPs with the E. coli OM was examined and evaluated on the basis of the change in NPN fluorescence. An increase in fluorescence, because of penetration of NPN into the OM, was recorded over time until there was no further increase in the emission intensity. The NPN uptake by E. coli exposed to CS–TPP NPs is shown in (Fig. 5). When CS–TPP NPs and NPN were mixed with E. coli suspensions, an instantaneous uptake of NPN was observed, with the uptake amount being maximal after about 6–7 min. Thereafter, NPN uptake was nearly unchanged until 10 min. NPN uptake was higher in the bacteria exposed to L1 and L2 than in those exposed to H1 and H2 (p < 0.05); this finding is in agreement with the cell membrane integrity results. The control showed no NPN uptake over 10 min (Helander et al. 2001), suggesting that high concentrations of chitosan were required to achieve a significant increase in NPN uptake.
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-014-2517-9/MediaObjects/11051_2014_2517_Fig5_HTML.gif
Fig. 5

The uptake of NPN (measured as fluorescence intensity in arbitrary units) by E. coli treated with either L1, L2, H1, and H2, or control (pH 6 acetic acid solution), n = 3

The inner membrane permeability of L1, L2, H1, and H2 was evaluated by treating E. coli, which caused cytoplasmic β-galactosidase release into the medium because of a compromised inner membrane. There was a lag time of 20 min, followed by a progressively higher release of β-galactosidase until a steady state was reached after 90 min (Fig. 6). In this study, a considerably higher release of β-galactosidase was observed from E. coli treated with L1 and L2, compared to that observed upon treatment of E. coli with H1 and H2. Moreover, the difference in release observed upon using different range of particle sizes of chitosan nanoparticles of the same molecular weight also indicated that the release was greatly affected by particle size (p < 0.05). Antibacterial mechanism of action of chitosan has been reported previously (Helander et al. 2001; Vishu Kumar et al. 2004; Chung et al. 2004; Rabea et al. 2003; Cuero et al. 1991) but the mode of antibacterial action of CS–TPP NPs is still unclear. Membrane permeability assays were performed, not only to explore the expected mechanism of action of CS–TPP NPs, but also to estimate the potential of damage and permeation through bacterial cell membranes. Many antimicrobial agents target the cytoplasm and bacterial cell membrane, which leads to reduced bacterial viability. Intracellular components such as K+ and PO43− tend to leach out first owing to their low molecular weights, followed by larger molecules such as DNA, RNA, and other materials. The release of these components, detected by absorption at 260 nm, indicates bacterial membrane damage (Chen and Cooper 2002).
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-014-2517-9/MediaObjects/11051_2014_2517_Fig6_HTML.gif
Fig. 6

Release of cytoplasmic β-galactosidase activity (indicated by absorbance at 420 nm) from E. coli cells treated with L1, L2, H1, and H2, or control (pH 6 acetic acid solution), n = 3

The E. coli OM is stabilized by LPSs and proteins, which are held together by electrostatic interactions with divalent cations. A hydrophobic probe such as NPN usually penetrates the damaged and functionally invalid OM, producing increased fluorescence (Liu et al. 2004; Jeon and Kim 2006). Inner membrane integrity was analyzed by evaluating cytoplasmic β-galactosidase release by using ONPG as a substrate. Other researchers have reported that the release of cytoplasmic β-galactosidase caused by chitosan was time- and dose-dependent (Liu et al. 2004; Xing et al. 2009). The present study’s finding of an increased rate of cytoplasmic β-galactosidase release from E. coli treated with smaller nanoparticles is in accordance with the findings of OM permeability studies.

Interaction of CTS–TPP NPs with an artificial phospholipid membrane

It has been suggested that the electrostatic interaction between the polycationic structure of chitosan and the primarily anionic components of the microorganism’s surface LPS plays a prime role in antibacterial activity (Kong et al. 2010). An artificial membrane of egg PC was, therefore, developed to investigate this interaction mechanism with CS–TPP NPs (Liu et al. 2004). The intramolecular interaction between CS–TPP NPs and PC liposomes was confirmed by FTIR spectroscopy (Fig. 7). CS–TPP NPs show specific amide I and amide II bands at 1,653 and 1,564 cm−1, respectively (Zhang et al. 2004). The existence of primary amines is indicated by a dominant N–H stretch at 3,324 cm−1 and an intense N–H bend from 1,650 to 1,440 cm−1, which shifted to 3,430 and 3,431 cm−1 in PC–L1 and PC–H1 complexes, respectively. Decreased absorption was also observed in the 1,650 to 1,440 cm−1 region in the PC–L1 and PC–H1 complexes.
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-014-2517-9/MediaObjects/11051_2014_2517_Fig7_HTML.gif
Fig. 7

FTIR spectra of PC liposomes, L1, H1, PC–L1 complex, and PC–H1 complex

It was previously reported that the asymmetric and symmetric bands of phosphoryl groups appeared around 1,240–1,220 and 1,085 cm−1, respectively (Gomez-Fernandez and Villalain 1998). In this study, the absorption bands from stretching of P=O bonds appeared at 1,240 cm−1 and shifted to 1,232 and 1,229 cm−1, and increased absorption at 1,089 and 1,087 cm−1 in the PC–L1 complex and PC–H1 complex, respectively. The C=O stretching band of PC liposomes appeared at 1,743 cm−1. Another important difference was the formation of new peaks by the coupling of this 1,743 cm−1 stretching band of PC liposomes with 1,653 cm−1 from CS–TPP NPs at 1,736 and 1,737 cm−1 for the PC–L1 complex and the PC–H1 complex, respectively. This provided evidence for an intramolecular interaction between PC liposomes and CS–TPP NPs. The FTIR analysis of CS–TPP NPs, PC liposomes, and their complexes demonstrated that mainly the carbonyl and phosphoryl groups of PC participated in the interaction with the amino group of CS–TPP NPs. This finding further supported the previous finding that suggested that the phosphoryl group in PC participated in the interaction with chitosan (Liu et al. 2004; Li et al. 2010).

SDS-PAGE

The cellular protein contents obtained by centrifugation from E. coli treated with L1 and H1 were analyzed (Fig. 8). As the incubation time increased (1–3 h), the SDS-PAGE patterns of E. coli treated with CS–TPP NPs appeared shallow and fade, while the normal bacteria displayed apparently strong and clear protein bands (Tao et al. 2011). The gradual decrease in the cellular protein content showed release of proteins into the cell-free supernatant over time. The results also depicted that protein bands were almost disappeared after 3 h and these characteristics were more distinct in the patterns from bacteria treated with L1 as compared the bacteria treated with H1. The SDS-PAGE analysis of bacterial proteins indicated that CS–TPP NPs decreased the cellular protein contents by cell disruption and permeation.
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-014-2517-9/MediaObjects/11051_2014_2517_Fig8_HTML.gif
Fig. 8

SDS–PAGE of E. coli cells treated with L1 and H1 (M indicates the protein marker)

Flow cytometric detection of dead and surviving bacteria

The cell viability kit contains a combination of dyes that can easily distinguish live and dead bacterial cells by flow cytometric analysis. Using BD Cell Quest software, the TO and PI fluorescence of E. coli and S. aureus cells were analysed. Following the kit instructions, the cells were gated in such a manner that all the interferences were removed and only the live, injured, and dead bacteria were analysed and quantified. The data showed that all the L1, L2, H1, and H2 killed both E. coli and S. aureus (Fig. 9). The killing efficacy was negatively correlated with the particle size. The ratio of the surviving and injured bacteria was more in the samples treated with larger particle size CS–TPP NPs as compared to the small particles. Fluorescent staining and flow cytometry are a commonly employed and excellent method for the analysis of mammalian cells by labeling with immunofluorescence or other affinities. Flow cytometry has now become a well-established tool for analysis of yeast and bacterial cells (Feldhaus and Siegel 2004; Daugherty et al. 2000). TO is a permeant dye and enters all live and dead cells, while PI is a large molecule that penetrates the bacteria when membrane integrity has been compromised (Boyd et al. 2003) and these cells are normally considered as dead. The uptake of PI is an indication of both cell injury and death (Berney et al. 2008). The TO solution can stain all cells while PI stains only dead cells.
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-014-2517-9/MediaObjects/11051_2014_2517_Fig9_HTML.gif
Fig. 9

Flow cytometric analysis of L1, L2, H1, and H2 treated with E. coli and S. aureus

Cytotoxicity against V79 cells

Cytotoxic effect of L1, L2, H1, and H2 on V79 cells was investigated by using Alamar Blue cell viability assay. There was no significant difference in the cytotoxicity of L1, L2, H1, and H2 after 24 h in V79 cell line; over 95 % cell viability was observed for all the samples in comparison to untreated cells, p < 0.05 (Fig. 10). In this study, particle size had no significant effect on cell viability. This study, therefore, illustrated that CS–TPP NPs were non-toxic and could be exploited as either an excellent candidate or a potential vehicle for the antimicrobial drugs.
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-014-2517-9/MediaObjects/11051_2014_2517_Fig10_HTML.gif
Fig. 10

Cytotoxic effect of L1, L2, H1, and H2 CS–TPP NPs on V79 cells. n = 3

Conclusions

The mode of antimicrobial action of chitosan nanoparticles is not a simple mechanism but a complex event driven process. The present approach has provided numerous evidences that CS–TPP NPs destabilize and disrupt bacterial membrane leading to leakage of cellular components. This might cause ultimate dysfunctions to the whole cellular apparatus. The study also demonstrated different antibacterial potential against both Gram-positive and Gram-negative bacteria with respect to their particle size. The particle size and surface charge of CS–TPP NPs increased with increasing concentration and molecular weight of chitosan. Particle size and molecular weight are significant factors as smaller nanoparticles had lower MIC and MBC values, indicating superior antibacterial effects. Studies of antibacterial assessment, cell membrane integrity, inner and OM permeability, TEM, SDS-PAGE, flow cytometry analysis, and cell viability indicated that between the different sizes, the L1 with particle size of 196 nm demonstrated higher antibacterial activity with better permeation through bacterial cell membranes. The ability to enhance the antibacterial activity by controlling the particle size which can be varied by concentration and molecular weight, exhibits that CS–TPP NPs can be used as an antibacterial adjunct. The effect of surface area and pH variation on the antibacterial activity CS–TPP NPs would be part of our future study to reinforce the presented work.

Acknowledgments

This work was supported and funded by a grant (02-01-02-SF0737) from the Ministry of Science, Technology and Innovation (MOSTI), Malaysia and Universiti Kebangsaan Malaysia.

Copyright information

© Springer Science+Business Media Dordrecht 2014