Synthesis, characterization and antibacterial activity of superparamagnetic nanoparticles modified with glycol chitosan

Abstract

Iron oxide nanoparticles (IONPs) were synthesized by coprecipitation of iron salts in alkali media followed by coating with glycol chitosan (GC-coated IONPs). Both bare and GC-coated IONPs were subsequently characterized and evaluated for their antibacterial activity. Comparison of Fourier transform infrared spectra and thermogravimetric data of bare and GC-coated IONPs confirmed the presence of GC coating on IONPs. Magnetization curves showed that both bare and GC-coated IONPs are superparamagnetic and have saturation magnetizations of 70.3 and 59.8 emu g⁻¹, respectively. The IONP size was measured as ∼8–9 nm by transmission electron microscopy, and their crystal structure was assigned to magnetite from x-ray diffraction patterns. Both bare and GC-coated IONPs inhibited the growths of Escherichia coli ATCC 8739 and Salmonella enteritidis SE 01 bacteria better than the antibiotics linezolid and cefaclor, as evaluated by the agar dilution assay. GC-coated IONPs showed higher potency against E. coli O157:H7 and Staphylococcus aureus ATCC 10832 than bare IONPs. Given their biocompatibility and antibacterial properties, GC-coated IONPs are a potential nanomaterial for in vivo applications.

Introduction

Given the increased bacterial resistance to conventional antibiotics, the development of alternative antibiotics has received considerable attention in the past decade [1, 2]. Also, infections resulting from the formation of bacterial slime or biofilms at the surface of an array of implantable devices present a major medical problem as a biofilm can promote bacterial resistance by restraining antibiotic treatment and host immune response [2]. The formation of microbial biofilms is mostly associated with the bacterial ability to adhere onto material surfaces through nonspecific interactions such as electrostatic, hydrophobic and van der Waals interactions; this ability is reversible at first and later becomes irreversible during the infection process [3]. Therefore, there is a need for materials that could reduce microbial adhesion.

Recent advancements in the field of nanotechnology have provided attractive solutions for synthesizing alternative antimicrobial agents and reducing biofilm formation [2, 4]. For instance, silver nanoparticles can reduce infections on implanted devices, preserve food and develop antimicrobial finished textile fabrics [5–7]. However, recent reports have raised concerns on potential environmental and health risks associated with a range of consumer products containing silver nanoparticles as antimicrobial agents [8, 9]. Superparamagnetic iron oxide nanoparticles (IONPs) with tailored surface chemistry have been widely used for various in vivo applications, including magnetic resonance imaging, tissue repair, immunoassay, detoxification, hyperthermia, drug delivery and cell separation [10]. The application of these IONPs in biomedical and bioengineering fields requires high magnetization and smaller size with narrow particle-size distribution to achieve uniform physical and chemical properties [11]. Most importantly, the surfaces of these IONPs need to be tailored by coating with nontoxic and biocompatible polymers, not only to overcome the agglomeration resulting from a large surface-to-volume ratio, but also to meet the demands of some specific applications.

Among various coating materials, chitosan has been widely used because of its nontoxic, biodegradable, biocompatible and antibacterial properties [12, 13]. However, its application is limited by its poor water solubility, as its highly deacetylated form is soluble only at low pH in organic acids [12, 13]. Efforts to improve the water solubility of chitosan include modifying its degree of acetylation to ∼50% and grafting with more hydrophilic groups [14–18]. Unfortunately, most approaches to improve its suitability for physiological conditions have resulted in a reduction of free amine-containing residues. For example, functional groups like succinyl, polyethylene glycol, dicarboxymethyl and carboxymethyl were all attached to chitosan through amine groups [15–18]. To remedy this problem, glycol chitosan (GC), a hydrophilic chitosan derivative soluble in water at any pH, was chosen as a coating material to modify the surface of iron nanoparticles. GC contains free amine groups along the polymer chain and behaves as a typical polycation at lower pH owing to the protonation of amine groups [18, 19]. Additionally, the coating of GC on IONPs can enhance both biocompatibility and antibacterial activity. The objectives of this study were to synthesize glycol-chitosan-modified superparamagnetic IONPs (GC-coated IONPs), characterize their structural and magnetic properties, and evaluate their antibacterial activity against four pathogens by comparing with two commercial antibiotics.

Materials and methods

Materials

All the chemicals were of analytical reagent grade and used without further purification. GC (≽60% purity as assayed by titration) with molecular weight above 400 (degree of polymerization indicating number of repeating monomeric units) was purchased from Sigma (St Louis, Missouri, USA). Ferric chloride (FeCl₃·6H₂O), ferrous sulfate (FeSO₄·7H₂O) and potassium bromide (KBr) were obtained from Nacalai Tesque (Kyoto, Japan), ammonium hydroxide (28%) was from J T Baker (Phillipsburg, USA) and hydrochloric acid was from Riedel de Haën (Seelze, Germany). An overhead mechanical stirrer from IKA^® (Staufen, Germany) was used for vigorous stirring during the synthesis of IONPs. Milli-Q deionized water was obtained from Millipore (Bedford, MA) and deoxygenated. For antibacterial study, two microbiological media, Mueller Hinton agar (MHA) and tryptic soy broth (TSB), were purchased from Difco (Michigan, USA). The American Type Culture Collection Centre (Virginia, USA) supplied the bacterial strains Escherichia coli (ATCC 8739) and Staphylococcus aureus (ATCC 10832), while E. coli O157:H7 (TWC 01) and Salmonella enteritidis (SE 01) were collected from clinical patients in Taiwan. The commercial antibiotics linezolid PZ0014 and cefaclor C6895 at a purity of ≽ 98% were from Sigma. A U1900 spectrophotometer (Hitachi, Tokyo, Japan) was used to determine the bacterial concentration via absorption at 610 nm, and a Shel Lab incubator (model 1535, Sheldon Manufacturing, Inc., Oregon, USA) was used for incubating the nanoparticles with bacterial cells.

Syntheses of bare and GC-coated IONPs

The syntheses of bare and GC-coated IONPs were based on a method reported by Kumar et al [20] and Inbaraj et al [21]. The method involves coprecipitation of ferrous and ferric iron salts in aqueous solution by adding a base. In brief, 6.1 g of ferric chloride and 4.2 g of ferrous sulfate were dissolved in deoxygenated deionized (DD) water followed by adding a few drops of concentrated hydrochloric acid. The mixture was then vigorously stirred at 2000 rpm with a concomitant bubbling of nitrogen gas to prevent the oxidation of ferrous ions, while raising the temperature to 85 °C. At 85 °C, 20 ml of ammonium hydroxide was rapidly added to precipitate iron oxide nanoparticles, and the color of the solution mixture turned from light orange to black. Next, 2.0 g of GC dissolved in 50 ml of DD water was added to the reaction mixture at 85 °C and the stirring was continued for 1 h. At the end of the reaction, the IONPs coated with GC were washed several times with DD water and decanted by using a permanent handheld magnet for vacuum drying at 0 mm Hg and 40 °C for 24 h. Bare IONPs were prepared by adopting the same procedure but skipping the addition of GC.

Structural and magnetic characterization of bare and GC-coated IONPs

Fourier transform infrared (FTIR) spectroscopy

The FTIR spectra of uncoated IONPs, GC-coated IONPs and pure GC were recorded on a Horiba FTIR spectrophotometer (model FT 730, Kyoto, Japan) in the wavenumber range of 4000–400 cm⁻¹ using KBr pellets prepared by homogenizing a small amount of nanoparticles with KBr crystals.

X-ray diffraction (XRD)

The XRD pattern was recorded on a Multiflex model Rigaku diffractometer (Tokyo, Japan) in the 2θ range of 20–70° using Cu–K_α radiation (λ=1.540 556 Å) at 40 kV and 40 mA. The average diameter of as-synthesized IONPs was calculated from the XRD linewidth using the Debye–Scherrer equation [20, 21].

Transmission electron microscopy (TEM)

The average size and morphology of bare and GC-coated IONPs were determined with a JEOL transmission field-emission electron microscope (JEM 2100F, Tokyo, Japan) operated at 80 kV.

Thermogravimetry (TGA)

Thermogravimetric analysis was carried out using a versa Therm HS Model Cahn TGA analyzer (Thermo Fischer Scientific, USA) to determine the weight percentage of GC coating on the surface of IONPs. The samples weighing between 5 and 15 mg were heated from 25 to 700 °C at a heating rate of 10 °C min⁻¹.

Vibrating sample magnetometry (VSM)

The magnetization measurements of as-synthesized IONPs were carried out at room temperature (301 K) using a vibrating sample magnetometer (DMS model 1660, ADE Technologies Inc., MA, USA). The applied magnetic field was varied between +13500 and −13500 Oe.

Total iron concentration

The total iron concentration was determined by dissolving the samples at room temperature (301 K), first with concentrated hydrochloric acid and then with concentrated nitric acid at a volume ratio of 2:1, and aspirating the solution into a flame atomic absorption spectrophotometer.

Antibacterial activity

Four pathogens including three Gram-negative bacteria (E. coli ATCC 8739, E. coli O157:H7 TWC 01 and S. enteritidis SE 01) and one Gram-positive bacterium (S. aureus ATCC 10832) were used to evaluate the antibacterial activities of bare and GC-coated IONPs by the agar dilution assay. The antibacterial activity was expressed as the minimum inhibitory concentration (MIC), that is, the lowest concentration of an antimicrobial agent that inhibits the growth of bacteria after overnight incubation. The test bacterial strains obtained in frozen form were thawed on ice for 20 min, followed by subculturing aerobically in 10 ml of TSB (pH 7.3) overnight, centrifuging at 5000 g for 10 min and suspending the pellet in 10 ml of phosphate-buffered saline (PBS, pH 7.2). Next, a uniform bacterial concentration of 10⁸ colony forming units (CFU) ml⁻¹ was obtained by adjusting the optical density at 610 nm. Antibacterial activity against test pathogens was evaluated using spot-on-lawn on MHA (pH 7.3). Soft agar was prepared by adding 0.75% agar in MH broth, inoculating the mixture with 1% of 10⁸ CFU ml⁻¹ of bacterial strains, and overlaying 10 ml of the product on MHA. Ten different concentrations of 0.5, 1, 2, 4, 8, 16, 32, 64, 128 and 256 μg ml⁻¹ were prepared by diluting bare and GC-coated IONPs separately in DD water. Similarly, the commercial antibiotics linezolid and cefaclor were also dissolved in DD water to obtain the same concentrations. Next, 100 μl of each concentration was dropped on the soft agar and incubated at 37 °C for 24 h. Ultrapure DD water without nanoparticles or antibiotics was used as the negative control (NC) and the MIC values were determined by measuring the zone of inhibition (ZOI) with a vernier caliper and substituting in the formula, ZOI_NP/AB−ZOI_NC, where NP and AB are nanoparticle and antibiotic, respectively. The results are reported as the mean of nine replicates (n=9).

Results and discussion

Characterization of bare and GC-coated IONPs

FTIR spectroscopy

Curves a and b in figure 1 show FTIR spectra of pure GC and GC-coated IONPs, respectively. There is a strong broad band at 3397 cm⁻¹ in GC-coated IONPs, which may be attributed to an overlap of O–H and N–H stretching vibrations [20, 21], while the peak at 2924 cm⁻¹ may be assigned to a C–H stretching vibration [21, 22]. The absorption band at 1628 cm⁻¹ corresponds to the amide bond of the undeacetylated part in GC and the N–H bending vibration of the free amine group of the deacetylated part in GC resulted in a peak at 1546 cm⁻¹ [19]. The peaks at 1110 and 1064 cm⁻¹ may be ascribed to a glycosidic linkage (ether bond) and C–N vibrations, respectively [22]. Interestingly, the amide absorption band in glycol chitosan shifted from 1649 to 1628 cm⁻¹ after interaction with IONPs. This shift of the amide IR band from higher to lower energy implies the interaction of IONPs with GC via nitrogen atoms, as pointed out by Bhattarai et al [23]. A sharp intense peak at 582 cm⁻¹ was unambiguously assigned to an Fe–O vibration of Fe₃O₄ [20, 21], which was absent in the IR spectrum of pure GC. However, all the other peaks described above were present for pure GC (figure 1(a)). The IR spectrum of bare IONPs (not shown) contained only two peaks at 3432 and 578 cm⁻¹, which may be attributed to an O–H vibration of adsorbed water and an Fe–O vibration, respectively [20, 21]. Altogether, these results confirm that the coating of GC did occur on the surface of IONPs.

Figure 1.

FTIR spectra of pure GC (a) and GC-coated IONPs (b).

X-ray diffraction

Figure 2 shows XRD patterns of uncoated and GC-coated IONPs. The six characteristic peaks at 30° (220), 35° (311), 43° (400), 53° (422), 57° (511) and 62° (440) correspond to iron oxide spinel structure (magnetite–Fe₃O₄ or maghemite–γ-Fe₂O₃) [24], indicating that the as-synthesized IONPs did not contain hematite (α-Fe₂O₃), goethite or other iron hydroxides. For bare and GC-coated IONPs, the lattice constant was estimated as 8.3945 and 8.3922 Å, respectively. These values are close to that of magnetite (8.3960 Å), implying that both the synthesized IONPs were mainly magnetite and not maghemite (8.3515 Å) [25]. This was further confirmed by comparing the 2θ value of the (311) peak at 35.45° for bare IONPs and 35.48° for GC-coated IONPs with the standard values for magnetite (35.423°) and maghemite (35.631°) [24]. By substituting the full width at half maximum of the main peak (311) into the Debye–Scherrer equation, the average particle sizes were calculated as 12.8 nm for bare IONPs and 10.3 nm for GC-coated IONPs [20, 21, 25].

Figure 2.

XRD patterns of bare (a) and GC-coated IONPs (b).

Transmission electron microscopy

Figures 3(a) and 3(b) show TEM images of bare and GC-coated IONPs, respectively, revealing nearly spherical shapes for both IONPs. From the particle-size distribution histograms, the average diameters of the bare and GC-coated IONPs were determined as 9.0 and 8.1 nm, respectively. These values are close to those estimated from the XRD line-width (12.8 and 10.3 nm), suggesting that the synthesized IONPs were nearly single crystals [11]. An increase in particle size was expected after the coating of GC on IONPs, and the observed slight decrease for GC-coated IONPs may be attributed to prolonged vigorous stirring at high speed (2000 rpm) after the addition of GC. A similar tendency was reported by Yu and Chow [26] who observed the same particle size of 8 nm for both bare and poly(methacrylic acid)-coated iron oxide nanoparticles. In addition, a slight difference in particle size between bare and GC-coated IONPs can be caused by instrumental error in TEM and XRD.

Figure 3.

TEM images of bare (a) and GC-coated IONPs (b) and the corresponding particle size distribution histograms.

Thermogravimetry

Figure 4 shows the TGA curves of bare and GC-coated IONPs. The initial weight loss of ∼0.5% occurring for both IONPs until 200 °C may be related to the removal of surface hydroxyl groups and/or adsorbed water [20, 21, 25]. Above 200 °C, the weight loss rose to 1% for bare IONPs and 9% for GC-coated IONPs, which may be attributed to the decomposition of amorphous iron hydroxides followed by iron oxide formation for the former and evaporation followed by the decomposition of GC coated on IONPs for the latter. By comparing the weight loss curves of bare and GC-coated IONPs, the amount of GC coated on IONPs was estimated as 8.5%.

Figure 4.

Thermogravimetric curves of bare (a) and GC-coated IONPs (b).

Vibrating sample magnetometry

Figure 5 shows the magnetization curves of bare and GC-coated IONPs recorded by VSM as a function of applied magnetic field (H, Oe). Both IONPs exhibit a well-defined superparamagnetic behavior at room temperature with zero coercivity or remanence, revealing that the thermal energy exceeded the anisotropic energy of IONPs [20, 21, 27]. Upon reducing the field, the magnetization value (M_s) dropped from the plateau and reached zero in zero field, implying that both IONPs are composed of single-crystal domains with aligned magnetic moments [10]. This result implies that the as-synthesized IONPs have high dispersibility (at H=0) and high sensitivity to the magnetic field (at H≠ 0), which are critical factors for their efficient separation and applications in biomedical and bioengineering fields. The saturation magnetization value for bare IONPs was 70.3 emu g⁻¹ at 15 kOe, which is much lower than that for bulk magnetite (89–92 emu g⁻¹) [20, 21]. This reduced magnetization could be explained on the basis of the small-particle surface effect [20, 25, 26]. The misalignment of atomic spins on the surface of particles is induced by reduced coordination and broken exchange bonds between surface spins. Consequently, the sum of disoriented surface spins is relatively insensitive to the applied magnetic field, with some of the surface spins being misaligned even at high magnetic fields. This surface effect becomes more pronounced for nanosized particles because of their large surface-to-volume ratio [20, 25, 26]. Thus, it is rational for the synthesized IONPs to show reduced M_s value as compared with the bulk material. Additionally, the amorphous fraction in the nanocrystals may contribute to the drop in the effective magnetic moment [20, 25]. For GC-coated IONPs, an M_s value of 59.8 emu g⁻¹ was obtained, which is smaller than that for the uncoated IONPs (70.3 emu g⁻¹). This difference can be attributed to the GC coating layer and the exchange of electrons between the surface Fe atoms and the polymer ligands [20, 25, 26], implying that the coating of GC did occur on IONPs. Nevertheless, the GC-coated IONPs still retain sufficient magnetism to facilitate separation within 20–30 s using a magnet. The difference in M_s value between bare and GC-coated IONPs indicates that the weight percentage of GC coating on IONPs is 10.5%, which is in agreement with the value of 8.5% estimated from TGA curves.

Figure 5.

Magnetization curves of bare (a) and GC-coated IONPs (b) recorded at room temperature (301 K).

Antibacterial activity

Superparamagnetic IONPs are attractive antibacterial agents because they can be magnetically driven and precisely targeted on the bacterial biofilm during in vivo applications. The antibacterial effects of both bare and GC-coated IONPs as well as the antibiotics linezolid and cefaclor were evaluated on the basis of the inhibition zone by agar dilution assay, and the MIC values obtained as a mean of nine replicates are summarized in table 1. Four pathogens were chosen in this study to evaluate the antibacterial activity of IONPs: E. coli and S. enteritidis are the most common Gram-negative food-borne pathogens, and E. coli O157:H7 is the most virulent of many E. coli strains. Whereas, S. aureus is a Gram-positive pathogen and is the prevalent cause of food intoxication. Of the three Gram-negative bacteria tested, both bare and GC-coated IONPs as well as GC were more efficient against the growths of E. coli and S. enteritidis than linezolid and cefaclor. For E. coli O157:H7 strain, both bare (256 μg ml⁻¹) and GC-coated IONPs (128 μg ml⁻¹) exhibited much higher MIC values than GC (4 μg ml⁻¹), linezolid (32 μg ml⁻¹) and cefaclor (16 μg ml⁻¹), implying that a relatively high dose of synthesized IONPs is necessary to attain the same antibacterial activity as GC and antibiotics. Nevertheless, it is worth pointing out that GC-coated IONPs were twice more efficient than bare IONPs in inhibiting the growth of E. coli O157:H7. Similarly, a lower antibacterial activity was observed for both bare and GC-coated IONPs (256 and 64 μg ml⁻¹, respectively) toward the Gram-positive S. aureus than GC (32 μg ml⁻¹), linezolid (4 μg ml⁻¹) and cefaclor (2 μg ml⁻¹). Yet, the GC-coated IONPs were four times more efficient against S. aureus than the bare ones. Compared with linezolid, the antibiotic cefaclor was twice more potent in inhibiting all the tested pathogens, probably due to the difference in the bactericidal mechanism [28]: cefaclor mainly inhibits cell wall synthesis by attaching to specific penicillin-binding proteins, whereas linezolid inhibits protein synthesis by binding to the bacterial 23S ribosomal RNA of the 50S subunit. Thus, linezolid prevents the formation of the functional 70S initiation complex, which is an essential component of the bacterial translation process [28].

Table 1.

Minimum inhibitory concentrations (μg ml⁻¹, mean of nine replicates) of GC, bare IONPs, GC-coated IONPs and two antibiotics against test pathogens determined by agar dilution assay.

Sample	E. coli	E. coli	S. enteritidis	S. aureus
	O157:H7
GC	4	<0.5	<0.5	32
Uncoated IONPs	256	<0.5	<0.5	256
GC-coated IONPs	128	<0.5	8	64
Linezolid	32	16	32	4
Cefaclor	16	8	16	2

The antibacterial activity of a material mainly depends on the surface charge and hydrophobic/hydrophilic nature of both the bacterial cell and adherent material [3, 12]. The antibacterial activity of bare IONPs against E. coli and S. enteritidis was much higher than those of the antibiotics. A plausible mechanism of the antibacterial action of the bare IONPs is that metal oxide (Fe₃O₄) may generate reactive oxygen species via the Fenton reaction, leading to lipid peroxidation, DNA damage and protein oxidation [4]. Additionally, owing to the positively charged surface, metal oxides may bind to negatively charged cell membranes through electrostatic interaction, thereby disrupting bacterial functions [2]. This phenomenon is proven by the fact that Gram-negative bacteria have a more negative charge on the surface than the Gram-positive ones [29]. Conversely, a relatively poor antibacterial activity shown for E. coli O157:H7 may be due to the reduced charge density, which concurred with the surface charge classification of ‘slightly charged’ by Li and McLandsborough [30]. A similar outcome of reduced antibacterial activity by bare IONPs towards S. aureus may be caused by the lower negative charge on the surface of Gram-positive bacteria.

Numerous studies have associated the antibacterial activity of chitosan and its derivatives with the interaction of protonated NH₂ groups in chitosan with the negative residues on the bacterial surface, resulting in the alteration of cell permeability, depletion of nutrient flow, and leakage of proteinaceous and intracellular constituents [12, 13]. More specifically, chitosan activates several defense processes in host tissue, binds more water, inhibits various enzyme activities and interferes with the synthesis of mRNA and proteins [12, 13]. Applying the same mechanism to this study, the electrostatic interaction can occur only when the NH₂ groups on GC are protonated, i.e. the pH of the bacterial medium is lower than the pK_a of GC (6.0) [31]. However, the pH of the bacterial medium was 7.3 in our experiment, implying that the NH₂ groups in GC may not be protonated and thus should prevent the electrostatic interaction between GC and the bacterial surface. In similar studies, Yang et al [14] and Tsai and Su [32] reported that the antibacterial activity of chitosan (pK_a = 6.5) towards E. coli and S. aureus declined following a rise in pH of the reaction medium from 6.0 to 7.0. Thus, factors other than surface charge may determine the antibacterial activity of GC-coated IONPs.

As mentioned earlier, the application of unmodified chitosan as coating material is limited owing to its poor solubility in water. However, the coating of IONPs with GC enhanced water solubility, and thus, the hydrophilic nature of GC-coated IONPs may significantly contribute to their antibacterial action, as the hydrophobic interaction between a microorganism and a substrate would facilitate microbial adhesion and proliferation [22, 33]. It has been well documented that the bacterial adhesion can be markedly enhanced when both bacterial and adherent surfaces are hydrophobic, but the adhesion is hindered if both surfaces are hydrophilic [32, 33]. In a recent study dealing with the bacterial adhesion on hydrophobic polystyrene and hydrophilic glass surfaces, Mafu et al [34] demonstrated that the adhesions of E. coli O157:H7, S. enteritidis and S. aureus were thermodynamically favorable on the former, but not on the latter surface. Likewise, Boulmedais et al [3] observed a 92% decrease in E. coli adhesion on hydrophilic anti-adhesive films composed of three bilayers of poly(l-lysine)/poly(l-glutamic acid)-grafted polyethylene glycol.

From the above discussion, the pronounced antibacterial activity of GC-coated IONPs toward E. coli and S. enteritidis may be accounted for by the hydrophilic nature of both the Gram-negative strains and nanoparticles. Similarly, the hydrophilic nature of E. coli O157:H7 may be responsible for the twofold higher antibacterial activity of GC-coated than bare IONPs. However, the difference in antibacterial activity between the two tested E. coli strains (ATCC 8739 and O157:H7) as affected by IONPs and antibiotics should be caused by the varying degree of hydrophilicity. Moreover, the extracellular polymeric substance produced by E. coli O157:H7 may mask the hydrophilic property by enhancing the adhesion ability of cells, as pointed out by Hassan and Frank [35]. Generally, the Gram-positive S. aureus strains are considered hydrophobic. However, according to the evaluation of 15 Staphylococcus strains by Reifsteck et al [36], the Wood 46 strain of S. aureus used in this experiment was classified as hydrophilic, which may account for a fourfold rise in antibacterial activity for GC-coated IONPs compared with the bare ones in this study. The bacterial adhesion involved in the antibacterial mechanism is a complex process, which primarily depends on the surface charge and hydrophobic/hydrophilic nature of both the bacterial cells and substrate. Thus, a detailed study is necessary to elucidate the molecular mechanism of bacterial growth inhibition by IONPs.

Although the antibacterial activity of glycol chitosan is reduced after coating with IONPs, particularly for E. coli O157:H7, S. enteritidis and S. aureus, the magnetic property of IONPs is vital for future biomedical applications as it may offer a precise delivery to infected organs or tissues. A rapid and efficient action can be expected as the magnetic field remains unabsorbed in the body, which facilitates access to deep regions in the living tissues [21]. Also, IONPs are advantageous over commercial antibiotics owing to their smaller dosage requirement and elimination of possible harmful side effects.

Conclusions

We have synthesized superparamagnetic iron oxide nanoparticles coated with glycol chitosan; the particles have an average size of 8 nm and a magnetization value of 59.8 emu g⁻¹. As shown in the MIC values, both bare and GC-coated IONPs were more efficient in inhibiting the growth of E. coli ATCC 8739 and S. enteritidis SE 01 than the antibiotics linezolid and cefaclor, whereas the antibiotics acted better against E. coli O157:H7 and S. aureus ATCC 10832. This study demonstrated the in vitro antibacterial activity of IONPs coated with glycol chitosan, which may have potential biomedical applications owing to their biocompatibility and antibacterial properties.