Functionalized Magnetic Nanoparticles and Their Effect on Escherichia coli and Staphylococcus aureus

Magnetite (Fe 3 O 4 ) nanoparticles were prepared using coprecipitation and subsequently surface-functionalized with 3aminopropyltriethoxysilane (APTS), polyethylene glycol (PEG), and tetraethoxysilane (TEOS). Nanoparticle morphology was characterized using scanning electron microscopy, while structure and stability were assessed through infrared spectroscopy and zeta potential, respectively. Average size of the nanoparticles analysed by dynamic light scattering was 89 nm, 123 nm, 109 nm, and 130 nm for unmodified magnetite and APTS-, PEG-, and TEOS-modified magnetite nanoparticles, respectively. Biological effect was studied on two bacterial strains: Gram-negative Escherichia coli CCM 3954 and Gram-positive Staphylococcus aureus CCM 3953. Most of modified magnetite nanoparticles had a significant effect on S. aureus and not on E. coli, whereas PEG-magnetite nanoparticles displayed no significant effect on the growth rate of either bacteria.


Introduction
Surface functionalized magnetic nanoparticles have been widely used in a range of biological applications [1][2][3].Magnetite (Fe 3 O 4 ) is easily degradable and is useful, therefore, in bioseparation and catalytic processes.Magnetite nanoparticles have also been extensively studied in biomedicine [4,5] due to their superparamagnetic properties, high biocompatibility, and lack of toxicity to humans.Magnetite nanoparticles possess high surface energy and thus tend to quickly aggregate.Such strong aggregation, however, may alter their adsorption properties and magnetic efficiency; hence the nanoparticles are frequently coated with an organic or inorganic layer to prevent aggregation.Such coatings not only stabilize the magnetite nanoparticles but can be easily used for further functionalization.Uncoated magnetite nanoparticles are also known to be highly susceptible to leaching under acidic conditions; hence several methods have been developed for the preparation of magnetic nanoparticles coated with a polymer, such as polyethylene glycol (PEG) and silica-containing organic material [6][7][8] in the form of a core/shell structure, with the silica/PEG shell coated onto magnetic nanoparticles [9,10].Described coating enhances hydrophilicity and improves biocompatibility [11,12]; the core/shell structure [13] has a number of attractive properties, including high adsorption capacity and chemical and thermal stability [14].As the shell provides active groups on its surface that make available binding sites for enzymes, proteins, or drugs [15], magnetic nanoparticles have the potential to serve as drug carriers that can selectively target cancer cells, for example, and provide controlled release of chemotherapeutics [16,17].Magnetite nanoparticles coated with aminopropyltriethoxysilane (APTS) have many applications as adsorbent layers for removal of aqueous heavy metals during waste water treatment [18,19].Finally, the nontoxic nature of PEG-modified nanoparticles may be useful for more efficient biotechnology application [20].
Coprecipitation, sol-gel, and microemulsion are some of the most common methods for superparamagnetic magnetite nanoparticle synthesis, with coprecipitation being the most simple and economic method [21,22].It is based on the mixing of probably Fe 3+ and Fe 2+ at a 2 : 1 molar ratio in a highly basic solution, with the size and shape of the magnetite nanoparticles produced depending on the type of salt used, the reaction temperature, pH, and ionic strength of the media.A common method for coating magnetite nanoparticles with a uniform silica shell is the sol-gel process.This process makes use of base-catalyzed hydrolysis and condensation of tetraethoxysilane (TEOS) [23].The shell thickness of these silica-coated magnetite nanoparticles can be adjusted by controlling the amount of TEOS used [24].
The use of nanoscale materials has also attracted increasing concern due to the potential for environmental risk of toxicity.Several studies have been performed to evaluate the toxicity of magnetite nanoparticles on eukaryotic organisms, with surfactant-modified magnetite nanoparticles displaying negligible toxicity [25,26].Few studies have been published on the toxicity of magnetite nanoparticles to bacteria [27][28][29][30][31][32][33]; hence it is important to study the toxic effects of modified magnetite nanoparticles on more bacterial strains in order to fill gaps in our knowledge.
In this study, we prepared a range of magnetite nanoparticles with different surface functionality by modifying the nanoparticle surface with either APTS, PEG, or TEOS.The functional groups, surface charge, diameter, and morphology of the nanoparticles were characterized, together with their biological effect on Gram-negative Escherichia coli and Gram-positive Staphylococcus aureus.O), PEG (PEG6000, ≥95%), APTS (≥97%), and TEOS (≥99%) were purchased from Sigma-Aldrich and used as received.Nanoparticle structure and stability were assessed through infrared spectroscopy and zeta potential, respectively.Fourier transform infrared spectroscopy (FT-IR) was performed using the Tensor 27 Infrared Spectrometer (Bruker, USA), while zeta potential measurements were performed using a Zetasizer Nano analyzer (Malvern Instruments, USA) at pH 7. Dynamic light scattering (DLS) analysis was employed to measure the hydrodynamic diameters of magnetic nanoparticle aggregates in DI water using a Zetasizer Nano DLS unit.Microscopy images were obtained through scanning electron microscopy (SEM) using a Zeiss ULTRA Plus field-emission SEM equipped with a Schottky cathode.The images were analyzed using Smart SEM software v5.05 (Zeiss, Germany) for imaging operated at 1.5 kV.

Synthesis of Unmodified Magnetite
Nanoparticles.The unmodified magnetite nanoparticles were produced from an aqueous solution of FeCl 3 ⋅6H 2 O and FeCl 2 ⋅4H 2 O using the coprecipitation method [21,22].FeCl 2 ⋅4H 2 O (1.9 g) and FeCl 3 ⋅6H 2 O (5.4 g) at an Fe 3+ /Fe 2+ molar ratio of 2 : 1 were dissolved in deionized water (DI; 100 mL) and heated to 70 ∘ C. Ammonium hydroxide (6 mL) was quickly added to the solution, which immediately produced a deep black magnetite precipitate.The suspension was stirred for 30 min at 70 ∘ C. The product was washed several times with distilled water, following which the magnetite nanoparticles were dried in a rotary evaporator at 40 ∘ C (25 mbar) until a powder was formed.The powdered nanomaterial was stored in the dark at room temperature prior to modification.
APTS-modified magnetite nanoparticles were prepared according to Ming et al. [34,35], with minor modification.A 0.0128 M of magnetite solution (25 mL), which was prepared from first experiment, was diluted to 150 mL with ethanol (absolute) and 1 mL DI water.This solution was then treated in an ultrasonic bath (28 kHz at 25 ∘ C) for 1 h, whereupon APTS (35 L) was added and stirred rapidly for 2 h.The resulting liquid was washed with ethanol five times and then dried in a vacuum at room temperature until a powder was formed.
PEG-modified magnetite nanoparticles were prepared by dissolving 1.99 g of FeCl 2 ⋅4H 2 O and 3.24 g of FeCl 3 ⋅6H 2 O in 50 mL of DI water (Beaker I) and 30 mL of ammonium hydroxide in 50 mL of DI water (Beaker II).Subsequently, 2.5 g of PEG 6000 was dissolved in 100 mL of DI water and the liquid stirred at room temperature.The PEG solution (25 mL) was added to both beakers (I and II) and stirred in order to obtain a homogenous solution.The contents of Beaker II were added dropwise to Beaker I until pH 9 was reached.Formation of the magnetic nanoparticles was confirmed through a color change in the solution [36,37].The nanoparticles were separated out by centrifuging (28 kHz for 30 minutes) and the resultant precipitate was dried for 24 hours at room temperature.The dried powder was then redissolved in the remaining 50 mL of PEG solution and placed in an ultrasonic bath for about 30 minutes, whereupon it was again centrifuged to obtain the magnetic PEG-nanoparticles, which were then washed several times with DI water and dried to a powder under vacuum at room temperature.
TEOS-modified magnetite nanoparticles were prepared by dispersing approximately 30 mg of freshly prepared magnetite nanoparticles in 30 mL of ethanol and 6 mL of DI water, with the dispersion then being homogenized in an ultrasonic bath (28 kHz at 25 ∘ C) for about 10 minutes.TEOS (3.3 mmol) was then added to the mixture and sonicated for a further 20 minutes.Finally, aqueous ammonia (30 mmol) was added and the mixture again was placed in an ultrasonic bath for 60 minutes.The magnetic nanoparticles were magnetically separated and washed several times with DI water and then dried to a powder under vacuum at room temperature [38][39][40][41].

Dispersion of Magnetite Nanoparticles.
To assess the effect of functionalized magnetite nanoparticles on bacteria, stock solutions (10 g/L) of unmodified magnetite and APTS-, PEG-, TEOS-modified magnetite nanoparticles were prepared from powdered material (see above) by dispersing in sterilized DI water in a sterile glass tube and vortexing (IKA vortex3, Germany) for five minutes.Each suspension was prepared freshly before testing for toxicity.

Bacterial Strains and Culture Media. Bacterial strains of
Gram-negative E. coli CCM 3954 and Gram-positive S. aureus CCM 3953 were obtained from the Czech Collection of Microorganisms, Masaryk University, Brno, Czech Republic.The bacterial inocula were always prepared freshly from a single colony growing overnight in a soya nutrient broth (Sigma Aldrich) at 37 ∘ C. The culture was adjusted to 0.01-0.02optical density (OD) at 600 nm (OD 600 ) using the DR6000 UV-Vis spectrophotometer (Hach Lange, Germany), immediately before performing the antibacterial experiments.
2.5.Bacterial Growth Rate.The freshly prepared bacterial cultures were transferred to 30 mL of soya broth and kept in 200 mL conical flasks.The magnetite suspensions (unmodified and APTS-, PEG-, TEOS-modified) were added to the bacterial culture at a range of final concentrations (0.05, 0.3, 0.6, and 1 g/L), each sample being prepared in replicate.Negative (bacterial cells only in growth media) and positive (magnetic nanoparticles only in growth media) controls were run in parallel.All samples were incubated for six hours at 37 ∘ C. Subsamples were taken every two hours for OD 600 measurement.To prevent cross contamination, each bacterial strain was tested on different days.Oxidative/reductive potential (ORP) and pH were measured during the experiments using a WTW multimeter (Germany).
Effect of nanoparticle concentration on bacterial growth rate () was calculated for each nanoparticle type based on the equation: (%) = (  −  )/  ×100, where  is inhibition,   is mean value of growth rate () of the control, and   value is the growth rate of the culture affected by the nanoparticles.The bacterial growth rate was defined by  linear regression of cell density (OD 600 ) versus incubation time (hour).The EC10 value (effective concentration at 10% inhibition) was obtained by plotting % versus concentration of nanoparticle tested.

Determination of Colony Forming Units (CFU).
Number of E. coli and S. aureus CFU was determined using the same unmodified and APTS-, PEG-, and TEOS-modified magnetite (1 g/L) samples used for growth rate measurement.The bacterial strains were exposed to the nanoparticles in the dark for six hours at 37 ∘ C. Following incubation, 1 mL of the culture was transferred to a sterile agar plate and incubated for 24 h under the same conditions as the liquid cultures.All samples were prepared in duplicate and cultures without nanoparticles were cultivated in growth media as controls.

Bacterial Cell Morphology.
Cell morphology of E. coli and S. aureus was determined using the same unmodified and APTS-, PEG-, and TEOS-modified magnetite (1 g/L) bacterial samples used for determination of growth rate.Cells with nanoparticles (1 g/L) were stained using 4  ,6-diamidino-2-phenylindole (DAPI) and observed under an AxioImager fluorescence microscope (Zeiss, Germany) at 365/461 nm excitation/emission.Length of E. coli cells was measured for samples with nanoparticles (incubated for six hours) and for a control without nanoparticles.

Statistical Analysis.
All results were analysed using ANOVA (GraphPad Prism software; CA, USA).Dunnett's multicomparison test was used to compare differences between the means of E. coli and S. aureus growth rate, while values for CFU and bacterial cell length were compared using Sidak's multicomparison test.

Characterization of Magnetic Nanoparticles.
In each case, functional groups on the surface of the magnetite nanoparticles were detected by IR (Figure 1), with absorption peak at 580 cm −1 confirming the presence of an Fe-O bond related to the magnetite phase of magnetite nanoparticles [42].Bands at 800 cm −1 and 1090 cm −1 were due to symmetric and asymmetric linear vibrations of Si-O-Si, indicative of formation of a silica shell with APTS-and TEOS-modified magnetite.The absorption contribution from the free -NH 2 group of APTS-modified magnetite appeared at 3440 cm −1 .Absorption bands at 3400 cm −1 and 1200 cm −1 were assignable to O-H stretching and C-O of PEG-modified magnetite.
DLS indicated that unmodified magnetite nanoparticles display a wider size range than modified magnetite nanoparticles.Presence of an APTS, PEG, or TEOS layer on the nanoparticle surface increased average particle size (Figure 2), with an average increase of 89 nm for unmodified nanoparticles and 123 nm, 109 nm, and 130 nm for APTS-, PEG-, and TEOS-modified magnetite nanoparticles, respectively.
The zeta potential for unmodified magnetite nanoparticles was 36.9 mV and 12.2 mV, 23.1 mV, and 36.2 mV for APTS-, PEG-, and TEOS-modified magnetite nanoparticles, respectively.The nanoparticle's zeta potential (surface charge) indicates how effectively nanoparticles form stable or aggregated colloids during the colloidal phase.At low zeta potentials (close to zero), particles are no longer repelled strongly and colloids will aggregate due to attractive surface forces.Conversely, stable dispersions are formed at high zeta potentials (above ∼30 mV).This is of particular importance in water treatment and biomedicine applications where stable colloidal systems are required.SEM indicated slight differences in the morphology of unmodified and modified magnetite nanoparticles (Figure 3).Unlike modified magnetite nanoparticles, unmodified nanoparticles, which showed a broader size distribution, were usually agglomerated due to the high surface energy between the nanoparticles and dipole-dipole interactions.In addition, we detected a higher number of modified magnetite nanoparticles, providing further evidence for insertion of a surface layer on the magnetite.

Effect of Surface-Modified Magnetite on Bacteria.
The growth rates of Gram-negative E. coli and Gram-positive S. aureus control solutions (nutrient broth without nanoparticles) were 0.24 and 0.18 doublings/h, respectively.Unmodified magnetite nanoparticles had no significant effect on the growth rate of either bacteria.Interestingly, however, the growth rates of both bacteria were negatively correlated with suspension concentration (0.05, 0.3, 0.6, and 1 g/L) in APTS-, PEG-, and TEOS-modified nanoparticle solutions (Figure 4).
E. coli growth rate dropped to 0.15 doublings/h when incubated with APTS-magnetite (1 g/L), while growth rates for PEG-and TEOS-magnetite (1 g/L) were 0.22 and 0.18 doublings/h, respectively.At all concentrations, APTSmagnetite had a greater effect on E. coli growth rate than the other modified magnetite nanoparticles (Figure 4).Growth rates of S. aureus exposed to APTS-, PEG-, and TEOSmagnetite were 0.09, 0.13, and 0.1 doublings/h, respectively (Figure 4).
The EC10 for growth inhibition on both E. coli and S. aureus in unmodified and modified nanoparticle solutions indicated that modified nanoparticles had a greater effect on S. aureus than E. coli, with PEG-magnetite nanoparticles having the least effect on growth rate of either bacteria (Table 1).Both APTS-magnetite and TEOS-magnetite (1 g/L) had a significant negative effect (ANOVA;  < 0.0001) on both E. coli and S. aureus growth.
The effect of modified magnetite on bacterial growth rate was further supported by results obtained from the colony-forming assay.Compared to control samples with no nanoparticles, E. coli CFU declined to 73% with APTSmagnetite and 63.6% with TEOS-magnetite, whilst that for S. aureus declined to 27% with APTS-magnetite and 38% with TEOS-magnetite, after six hours of exposure (Figure 5).This confirms a similar effect determined for growth rate in both bacterial strains.Comparison of viable cells using ANOVA indicated that APTS-magnetite had a greater impact on S. aureus ( = 0.0019) than TEOS-magnetite ( = 0.0086).
On the other hand, both APTS-and TEOS-magnetite had no significant effect on E. coli ( = 0.2 and  = 0.13).
In order to elucidate these results further, we compared bacterial cell morphology using fluorescence microscopy after DAPI staining.Notably, E. coli showed rapid cell elongation (3.8 ± 0.12 m to 6.5 ± 0.5 m after six hours;  = 0.0001) when exposed to APTS-magnetite in soya broth media (Figure 6).Moreover, E. coli cells were clearly attached to the APTS-magnetite particles while the grape-like clusters of S. aureus were irregularly grouped with more single cells than cell clusters.

Factors Affecting Biological Stress.
Our results indicate that APTS-, TEOS-, and PEG-magnetite nanoparticles had a significantly greater biological effect on S. aureus than on E. coli.Many variables may impact on biological effect, including both biological (e.g., bacterial cell structure, cell growth rate, biofilm formation, stress/toxicity mechanisms) and chemical parameters (e.g., pH, ORP).Bacterial cell walls (peptidoglycan layers) are designed to protect the intracellular matrix while allowing for nutrient transport.In doing so they help to maintain the structural strength of the cell and stabilize the osmotic pressure of the cytoplasm and they are involved in binary fission during bacterial cell reproduction [43,44].Gram-positive cells possess a thick peptidoglycan layer of 20-50 nm [45], while Gram-negative cells contain a thin peptidoglycan layer.Notably, in our study, E. coli cells in contact with APTS-magnetite were unable to divide as the cells had increased in length from 3.8 m to 6.5 m.Further, Gram-negative cell walls comprise an outer membrane, which covers the surface membrane and is often resistant to compounds such as detergents, and a lipopolysaccharide layer, which is essential for cell viability and contributes to the negative charge of the membrane [46].An increased negative surface charge could reduce the likelihood of E. coli interacting with nanoparticles, as reflected in the lowered effect on Gram-positive S. aureus in this study.
Bacterial growth rate is a good method for indicating tolerance of bacteria to nanoparticles.Fast-growing bacteria, for example, are more susceptible to antibiotics and nanoparticles than slow-growing bacteria [47,48].The hindered growth observed in E. coli and S. aureus in this study could be related to expression of stress-response genes [49,50].Grampositive bacteria have the ability to form a biofilm to protect  themselves under stressful conditions (Pseudomonas sp., e.g., develop a biofilm in the presence of heavy metals) [51].In our study, Gram-positive S. aureus aggregated into large grapelike clusters after six hours of incubation with unmodified and APTS-, TEOS-, and PEG-modified magnetite.Previous studies have also reported that magnetite nanoparticles have a considerable capacity to penetrate biofilms [52,53].By interacting with the outer membrane, nanoparticles may cause loss of membrane integrity in bacteria and change the cell's structure.Such damage could lead to an increase in membrane permeability and leakage of intracellular constituents [54][55][56] and, indirectly, generate reactive oxidation species (ROS) through the Fenton reaction [57].Zero-valent iron, for example, causes cell membrane disruption in E. coli [54] and adsorption on cells and ROS generation in Bacillus subtilis var.niger and Pseudomonas fluorescens [58].In this study, magnetite nanoparticles are likely to have created  stressful conditions via ROS generation, which significantly inhibited S. aureus growth.Additionally, pH values obtained in parallel with bacterial growth experiments at the start and end of the experiment were similar to those of the controls (E. coli or S. aureus culture alone), with initial pH being neutral and dropping to around pH 5 after six hours, with or without the presence of magnetic nanoparticles (Figure 7).Only in the case of E. coli cultured with APTSmagnetite did the neutral pH remain unchanged after six hours, remaining within the optimum pH range of 6-7 for E. coli growth [59].Furthermore, the negative ORP of E. coli and positive ORP of S. aureus both showed similar trends in both treated and untreated cultures (Figure 8).In our biological effect tests, therefore, pH and ORP could not be considered as indicators for stressed bacterial states (Figures 7 and  8).In summary, magnetite nanoparticles surface-functionalized with APTS and TEOS had a significant biological effect on Gram-positive S. aureus, while PEG-nanoparticles did not.In contrast, none of the functionalized magnetite nanoparticles showed any statistically significant effect on Gramnegative E. coli.

Conclusion
Functionalized modified magnetite nanoparticles (TEOS-, PEG-, and APTS-magnetite) were prepared using coprecipitation and characterized with SEM and IR.Testing for biological effect indicated that PEG-magnetite can be considered as safe nontoxic material.TEOS-magnetite showed stronger effect only towards S. aureus.APTS-magnetite nanoparticles display a degree of antimicrobial activity, allowing for their use in bioapplications such as drug nanocarriers, where bacterial growth is undesirable.

Figure 4 :Figure 5 :Figure 6 :
Figure 4: Growth rate (doublings/h) of Escherichia coli (a) and Staphylococcus aureus (b) after six hours of incubation with unmodified and modified APTS-, PEG-, and TEOS-magnetite.The error bars were determined from  = 2.

Figure 8 :
Figure 8: pH values for Escherichia coli (a) and Staphylococcus aureus (b) cultures at time zero and after six hours of incubation with modified magnetite nanoparticles. 2

Table 1 :
The effective concentration at 10% inhibition, EC10 (g/L) of unmodified and modified magnetite nanoparticles determined for Gram-negative Escherichia coli and Gram-positive Staphylococcus aureus.