Abstract

Unmodified magnetic nanoparticles (MNPs) lack antibacterial potential. We investigated MNPs surface modifications that can impart antibacterial activity. Six MNPs species were prepared and characterized. Their antibacterial and antibiofilm potentials, surface affinity, and cytotoxicity were evaluated. Prepared MNPs were functionalized with citric acid, amine group, amino-propyl trimethoxy silane (APTMS), arginine, or oleic acid (OA) to give hydrophilic and hydrophobic MNPs with surface charge ranging from −30 to +30 mV. Prepared MNPs were spherical in shape with an average size of 6–15 nm. Hydrophobic (OA-MNPs) and positively charged MNPs (APTMS-MNPs) had significant concentration dependent antibacterial effect. OA-MNPs showed higher inhibitory potential against S. aureus and E. coli (80%) than APTMS-MNPs (70%). Both particles exhibited surface affinity to S. aureus and E. coli. Different concentrations of OA-MNPs decreased S. aureus and E. coli biofilm formation by 50–90%, while APTMS-MNPs reduced it by 30–90%, respectively. Up to 90% of preformed biofilms of S. aureus and E. coli were destroyed by OA-MNPs and APTMS-MNPs. In conclusion, surface positivity and hydrophobicity enhance antibacterial and antibiofilm properties of MNPs.

1. Introduction

Nanoparticles (NPs) are gaining attention as a new antimicrobial and antibiofilm approach. They may directly exhibit antibacterial effect or be used as antibiotic carriers [1]. Metal oxides NPs can be prepared easily and stand harsh conditions such as high temperature during sterilization. Metal oxide NPs are expected to overcome the organic molecule drawbacks such as pollution, residence in tissues, high cost, toxicity, and low stability. Their mechanism of action includes production of reactive oxygen species which damages cellular structures, alteration of membrane permeability, interruption of energy transduction, alteration of enzymatic activity, and DNA replication [2].

Iron oxide magnetic nanoparticles (MNPs) have several advantages due to their stability, low preparation cost, and biocompatibility, as well as their manipulation by a magnetic field [3]. MNPs types include hematite (α-Fe2O3), magnetite (Fe3O4), wüstite (FeO), and maghemite (γ-Fe2O3). The magnetite and maghemite types have proven biocompatibility. Both types are produced by thermal coprecipitation method [4, 5]. Magnetite Fe3O4 (black) is produced under anaerobic conditions while maghemite γ-Fe2O3 is the magnetite oxidation product (brown). MNPs range in size from few to hundreds of nanometers [6]. They have many bioapplications, such as magnetic bioseparation and detection of biological entities, diagnostic applications as magnetic resonance imaging, and therapeutic applications as targeted drug delivery and biological labels [7].

MNPs are generally very reactive and tend to aggregate quickly to decrease their surface energy which leads to alteration in their size and magnetic properties. Surface coating of MNPs modulates their aggregation, stability, and dispersion ability [4, 6].

Owing to their unique properties, several attempts were performed to investigate the antibacterial and antibiofilm potentials of MNPs. Previous studies showed that MNPs exhibited insignificant or no antibacterial activity, while some surface modifications were successful to impart antibacterial potential (Table 1). Understanding the effect of surface modification of MNPs will enable the optimum selection of surface coating materials offering the greatest antibacterial activity with the least toxicity.

Many bacterial strains produce slime which serves as a matrix in which bacteria are embedded leading to the formation of bacterial biofilm. Bacterial adhesion is mediated by electrostatic, dipole-dipole, H-bond, hydrophobic, and van der Waals interactions [8]. Biofilms promote antibiotic tolerance by reducing antibiotic entry into the bacterial cells. Bacteria in the biofilm can also grow slowly to adapt to depletion of nutrient and accumulation of waste [9]. Bacterial biofilm can adhere to surfaces or exist in flowing system like water columns [10]. Bacterial infection which forms a biofilm will be transformed from an acute to a chronic infection which is difficult to eradicate. Eradicating bacteria in a biofilm will require either mechanical removal or long time combination of high doses of antibiotics [11]. NPs with antibiofilm ability will greatly reduce the antibiotic use. Antibacterial NPs having ability to reduce bacterial biofilm formation can exert their function by preventing bacterial adhesion to surfaces and increasing the bacterial cell exposure to surrounding environment [9].

The present study aimed to compare the effect of hydrophobicity and surface charge modulation on the antibacterial and antibiofilm potentials of MNPs. MNPs with different surface fictionalizations that result in different surface charge and hydrophobicity were synthesized and compared in terms of antibacterial activity, inhibition of bacterial biofilm formation, destruction of preformed biofilm, and their safety to human cells. Comparison was performed against Gram-positive S. aureus and Gram-negative E. coli. To the best of our knowledge, this is the first attempt to compare the effect of hydrophobicity and charge modification on MNPs’ antibacterial potential and antibiofilm formation ability.

2. Materials and Methods

2.1. Material and Instrumentation

FeCl3, FeSO4, ammonia solution (25%), citric acid, and oleic acid were obtained from Al-Gomhoreya for Chemical Industries in Cairo, Egypt. Arginine (Arg) and amino-propyl trimethoxy silane (APTMS) were obtained from Sigma Aldrich (St. Louis, MO). Deionized (DI) water was produced in house by a Milli-Q system (Milford, Connecticut, USA). Mueller Hinton broth was obtained from Sigma Aldrich. E. coli (ATCC-8739) and S. aureus (ATCC-6538) strains and normal human epithelial (WISH) cells (ATCC-CCL25) were generously provided by VACSERA (Cairo, Egypt). Phosphate buffer saline (PBS) was purchased from Biowhittaker-Belgium. Dynatech Microplate Reader (MR 5000Er, West Sussex, UK) and a Jenway 6850 Spectrometer (Staffordshire, UK) were used.

2.2. Methods of MNPs Synthesis
2.2.1. Synthesis of Magnetic Iron Nanoparticles (MNPs)

Thermal coprecipitation method was adopted for the preparation of MNPs using FeCl3, FeSO4, and 35% ammonia solution (Figure 1). Briefly, 5.5 g of FeCl3 and 2.75 g of FeSO4 were weighed and dissolved in DI water (1 L). The solution was heated at 70°C for 30 min [12]. Five mL of ammonia solution was added until a black precipitate was formed (Magnetite Fe3O4). The reaction was continued for additional 10 min. Solution was evaporated to dryness in rotavap to remove adsorbed ammonia. Residue was then washed six times with DI water and surface charge was measured until shifted to negative (−20 mV) indicating complete desorption of surface ammonia. The MNPs were then dried in oven at 180°C and grinded before use. The powder turned into the brown maghemite (γ-Fe2O3) solid (supplementary information 1, in Supplementary Material available online at https://doi.org/10.1155/2017/3528295). Prepared MNPs (200 mg) were weighed and suspended in 100 mL DI water. The X-ray diffraction (XRD) analysis (supplementary materials 2) and transmission electron microscope (TEM) imaging (Figure 2) of the prepared particles were performed.

2.2.2. Preparation of Oleic Acid (OA) Functionalized MNPs (OA-MNPs)

OA-MNPs were synthesized according to the methods of [13, 14], with some modifications. Excess oleic acid (OA) was added (3 mL) to black magnetite particles with stirring for 1 h at 70°C. Two layers were formed: the upper OA layer was separated in a separating funnel and washed three times with DI water. MNPs were collected by a magnet and then washed 6 times with ethanol to remove excess OA. OA-MNPs were then tested for removal of uncoated OA by FTIR in terms of absence of C=O peak of OA indicating chemosorption of OA on MNPs (Figure 3(a)) [14, 15]. Chemosorption of OA was further verified by the shift in zeta potential (Figure 3(b)) and the lack of dispersion ability of particles in water (Supplementary material 3a and b). OA-MNPs were air dried to remove excess alcohol grinded and weighed (200 mg). OA-MNPs were then mixed with 40 μL of Tween 80 in a glass mortar and DI water was added dropwise for the first 10 mL and then portionwise until the volume was completed to 100 mL [13].

2.2.3. Preparation of Amine Coated MNPs (A-MNPs)

The black magnetite particles were extracted from their reaction mixture after being left for aging at room temperature for 3 h with the excess ammonia in closed condition. A-MNPs were washed with acidified DI water and then with cold DI water to remove excess ammonia and other inorganic components. The particles were charge measured and FTIR scanned. A-MNPs were then reconstituted in DI water (2 mg/mL) [16].

2.2.4. Preparation of Citrate Coated MNPs (CA-MNPs)

CA-MNPs were prepared as described in [17, 18] with some modifications. Solution of citric acid was prepared by dissolving 38.4 mg of citric acid in 1 L of DI water. The solution was heated at 80°C (Solution 1). FeCl3 (5.5 g) and FeSO4 (2.75 g) were weighed and dissolved in DI water (1 L). The solution was heated at 70°C for 30 min. Five mL of ammonia solution was added until black precipitate is formed. The reaction was continued for 10 min (Solution 2). Solution 2 was added on solution 1 dropwise with stirring. The mixture was heated for 20 min and washed with DI water. The resulting CA-MNPs were air dried, grinded, and examined for FTIR and zeta potential change. Finally, a suspension of 2 mg/mL was prepared for application [17].

2.2.5. Preparation of MNPs Coated with Arginine (Arg-MNPs)

Unfunctionalized MNPs (100 mg) were transferred into a glass mortar and 20 mg of arginine (Arg) was added followed by 1 mL HCl. The mixture was mixed using a pestle and diluted with DI water (100 mL) and then transferred into a stoppered conical flask for 6 h reflux. Arg-MNPs were separated using an external magnet and washed with DI water six times. Arg-MNPs were air dried followed by grinding. A 2 mg/mL solution was prepared [19].

2.2.6. Preparation of Silane Coated MNPs (APTMS-MNPs)

Amino-propyl trimethoxy silane (APTMS) was used as a functionalizing agent for MNPs. Two protocols were followed. In the first protocol (cold synthesis method), 200 mg of MNPs was weighed and suspended in 3 mL of DI water. The suspension was sonicated for 30 min for hydration and ensuring homogeneity (I). The MNPs suspension was then poured (100 μL portions) on a glass vial containing 800 μL of APTMS placed on an ice path. The mixture was left in ice bath for 3 h and then washed with DI water for 6 successive times. In the second method (hot synthesis method), 200 mg MNPs were suspended in 10 mL of acetone (I). APTMS (400 μL) were mixed with 10 mL of acetone (II). Solutions I and II were mixed by sonication for 90 min. The contents were then transferred to a glass stoppered conical flask and refluxed for 3 h followed by washing. From both methods, 2 mg/mL suspensions of MNPs were prepared in DI water. Zeta potential scanning was used to compare higher potential change induced by APTMS using both methods (supplementary information 4).

2.3. Antibacterial and Antibiofilm Activity of Synthesized MNPs
2.3.1. Antibacterial Activity Screening

Antibacterial effect of MNPs was examined against E. coli (ATCC-8739) and S. aureus (ATCC-6538). MNPs were diluted in Mueller Hinton broth to reach a final concentration of 100 μg/mL and incubated with 18–20 h subcultured bacterial strains at 37°C on a shaker (200 rpm) for 24 h. At the end of the incubation period, samples were obtained from each flask and 10-fold serially diluted in sterile saline. One hundred μL of each dilution as well as control was spread on the surface of 3 nutrient agar plates, incubated at 37°C for 24 h, and the average numbers of colony forming units (CFU/mL) were counted [20]. MNPs species showing promising antibacterial screening were imaged by scanning electron microscope (SEM) to compare their surface morphology.

2.3.2. Effect of Concentration on the Antibacterial Activity of MNPs

The OA-MNPs and APTMS-MNPs were prepared in different dilutions (25–400 μg/mL) and incubated with subcultured bacteria. CFU/mL was counted on surface of agar plates after 24 h incubation periods.

2.3.3. MNPs Inhibitory Potentials and Growth Kinetics

To evaluate the inhibitory activity of MNPs as well as growth kinetics on tested bacterial strains, culture turbidity was used as a measure of bacterial growth. OA-MNPs and APTMS-MNPs were double fold serially diluted in 96-well plate. Positive control wells were double fold serially diluted with DI water. All plates were inoculated with test bacteria as 10 μL/well (105 CFU/mL) except for negative control wells (blank); this was carried out to avoid interference caused by light-scattering properties of NP [21]. Plates were incubated at 37°C for 24 h with continuous shaking. Percentage inhibition was calculated according to Sachidananda et al. as follows [22]: is optical density of test bacteria with MNPs; is optical density of positive control wells containing DI water, bacteria, and media.

MIC50 is the lowest MNPs concentration that reduces the bacterial growth by >50%. Optical density (OD) was measured at 1, 2, 4, 6, and 24 h intervals using an ELISA plate reader at 600 nm.

2.3.4. Surface Affinity of MNPs to Bacteria

OA-MNPs and APTMS-MNPs were mixed with concentrated bacterial suspension in DI water (OD of suspension = 1.2) to reach a final concentration of 500 μg/mL. The suspension was prepared in DI rather than broth to reduce the bacterial growth during the experiment. However, all measurements were performed against suspension control of the same age. The mixture of MNPs and bacteria were allowed to interact together through incubation on a shaker (200 rpm) for a predetermined period of time. MNPs were collected using an external magnet. The bacterial suspension was mixed to resuspend the bacterial cells and samples were collected to measure their OD at 600 nm using spectrometer. The affinity was calculated from the decrease in OD which occurred upon treating bacterial suspension with MNPs relative to positive control bacteria without MNPs. Optical densities obtained after collecting MNPs alone with a magnet were considered as negative control [23]. The collected MNPs were scanned by FTIR and compared to control MNPs [24].

2.4. Antibiofilm Activity
2.4.1. Effect on Biofilm Formation

The effect of MNPs on biofilm formation was carried out by allowing bacterial strains to grow in presence of double fold serially diluted MNPs (OA-MNPs and APTMS-MNPs) in 96 multiwell plates. Plates were incubated at 37°C for 24 h. After incubation period, contents of the plates were discarded and plates were washed 3 times with PBS to remove unbound bacteria. Plates were inoculated with 100 μL/well crystal violet (0.15%) and incubated at room temperature for 30 min. Crystal violet (CV) was discarded; plates were washed again 3 times with phosphate buffer saline (PBS) and allowed to air dry completely. Biofilm formed in each well was resuspended in 97% ethanol (200 μL/well) and incubated at room temperature for 10 min. The solubilized biomass (150 μL) was transferred to sterile 96 well plates to be measured spectrophotometrically at 590 nm. Data were presented as percentage inhibition in biofilm growth in presence as well as in absence of MNPs [25].

2.4.2. Effect on Preformed Biofilm

In 96 multiwell plates, 100 μL bacterial suspension (105 CFU/mL) was inoculated in each well and incubated at 37°C for 24 h to initiate biofilm formation. After incubation period, plates were washed 3 times with sterile PBS to remove any unattached cells. Double fold serially diluted MNPs (OA-MNPs and APTMS-MNPs) were added in all wells except in negative control wells and incubated at 37°C for 24 h to evaluate the effect of MNPs on preformed biofilms. MNPs were discarded and the remaining biofilms were stained with CV as previously described [26].

2.5. Cytotoxicity Assay

Ninety-six well-plates were inoculated with normal human epithelial (WISH) cells (ATCC-CCL25) at 104 cells/well. On confluency, culture media were discarded and plates were inoculated with double fold serially diluted MNPs (except negative control wells) and incubated at 37°C for 24 h. Residual living cells were treated with 20 μL sterile 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) dye (5 mg/mL) at 37°C for 4 h. MTT was discarded and plates were washed with PBS three times. Aliquots of 50 μL DMSO were added to each well. Plates were left on plate shaker for 30 min to allow dissolution of the intracellular formed blue formazan complex. OD was measured at 570 nm using an ELISA plate reader [27]. Viability percentage was calculated as follows [27]:

3. Results and Discussion

3.1. Results
3.1.1. Synthesis and Characterization of MNPs

MNPs were prepared using thermal coprecipitation method [12, 28]. Results of XRD analysis shown in supplementary materials (2) are in agreement with the typical pattern for γ-Fe2O3 as initially predicted from the change of their color from black to brown [29, 30]. Transmission electron microscope imaging revealed that the size of MNP was 6–15 nm and that they were spherical in shape (Figure 2). Functionalization of the MNPs was performed to prepare hydrophobic MNPs and charge modulated ones (ranging approximately from −30 mV to +30 mV).

Six types of MNP were prepared (Figure 1). Three positive charge shifted MNPs (A-MNPs, Arg-MNPs, and APTMS-MNPs) and three negatively charged particles (MNPs, OA-MNPs, and CA-MNPs) were prepared and their surface charge was measured (Table 2). Figure 3 compares the FTIR spectrum (a), the surface charge (b), and the scanning electron micrograph of particles (c).

3.1.2. Antibacterial and Antibiofilm Potential of MNPs

(1) Antibacterial Activity

(a) Antibacterial Activity Screening. Synthesized MNPs (100 μg/mL) with different surface charge and functional groups were examined for their antibacterial activities by counting the number of bacterial colonies developed over the surface of agar plates. The highest antibacterial effect was observed for OA-MNPs and APTMS-MNPs. Hydrophobic negatively charged OA-MNPs showed greater reduction in the number of CFU/mL (61%, 54%) compared to that obtained posttreatment with hydrophilic positively charged APTMS-MNPs (43%, 35%) for S. aureus and E. coli, respectively (Figure 4).

Investigation of the APTMS-MNPs and OA-MNPs antibacterial potential using different concentrations (25–400 μg/mL) revealed a concentration dependent reduction in the CFU 24 h posttreatment (supplementary information 5). At a concentration of 400 μg/mL, the OA-MNPs showed inhibition of 83% and 79%, while the hydrophilic MNPs showed inhibition of 73% and 72% for S. aureus and E. coli, respectively (Figures 5(a) and 5(b)).

(b) Effect of MNPs on Bacterial Growth. Inhibitory activity of MNPs was determined using microdilution method by calculating MIC50 after 24 h treatment. Treating S. aureus and E. coli with OA-MNPs resulted in an MIC50 value of 31 and 63 μg/mL, respectively. On the other hand, greater concentration of APTMS-MNPs (125 μg/mL) showed 50% inhibition towards both bacterial starins (Table 3). MIC50 results indicated that OA-MNPs and APTMS-MNPs exhibited a greater inhibitory potential on S. aureus than that on E. coli (Table 3).

Analysis of growth kinetics reflects a significant time and concentration dependent decrease in OD after OA-MNPs and APTMS-MNPs treatment and revealed the growth inhibitory potentials of these particles on test bacteria (Figure 6).

(c) Surface Affinity of MNPs to Bacteria. After bacteria were allowed to interact with MNPs, MNPs were collected using an external magnet at varying time intervals (3–240 min). The reduction in OD as compared to initial OD of treated bacterial suspension represents the bacterial cells bound to MNPs.

MNPs bound to bacteria instantly (after 3 min). The period of interaction did not affect the binding capacity. OA-MNPs showed the highest percentage reduction ranging between 15%–46% and 30%–83% for S. aureus and E. coli, respectively. Positively charged hydrophilic APTMS-MNPs recorded a lower reduction percentage between 14%–32% and 12%–23% for S. aureus and E. coli, respectively (Figure 7).

The FTIR spectral difference between OA-MNPs before and after attracting bacteria is shown (supplementary information 6). Arrows indicate the position of the characteristic peaks for E. coli and S. aureus, where (a) indicates polysaccharide (900–1200 cm−1) and (b) indicates the band attributed to primary amine (1640–1560 cm−1) [31] for E. coli. For S. aureus, the peaks denoted by (c) and (d) may represent the C=N and the C-H, respectively [32, 33].

(2) Effect on Biofilm Formation and Preformed Biofilm. Table 4 and Figure 8 display the percentage reduction in absorbance of stained biofilm by CV assay after treatment of S. aureus and E. coli with different concentrations of OA-MNPs and APTMS-MNPs.

OA-MNPs (15.6–2000 μg/mL) reduced biofilm formation by 62–94% and 48–96% for S. aureus and E. coli, respectively. APTMS-MNP showed 30–91% and 34–74% reduction of the ability of S. aureus and E. coli to form a bacterial biofilm, respectively.

OA-MNPs (15.6–2000 μg/mL) destroyed the preformed biofilm by 29–94% for S. aureus and 17–93% for E. coli. APTMS-MNPs (15.6–2000 μg/mL) were able to destroy the preformed biofilm by 19–89% and 9–77% for S. aureus and E. coli, respectively.

The inhibitory potential of OA-MNPs and APTMS-MNPs on the bacterial biofilm during its development increased by increasing the concentration of MNPs. OA-MNPs and APTMS-MNPs were able to destroy the preformed biofilm. A higher inhibitory biofilm potential was observed for OA-MNPs over APTMS-MNPs (Table 4).

3.1.3. Cytotoxicity Assay

The degree of cytotoxicity of MNPs to normal human (WISH) cells was carried out using MTT assay (Table 5). APTMS-MNPs had no cytotoxic effect up to 250 μg/mL. The viability decreased to 85.2% at a concentration of 500 μg/mL. Cell viability deceased from 100% to 52% upon treatment with OA-MNPs concentrations between 62.5 and 1000 μg/mL, respectively (Figure 9).

3.2. Discussion

There is a need to reduce the use of conventional antibiotics and find alternatives to combat bacterial infections and biofilms. The ability of bacteria to form a biofilm reduces its vulnerability to antibiotics and complicates eradication efforts. In this study, we have investigated the influence of surface functionalization on the antibacterial and antibiofilm properties of MNPs.

MNPs were synthesized by the classical thermal coprecipitation method using ammonia as the alkylating agent. The prepared MNPs were spherical and of average size of 6–16 nm. The charge of the synthesized MNPs was −18 mV [8, 31]. The XRD analysis (supplementary materials 2) showed typical pattern for maghemite (γ-Fe2O3).

Surface functionalization of MNPs with citric acid was successful to shift surface charge into a more negative potential (−31 mV) [17]. Functionalizing MNPs with OA resulted in an increase in negative potential to −29.2 mV [35, 44]. To synthesize positive charge shifted MNPs, Arg (amino acid) induced a charge shift from −18 to −6 mV [4547]. Ammonia adsorbed on MNPs’ surface was able to shift surface charge to −9.6 mV, while APTMS shifted surface charge of MNPs from −18 to 24.5 mV [23]. Hydrophobicity was induced on MNPs by surface functionalization with OA.

The functionalization of the synthesized MNPs was confirmed by FTIR scanning. The unfunctionalized MNPs showed Fe-O characteristic absorption band at 634.4 and 557.5 cm−1 [48]. Two other bands around 3422.7 cm−1 and 1634.9 cm−1 correspond to stretching and bending vibrations of surface hydroxyl groups on surface of MNPs [49]. OA-MNPs showed OA coating characteristic bands at 2923.8 and 2852.9 cm−1 which correspond to CH3 stretching. The C=O appeared as a peak at 1638 cm−1 and at 1618.3 cm−1 which supports the formation of a chelating bidentate interaction between the COO and the MNPs which resulted in the blue shift of the original C=O peak of OA which appears at 1731 cm−1. Harris et al. also reported the COO of OA to appear as a broad band between 1541 and 1639 cm−1 and concluded that the bonding pattern of the carboxylic acids on the surface of the NP was at an angle to the surface [50].

A-MNPs showed an amine peak between 3410 and 3457 cm−1. CA-MNPs showed a peak at 2958.5 cm−1 which corresponds to CH2 stretching. The C=O of citric acid appeared at 1638 cm−1 [51]. Arg-MNPs showed amine peak between 3410 and 3450 cm−1, a C=O peak at 1634.9 cm−1, and a CH2 stretching band at 2923.2 cm−1 [52]. APTMS-MNPs were characterized by the CH2 stretching which appeared at 2920.8 and 2851.3 cm−1, N-H bending appeared at 1631 cm−1, C-N bending appeared at 1384 cm−1, and absorption bands in the region 1000–1227 cm−1 can be due to Si-O-Si and Si-OH [53]. FTIR spectra of the synthesized MNPs agree with those obtained in previous reports indicating successful fictionalization of the prepared MNPs. A proposed mechanism of the synthesis for the six species of MNPs is shown in Figure 1.

MNPs showed insignificant antibacterial activity [8, 37, 39]. Two physicochemical attributes were compared, namely, the surface charge and the hydrophobicity of the MNPs. This was done in the initial antibacterial screening experiment performed by counting the colony forming units. To test the effect of hydrophobicity, the effect of OA-MNPs (negatively charged hydrophobic species) was compared to that of other negatively charged hydrophilic species (MNP and CA-MNPs). The bacterial count was reduced in case of treatment with OA-MNPs, while it did not upon treatment with other negatively charged hydrophilic MNPs. Particle hydrophobicity may, therefore, play a role in facilitating the interaction between bacterial cells and MNPs.

In case of hydrophilic particles, increasing the positive charge appears to induce the antibacterial potential of particles. Hydrophilic negatively charged particles showed insignificant antibacterial activity, while modulation of surface functional group to impart a strong positive charge (e.g., ATPMS-MNPs; 24.5 mV) enhanced the antibacterial activity [8, 54, 55]. This indicated that surface positivity imparts antibacterial potential on the particles. Both conclusions may be justified by the fact that bacterial cells have a net negative charge on their cell wall and are relatively hydrophobic [56]. It was observed that hydrophobicity played a stronger role than charge modulation in the antibacterial potential of MNPs.

Growth kinetic analysis revealed a significant time and concentration dependent growth inhibitory potentials of both OA-MNPs and APTMS-MNPs on the tested bacterial strains.

The initial interaction of MNPs and bacteria is crucial for the NPs to exert their function as it is highly affected by surface affinity. The MNPs were mixed with bacterial suspension and were subsequently removed from the suspension by an external magnet. The number of bacterial cells per mL of suspension was reduced as compared to original OD. The reduction was observed directly after removing the MNPs (within 3 min). This time is dramatically less than the duplication time of most bacteria noting that the duplication times are 15–20 min for E. coli and 27–30 min for S. aureus [57]. Thus, OD reduction is due to reduction in bacterial count due to binding of bacteria to the removed MNPs. This result supports previous reports showing strong affinity of positively charged MNPs to bacterial pathogens due to electrostatic attraction [23]. In addition, results of this study indicate that bacteria had a higher affinity to hydrophobic MNPs than hydrophilic ones. For additional confirmation, the OA-MNP was FTIR scanned before and after contacting bacteria. The difference in OA-MNP FTIR spectra represents the deposited bacteria [3133].

Combating biofilm may be done by surface coating with a material that repels the bacterial cells preventing their attachment [58] or more effectively by inclusion of an agent which prevents biofilm formation. In the latter, the biofilm formation is reduced even in flowing system in addition to preventing its attachment to surfaces. The ability of MNPs to destroy preformed biofilm and inhibit new biofilm formation can reduce the need to use antibiotics [59]. Additionally, MNPs can simply be recovered from the medium.

Various modified NPs with antibiofilm activity such as gold NPs loaded with gentamycin [60], silver NPs surface treated with Allophylus cobbe extract [61], and copper oxide NPs [62] were reported. The effect of MNPs on preformed biofilm was previously described for OA-MNPs; however, its ability to inhibit new biofilm formation was not investigated [35]. Successful inhibition of biofilm development was reported using MNPs surface treated with polyvinylpyrrolidone and a thiourea derivative but this method included the use of an antibacterial agent [36]. Although MNPs with lactobacillus fermentation extract were also shown to have antibiofilm formation ability, it was shown to induce E. coli growth [55]. On the other hand, a weak effect against preformed biofilm was observed for glycerol coated MNPs [36].

In the current study, OA-MNPs (hydrophobic) and APTMS-MNPs (hydrophilic) showed a promising effect on the bacterial biofilm especially in the initial stages of biofilm development. OA-MNPs showed a stronger antibiofilm activity than the APTMS-MNPs. Our results indicated that the biofilm formed by the Gram-negative E. coli was more resistant than that formed by the Gram-positive S. aureus.

The electrostatic properties as well as hydrophobicity of both NPs and biofilms influence how they interact, taking in consideration that the majority of bacteria have negatively charged and hydrophobic biofilm matrixes which explains the antibiofilm ability of APTMS-MNPs and OA-MNPs [6365]. The higher suitability of S. aureus biofilm may be explained by its reported less negative charge than that of E. coli which may facilitate its interaction with the negatively charged OA-MNPs [35].

APTMS-MNP and OA-MNPs were safe to WISH cells up to 250 and 125 μg/mL, respectively. At these concentrations, APTMS-MNPs were able to destroy preformed biofilm by 60% and 58% for S. aureus and E. coli, respectively. OA-MNPs destroyed preformed biofilms by 76% and 55% for S. aureus and E. coli, respectively. APTMS-MNPs and OA-MNPs were also able to reduce biofilm formation by 74% and 86% for S. aureus and 44% and 81% for E. coli, respectively. Higher concentrations resulted in more effective eradication and prevention of biofilms but showed cytotoxicity which should not be a concern if MNPs are used to treat surfaces.

4. Conclusions

In this study, MNPs were prepared and their surface properties were successfully modified by functionalization with different chemical groups. The resulting particles had different hydrophobicity and surface charge.

Our results demonstrated that surface functional groups that induce positive charge or impart hydrophobicity could potentiate the antibacterial and antibiofilm activities of MNPs, possibly by changing the interaction at the NPs bacteria interface.

Surface modified MNPs can reduce bacterial growth. They can also reduce the ability of bacteria to form biofilms and subsequently weakens bacteria and inhibit their attachment to surfaces.

In our study, we also showed that both hydrophobic and positive charged MNPs do not only reduce the biofilm formation but also can destroy the preformed biofilms and thus crack the bacterial protecting matrix.

We demonstrated the surface affinity of hydrophobic and positively charged MNPs to bacterial cells and their ability to capture bacteria from liquid system and subsequent collection and manipulation by an external magnet. The capturing efficiency of hydrophobic positively charged MNPs can be optimized as bacterial filters or antibacterial coating materials with possibility of regeneration. OA-MNPs have a superior efficiency over APTMS-MNPs because they will not retain flowing salts or electrolytes through electrostatic interactions.

Surface modified MNPs showed promising antibiofilm potential in the safe concentration range for mammalian cells. Higher concentrations showed higher antibiofilm effects but increased cytotoxicity. This should not be a concern for applications involving treating surfaces due to the possibility of recovering MNPs using a magnet. However, it should be taken into consideration when designing MNP-based antibiofilm solutions for other applications.

From this study, we conclude that modifying MNPs surface represents a promising antibacterial and antibiofilm approach. As shown in this study, the synthesis of MNPs is easy, cheap, and a relatively simple procedure which can be performed using basic laboratory equipment. Their surface functionalization is feasible and can provide a variety of particles with adjustable physicochemical properties.

In conclusion, MNP surface properties can be tailored to fit hydrophobic and hydrophilic systems and achieve the required antibacterial and antibiofilm activity.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Supplementary Materials

Supplementary material 1: Change in color of MNPs as induced by changing their status from magnetite (a) to meghemite (b).

Supplementary material 2: XRD data of the prepared MNPs.

Supplementary material 3: (a) OA-MNPs (after washing with water and before wash with ethanol) showing OA-NPs not dispersed in water instead floating on water surface. (b) After successive washing with ethanol to remove excess OA from surface of MNPs; OA-MNPs are still floating and unable to disperse.

Supplementary information 4: Evaluation of the APTMS-MNPs method was performed by comparing the Zeta potential shift imparted on the MNPs. (a) Charge on surface of MNPs by the cold synthesis method (+25 mV). (b) Surface charge of particles prepared by the hot synthesis method (-11 mV). The cold synthesis method was more efficient to coat the MNPs with a surface positive charge.

Supplementary material 5: Growth inhibitory potential of MNPs on test bacteria 24 h post incubation with variable concentrations of MNPs. Subcultured S. aureus (ATCC-6538) and E. coli (ATCC-8739) were incubated with varying concentrations of APTMS-MNPs and OA-MNPs for 24 h at 37°C on a shaker (200 rpm) for 24 h. At the end of the incubation period, samples were obtained from each flask and 10 fold serially diluted in sterile saline. One hundred µL of each dilution as well as control were spread on the surface of 3 nutrient agar plates, incubated at 37°C for 24 h and the average numbers of colony forming units (CFU/mL) were counted. (a) and (c) showed the effect of APTMS-MNPs on S. aureus and E. coli. (b) and (d) showed the effect of OA-MNPs on S. aureus and E. coli.

Supplementary information 6: FTIR spectrum of control OA-MNPs before and after contact with S. aureus and E. coli. For E. coli (a) indicates the polysaccharide (900 -1200 cm-1) and (b) indicates the band attributed to primary amine (1640 – 1560 cm-1). For S. aureus, the peaks denoted by (c) and (d) represent the C=N and the C-H, respectively.

Supplementary material 7: FTIR spectra of MNPs, Arg-MNPs, CA-MNPs and A-MNPs. (*) represent characterizing bands.

  1. Supplementary Material