Multifunctional Fe3O4 Nanoparticles Filled Polydopamine Hollow Rods for Antibacterial Biofilm Treatment

This work reports the use of mesoporous silica rods as templates for the step-wise preparation of multifunctional Fe3O4 NPs filled polydopamine hollow rods (Fe3O4@PDA HR). The capacity of as-synthesized Fe3O4@PDA HR as a new drug carrier platform was assessed by its loading and the triggered release of fosfomycin under various stimulations. It was found that the release of fosfomycin was pH dependent with ~89% of fosfomycin being released in pH 5 after 24 h, which was 2-fold higher than that in pH 7. The magnetic properties of Fe3O4 NPs and the photothermal properties of PDA enabled the triggered release of fosfomycin upon the exposure to rotational magnetic field, or NIR laser irradiation. Additionally, the capability of using multifunctional Fe3O4@PDA HR to eliminate preformed bacterial biofilm was demonstrated. Upon exposure to the rotational magnetic field, the biomass of a preformed biofilm was significantly reduced by 65.3% after a 20 min treatment with Fe3O4@PDA HR. Again, due to the excellent photothermal properties of PDA, a dramatic biomass decline (72.5%) was achieved after 10 min of laser exposure. This study offers an alternative approach of using drug carrier platform as a physical mean to kill pathogenic bacteria along with its traditional use for drug delivery.


Introduction
The utilization of mesoporous silica materials for biomedical applications, especially for controlled drug delivery purposes, has been well-documented in a number of research studies [1][2][3]. The properties that make this kind of material popular include the porous structure with a high surface area, good biocompatibility, the capability of tuning particle size and forming desired structures under specific synthesis conditions. Besides their common use as drug carrier platforms, those materials are often employed as sacrificial templates for the fabrication of polymeric hollow structures with high uniformity and a well-defined shape; however, it was commonly followed by the etching of the silica core via harsh chemical treatments [4,5].
The incorporation of Fe 3 O 4 nanoparticles (NPs) into drug carrier systems to form a magnetic guidance system has gained substantial interest due to its low toxicity and biocompatibility [6,7]. For instance, Lu et al. [8] reported their work on the synthesis of triple porous microspheres, in which porous Fe 3 O 4 particles served as magnetic cores being engulfed by dual porous SiO 2 shells for the controlled delivery of 5-fluorouracil. Javanbakht et al. [9] reported the use of magnetic Fe 3 O 4 NPs as a core structure engulfed by double-shell layers of SiO 2 and tannic acid to co-deliver doxorubicin and methotrexate for anticancer treatment. In the majority of Fe 3 O 4 based drug delivery systems, it is common that Fe 3 O 4 NPs served as the internal magnetic core which is engulfed by an outer layer of silica for drug binding. The utilization of magnetic activated metal nanomaterials for antimicrobial purposes has drawn attention from multiple researchers over the last few years [16]. For instance, Elbourne et al., reported the use of a magnetic field to attenuate and transform galliumbased liquid metal droplets to develop sharp-edged structures for the disruption of bacterial biofilms [17], or the utilization of magnetic nanoparticles under a magnetic field to create artificial channels for the improvement of antibiotic penetration within the biofilm structure was reported by Quan et al. [18].
Taking these into consideration, several questions have arisen regarding the movement of embedded Fe3O4 NPs under rotational magnetic fields and whether they could be used as a physical means to break down the PDA shell and induce damage to preformed biofilms, and whether the drug carrier itself could be used as a means to eliminate preformed bacterial biofilm in vitro. To answer those questions, further experiments are designed to evaluate the possibility of using Fe3O4@PDA HR to disrupt preformed bacterial biofilm under the exposure to a rotational magnetic field without using any therapeutic molecules as a proof-of-concept. Besides that, the photothermal conversion of Fe3O4@PDA HR inherited from a PDA component was also employed to evaluate its efficiency in terms of killing preformed bacterial biofilm under the exposure to NIR laser irradiation.

Synthesis of Monodisperse Fe3O4 NPs and Its Grafting onto MSR
The fabrication of MSR was made use of a binary surfactant system with the aid of Pluronic F127 and CTAB, and TEOS served as the silica precursor. According to the SEM image as presented in Figure 2A, the obtained silica particles exhibit uniform rod structures with the average size of 393 nm in length and 121 nm in width. The BET analysis The capability of using as-synthesized Fe 3 O 4 @PDA HR for drug loading and the controlled release of the payload under various stimuli, such as pH, NIR laser irradiation, and rotational magnetic field is then examined using fosfomycin as a model drug. Fosfomycin is a broad-spectrum antibiotic derived from cultures of Streptomyces spp. which has been used in clinical treatment to combat antibiotic resistant bacteria and their biofilm formation [13][14][15]. Additionally, the efficiency of using Fe 3 O 4 @PDA HR to deliver fosfomycin for antimicrobial purposes in vitro is further evaluated.
The utilization of magnetic activated metal nanomaterials for antimicrobial purposes has drawn attention from multiple researchers over the last few years [16]. For instance, Elbourne et al., reported the use of a magnetic field to attenuate and transform galliumbased liquid metal droplets to develop sharp-edged structures for the disruption of bacterial biofilms [17], or the utilization of magnetic nanoparticles under a magnetic field to create artificial channels for the improvement of antibiotic penetration within the biofilm structure was reported by Quan et al. [18].
Taking these into consideration, several questions have arisen regarding the movement of embedded Fe 3 O 4 NPs under rotational magnetic fields and whether they could be used as a physical means to break down the PDA shell and induce damage to preformed biofilms, and whether the drug carrier itself could be used as a means to eliminate preformed bacterial biofilm in vitro. To answer those questions, further experiments are designed to evaluate the possibility of using Fe 3 O 4 @PDA HR to disrupt preformed bacterial biofilm under the exposure to a rotational magnetic field without using any therapeutic molecules as a proof-of-concept. Besides that, the photothermal conversion of Fe 3 O 4 @PDA HR inherited from a PDA component was also employed to evaluate its efficiency in terms of killing preformed bacterial biofilm under the exposure to NIR laser irradiation.

Synthesis of Monodisperse Fe 3 O 4 NPs and Its Grafting onto MSR
The fabrication of MSR was made use of a binary surfactant system with the aid of Pluronic F127 and CTAB, and TEOS served as the silica precursor. According to the SEM image as presented in Figure 2A, the obtained silica particles exhibit uniform rod structures with the average size of 393 nm in length and 121 nm in width. The BET analysis revealed MSR particles have a general pore diameter of 2.6 nm, and a high surface area of approximate 1600 m 2 /g. revealed MSR particles have a general pore diameter of 2.6 nm, and a high surface area of approximate 1600 m 2 /g.
In this work, the monodisperse Fe3O4 NPs were prepared by thermal decomposition of the iron precursor iron oleate in 1-octadecene at 320 °C. This was followed by surface functionalization with APTES to enhance its water dispersity. The morphology of monodisperse Fe3O4 was characterized using TEM ( Figure 2B). As could be seen, the obtained Fe3O4 NPs have a high degree of uniformity in terms of a spherical shaped structure with an approximate diameter of 15.6 nm. The Fe3O4 filled PDA HR was prepared through several steps including grafting Fe3O4 NPs onto carboxylic functionalized MSR via an amidation reaction, followed by the PDA coating and silica core removal. To begin with, the carboxylic functionalization of MSR was achieved through the amidation between amine functionalized MSR and the carboxyl group of succinic acid anhydride. The successful grafting of amine and subsequent carboxylic functionalized MSR was indicated by an observed zeta potential change. The zeta potential of MSR had a negative value of −33.18 mV, which was attributed by the abundance of the OH group on the silica surface. Upon amine grafting, the zeta potential shifts significantly to a positive value of 26.48 mV, which indicates the successful immobilization of the amine group on the surface of MSR. The zeta potential shifts back to a negative value of −34.25 mV after the amidation of the amine functional group and In this work, the monodisperse Fe 3 O 4 NPs were prepared by thermal decomposition of the iron precursor iron oleate in 1-octadecene at 320 • C. This was followed by surface functionalization with APTES to enhance its water dispersity. The morphology of monodisperse Fe 3 O 4 was characterized using TEM ( Figure 2B). As could be seen, the obtained Fe 3 O 4 NPs have a high degree of uniformity in terms of a spherical shaped structure with an approximate diameter of 15.6 nm.
The Fe 3 O 4 filled PDA HR was prepared through several steps including grafting Fe 3 O 4 NPs onto carboxylic functionalized MSR via an amidation reaction, followed by the PDA coating and silica core removal. To begin with, the carboxylic functionalization of MSR was achieved through the amidation between amine functionalized MSR and the carboxyl group of succinic acid anhydride. The successful grafting of amine and subsequent carboxylic functionalized MSR was indicated by an observed zeta potential change. The zeta potential of MSR had a negative value of −33.18 mV, which was attributed by the abundance of the OH group on the silica surface. Upon amine grafting, the zeta potential shifts significantly to a positive value of 26.48 mV, which indicates the successful immobilization of the amine group on the surface of MSR. The zeta potential shifts back to a negative value of −34.25 mV after the amidation of the amine functional group and carboxylic acid group due to the ring-opening of succinic acid. This suggests that the amine functionalized MSR was successfully converted to carboxylic MSR.
The grafting of Fe 3 O 4 -NH 2 onto MSR-COOH was achieved via the conjugation between amine functional groups and carboxylic groups to form amide bonds. It is noteworthy that the whole conjugation process was conducted at room temperature with high efficiency, as evidenced by the very small amount of unbound MSR remaining in the solvent after magnetic separation.
The successful grafting of Fe 3 O 4 onto MSR was further confirmed by X-ray diffraction (XRD) analysis. The XRD peaks derived from monodisperse Fe 3 (Figure 3), which are in agreement with previous studies with regards to the XRD pattern of Fe 3 O 4 NPs [19,20]. Meanwhile, a broad peak ranging from 18 • to 30 • is ascribed to amorphous silica which could be observed in the samples containing silica rods [21,22]. Upon the incorporation of Fe 3 O 4 NPs, additional peaks similar to that of bare Fe 3 O 4 -NH 2 emerged in the XRD pattern of MSR-COOH, which implied the successful grafting of magnetite onto silica rod templates.
Molecules 2023, 28, x FOR PEER REVIEW 4 of 16 carboxylic acid group due to the ring-opening of succinic acid. This suggests that the amine functionalized MSR was successfully converted to carboxylic MSR. The grafting of Fe3O4-NH2 onto MSR-COOH was achieved via the conjugation between amine functional groups and carboxylic groups to form amide bonds. It is noteworthy that the whole conjugation process was conducted at room temperature with high efficiency, as evidenced by the very small amount of unbound MSR remaining in the solvent after magnetic separation.
The successful grafting of Fe3O4 onto MSR was further confirmed by X-ray diffraction (XRD) analysis. The XRD peaks derived from monodisperse Fe3O4 NPs featured very prominently at 2ϕ values of 30.4°, 35.6°, 43.5°, 57.4°, and 63.1° (Figure 3), which are in agreement with previous studies with regards to the XRD pattern of Fe3O4 NPs [19,20]. Meanwhile, a broad peak ranging from 18° to 30° is ascribed to amorphous silica which could be observed in the samples containing silica rods [21,22]. Upon the incorporation of Fe3O4 NPs, additional peaks similar to that of bare Fe3O4-NH2 emerged in the XRD pattern of MSR-COOH, which implied the successful grafting of magnetite onto silica rod templates. The Fe3O4 grafted MSR was then subjected to PDA coating and silica removal according to the PDA mediated silica dissolution process developed by our previous work [23]. The resulting structure is shown in Figure 4A. As can be seen, PDA hollow rods (HR) were obtained, which collapsed due to the loss of the silica core and drying for SEM imaging. In addition, the successful encapsulation of Fe3O4 within PDA shell was confirmed by TEM. As could be seen in Figure 4B, Fe3O4 NPs were found to remain intact within the interior of PDA HR after the removal of MSR, and this was further confirmed by EDX analysis ( Figure 4C). The Fe 3 O 4 grafted MSR was then subjected to PDA coating and silica removal according to the PDA mediated silica dissolution process developed by our previous work [23]. The resulting structure is shown in Figure 4A. As can be seen, PDA hollow rods (HR) were obtained, which collapsed due to the loss of the silica core and drying for SEM imaging. In addition, the successful encapsulation of Fe 3 O 4 within PDA shell was confirmed by TEM. As could be seen in Figure 4B, Fe 3 O 4 NPs were found to remain intact within the interior of PDA HR after the removal of MSR, and this was further confirmed by EDX analysis ( Figure 4C).

Loading and Controlled Release of Fosfomycin from Fe3O4@PDA HR
Fosfomycin was chosen as a model drug to be loaded onto Fe3O4@PDA HR. With the aid of LC-MS/MS, the amount of drug being loaded onto Fe3O4@PDA HR was calculated to be 5.7 ± 0.8 μg per mg of Fe3O4@PDA HR. Upon the successful loading of fosfomycin,

Loading and Controlled Release of Fosfomycin from Fe 3 O 4 @PDA HR
Fosfomycin was chosen as a model drug to be loaded onto Fe 3 O 4 @PDA HR. With the aid of LC-MS/MS, the amount of drug being loaded onto Fe 3 O 4 @PDA HR was calculated to be 5.7 ± 0.8 µg per mg of Fe 3 O 4 @PDA HR. Upon the successful loading of fosfomycin, the capability of triggered release of the payload from Fe 3 O 4 @PDA HR was further examined by employing a number of simulations, including pH, NIR laser irradiation and exposure to a rotational magnetic field.
As presented in Figure 5A, the release of fosfomycin from the drug carrier was pH dependent. For both pH 5 and 7, there was an initial burst release of fosfomycin, with approximately 53.4 ± 0.3% (pH 5) and 24.4 ± 2.7% (pH 7) of fosfomycin being detected after 30 min. After that, the releasing trend of fosfomycin slowed down and gradually increased over time with nearly 89.2 ± 3.8% and 46.4 ± 3.6% of the loaded drug being released after 24 h at pH 5 and pH 7, respectively. It was noted that the amount of fosfomycin being released at pH 5 was nearly 2-fold higher than that at pH 7 at any time interval being tested. The phenomena could be rationalized by the instability of the PDA structure under acidic conditions [24,25] which might change the permeability of the PDA coating, and pH 7 was chosen as it is neutral and quite close to body pH.

Fosfomycin Loaded Fe3O4@PDA HR for Biofilm Inhibition
In order to evaluate the efficacy of using Fe3O4@PDA HR for drug delivery purposes, the minimal inhibitory concentration (MIC) of fosfomycin and fosfomycin loaded Fe3O4@PDA HR against S. aureus was examined by a microbroth dilution assay. For fosfomycin, when its concentration reaches 8 μg/mL, the growth of S. aureus was not able to be observed after 24 h incubation. Whereas the MIC of fosfomycin loaded Fe3O4@PDA HR against S. aureus was determined at 1.25 mg/mL, which corresponds to the same amount of fosfomycin that was loaded onto Fe3O4@PDA HR.
Further experiments were then conducted to evaluate the inhibition effect of fosfomycin and fosfomycin-loaded Fe3O4@PDA HR on biofilm formation. As to the results, fosfomycin exhibits an excellent antibiofilm effect regarding to S. aureus biofilm formation. Specifically, at the final concentration 0.25 × and 0.5 × MIC, nearly 62.9 ± 1.0% and 72.4 ± 6.3% of biofilm formation was inhibited, respectively. The inhibition rate further increased to ~85% at 1 × MIC ( Figure 6). Similar results were observed in terms of fosfomycin-loaded Fe3O4@PDA HR with 68.7 ± 2.0%, 73.1 ± 7.1%, and 84.4 ± 6.0% at the final concentration of 0.25 ×, 0.5 ×, and 1 × MIC, respectively. The obtained results highlight that the fosfomycin-loaded Fe3O4@PDA HR are effective to inhibit biofilm formation, with an efficacy similar to that of free drug molecules, indicating fosfomycin retained its bio functionality after being loaded on PDA HR. Due to the incorporated magnetic Fe 3 O 4 NPs, it was hypothesized that the movement of Fe 3 O 4 @PDA HR under rotational magnetic force might facilitate the release of loaded drugs. To test this hypothesis, the release of fosfomycin-loaded Fe 3 O 4 @PDA HR was carried out with and without exposure to a rotational magnetic field. Results indicated that the release of fosfomycin from Fe 3 O 4 @PDA HR being placed under the rotational magnetic field was higher than its counterpart. The released amount of fosfomycin under the rotational magnetic field was found to be 41.3 ± 1.6%, 56.4 ± 3.0% and 64.0 ± 3.2% after 3, 5 and 24 h, respectively, while these figures were 37.0 ± 2.9%, 40.2 ± 3.9% and 46.4 ± 3.6% without the assistance of the magnetic field ( Figure 5A). We believe that the movement of Fe 3 O 4 @PDA HR upon the exposure to rotational magnetic field facilitates the detachment and release of fosfomycin from PDA HR.
In addition to the above, the ability of using NIR laser to trigger the release of fosfomycin from Fe 3 O 4 @PDA HR was examined. Under the same experimental conditions at pH 7, the release of fosfomycin under laser treatment was found to be more efficient compared to that of its counterpart. For example, after 1 h of testing, approximately 38.1 ± 1.3% of fosfomycin was detected upon laser treatment, whereas this figure in the absence of the laser treatment was just around 31.6 ± 1.8% ( Figure 5B).

Fosfomycin Loaded Fe 3 O 4 @PDA HR for Biofilm Inhibition
In order to evaluate the efficacy of using Fe 3 O 4 @PDA HR for drug delivery purposes, the minimal inhibitory concentration (MIC) of fosfomycin and fosfomycin loaded Fe 3 O 4 @PDA HR against S. aureus was examined by a microbroth dilution assay. For fosfomycin, when its concentration reaches 8 µg/mL, the growth of S. aureus was not able to be observed after 24 h incubation. Whereas the MIC of fosfomycin loaded Fe 3 O 4 @PDA HR against S. aureus was determined at 1.25 mg/mL, which corresponds to the same amount of fosfomycin that was loaded onto Fe 3 O 4 @PDA HR.
Further experiments were then conducted to evaluate the inhibition effect of fosfomycin and fosfomycin-loaded Fe 3 O 4 @PDA HR on biofilm formation. As to the results, fosfomycin exhibits an excellent antibiofilm effect regarding to S. aureus biofilm formation. Specifically, at the final concentration 0.25 × and 0.5 × MIC, nearly 62.9 ± 1.0% and 72.4 ± 6.3% of biofilm formation was inhibited, respectively. The inhibition rate further increased to~85% at 1 × MIC ( Figure 6). Similar results were observed in terms of fosfomycin-loaded Fe 3 O 4 @PDA HR with 68.7 ± 2.0%, 73.1 ± 7.1%, and 84.4 ± 6.0% at the final concentration of 0.25 ×, 0.5 ×, and 1 × MIC, respectively. The obtained results highlight that the fosfomycin-loaded Fe 3 O 4 @PDA HR are effective to inhibit biofilm formation, with an efficacy similar to that of free drug molecules, indicating fosfomycin retained its bio functionality after being loaded on PDA HR.

Fe3O4@PDA HR as Physico-Mechanical Means against Bacterial Preformed Biofilm
Preliminary experiments were carried out to prove whether Fe3O4 NPs could disrupt preformed bacterial biofilm under a rotational magnetic field. A preformed S. aureus biofilm was prepared and incubated with bare Fe3O4 NPs. After removal of un-attached Fe3O4 NPs, the treated biofilm was exposed to a rotational field at the speed of 2000 rpm. Strikingly, the biomass of preformed biofilm significantly declined to 52.0 ± 8.7%, 46.6 ± 6.4%, and 34.7 ± 7.2% after 10-, 15-and 20-min exposure to the magnetic rotational field, respectively ( Figure 7A).
The disruption of biofilm caused by the movement of Fe3O4 NPs under the magnetic field was also examined using SEM imaging. As seen in Figure 7(Bii), the biofilm treated with bare Fe3O4 NPs was significantly reduced under the rotational magnetic field. The breakage of the cell membrane and the leakage of the internal cell content caused by Fe3O4 NPs was observed (as indicated with white arrows). This shows that Fe3O4 NPs induce damage to pathogenic bacteria either from within the interior in terms of biofilm's uptake of Fe3O4 NPs or the exterior in case they escape from dead cells and disrupt adjacent cells under an applied magnetic field.
The use of Fe3O4@PDA HR to induce damage to S. aureus biofilm was then examined in the same way. After 10 min of exposure to the rotational magnetic field, the S.aureus

Fe 3 O 4 @PDA HR as Physico-Mechanical Means against Bacterial Preformed Biofilm
Preliminary experiments were carried out to prove whether Fe 3 O 4 NPs could disrupt preformed bacterial biofilm under a rotational magnetic field. A preformed S. aureus biofilm was prepared and incubated with bare Fe 3 O 4 NPs. After removal of un-attached Fe 3 O 4 NPs, the treated biofilm was exposed to a rotational field at the speed of 2000 rpm. Strikingly, the biomass of preformed biofilm significantly declined to 52.0 ± 8.7%, 46.6 ± 6.4%, and 34.7 ± 7.2% after 10-, 15-and 20-min exposure to the magnetic rotational field, respectively ( Figure 7A). To further confirm the above observation, the live/dead cell staining method was employed. As seen in Figure 8, in the case of bare Fe3O4, the fluorescent signals from biofilms post treatment are lower than those of the control experiment as well as the biofilm treated with Fe3O4@PDA HR. This might imply that the rotational force of the bare Fe3O4 under magnetic field might cause the detachment of the biofilm from the utilized substrate along with its wall disruption property to eliminate preformed biofilm. Furthermore, a significant amount of cell death was observed in either the biofilm treated with bare Fe3O4 or Fe3O4@PDA HR after exposure to a magnetic field, which further confirms the feasibility of utilizing magnetite at the nanoscale to control the growth of bacteria as well as their biofilm. The use of Fe 3 O 4 @PDA HR to induce damage to S. aureus biofilm was then examined in the same way. After 10 min of exposure to the rotational magnetic field, the S. aureus biomass biofilm declined slightly by approximately 10%. However, the biomass was reduced dramatically to 53.0 ± 7.2% and 40.0 ± 8.4% after prolonged exposure times of 15 and 20 min, respectively ( Figure 7A). This slow start phenomenon could suggest that the engulfment of the PDA wall might hinder the movement of Fe 3 O 4 NP; hence, in order to break and escape from PDA HR carriers, longer exposure times to the rotational magnetic field might be required. The results indicate that Fe 3 O 4 @PDA HR has a similar effect to bare Fe 3 O 4 NPs on disrupting biofilms in terms of inducing damage to cell walls, which could also be observed under SEM as indicated by white arrows in Figure 7(Biii).
To further confirm the above observation, the live/dead cell staining method was employed. As seen in Figure 8, in the case of bare Fe 3 O 4 , the fluorescent signals from biofilms post treatment are lower than those of the control experiment as well as the biofilm treated with Fe 3 O 4 @PDA HR. This might imply that the rotational force of the bare Fe 3 O 4 under magnetic field might cause the detachment of the biofilm from the utilized substrate along with its wall disruption property to eliminate preformed biofilm. Furthermore, a significant amount of cell death was observed in either the biofilm treated with bare Fe 3 O 4 or Fe 3 O 4 @PDA HR after exposure to a magnetic field, which further confirms the feasibility of utilizing magnetite at the nanoscale to control the growth of bacteria as well as their biofilm.

Fe3O4@PDA HR against Pre-Formed Biofilm under NIR Laser Irradiation
So far, PDA has been well-known for its excellent photothermal conversion, whic has been utilized in various studies [26][27][28]. Taking this into consideration, further expe iments exploiting the photothermal conversion effect of Fe3O4@PDA HR against pr formed S. aureus biofilm was conducted.
To examine the photothermal conversion efficacy, a similar amount of materials w dispersed in water, and exposed to NIR laser irradiation at the wavelength of 795 nm. Th water temperature containing either bare Fe3O4 or Fe3O4@PDA HR escalates upon the e posure to NIR laser irradiation at the wavelength of 795 nm. Specifically, after 20 min laser irradiation, the temperature of water containing Fe3O4@PDA HR rises significant by 8.4 °C, whereas the figure of bare Fe3O4 NPs was just around 4.9 °C ( Figure 9A). Th difference in temperature increase indicated that the presence of PDA enhances the he conversion capacity of Fe3O4 NPs. However, the heat conversion of Fe3O4@PDA HR seem to be less effective compared to the bare hollow PDA HR (8.4 vs. 9.6 °C). This could b attributed to the presence of Fe3O4 NPs within the PDA network, which might reduce th synergistically acoustic vibration between PDA walls. In addition, the photostability Fe3O4@PDA HR was quite stable even after being exposed to the laser source multip times ( Figure 9B).

Fe 3 O 4 @PDA HR against Pre-Formed Biofilm under NIR Laser Irradiation
So far, PDA has been well-known for its excellent photothermal conversion, which has been utilized in various studies [26][27][28]. Taking this into consideration, further experiments exploiting the photothermal conversion effect of Fe 3 O 4 @PDA HR against preformed S. aureus biofilm was conducted.
To examine the photothermal conversion efficacy, a similar amount of materials was dispersed in water, and exposed to NIR laser irradiation at the wavelength of 795 nm. The water temperature containing either bare Fe 3 O 4 or Fe 3 O 4 @PDA HR escalates upon the exposure to NIR laser irradiation at the wavelength of 795 nm. Specifically, after 20 min of laser irradiation, the temperature of water containing Fe 3 O 4 @PDA HR rises significantly by 8.4 • C, whereas the figure of bare Fe 3 O 4 NPs was just around 4.9 • C ( Figure 9A). The difference in temperature increase indicated that the presence of PDA enhances the heat conversion capacity of Fe 3 O 4 NPs. However, the heat conversion of Fe 3 O 4 @PDA HR seems to be less effective compared to the bare hollow PDA HR (8.4 vs. 9.6 • C). This could be attributed to the presence of Fe 3 O 4 NPs within the PDA network, which might reduce the synergistically acoustic vibration between PDA walls. In addition, the photostability of Fe 3 O 4 @PDA HR was quite stable even after being exposed to the laser source multiple times ( Figure 9B). the area that received NIR treatment within biofilm could be observed ( Figure 9D). The heat conversion capability of Fe3O4@PDA HR was triggered via laser treatment, and induced hyperthermia to bacterial cells, resulting in cell death as indicated by red signals, which represent the staining of nuclei of dead cells by propidium iodide [29,30], whereas virtually no effect on the growth of bacterial cells in the untreated area was observed, as indicated by strong green fluorescent signals, that had arisen from the metabolism of the SYTO 9 dye by the living cells [31]. Compared to previous studies on the use of Fe3O4 for biofilm eradication [18,32], the utilization of PDA HR to encapsulate and deliver monodisperse Fe3O4 NPs has several advantages as it could simultaneously co-deliver therapeutic drugs in a controllable manner. In addition, PDA based material could facilitate the localization of Fe3O4 NPs and optimize their quantity being uptaken by a particular cell of interest through the binding of catechol and amine groups distributed on PDA surface to the dopamine receptor family, D2DR as previously suggested by Liu's group [33]. Even though the use of Fe3O4 filled PDA HR for biofilm elimination in this study was just carried out on gram positive As presented in Figure 9C, the biomass of preformed biofilms after the uptake of either bare PDA HR or Fe 3 O 4 @PDA HR particles was significantly diminished upon NIR laser treatment. In detail, the biomass of the biofilm treated with bare PDA HR and Fe 3 O 4 @PDA HR reduced to nearly 30.0 ± 8% and 32.6 ± 9%, respectively, from an initial 100% biomass of controlled biofilm after 5 min of NIR laser irradiation, and a further reduction in biomass to around 25.0 ± 6.6% and 27.5 ± 8.3%, respectively, could be observed within the next 5 min of laser treatment. Even though bare PDA HR has a better photothermal conversion compared to that of Fe 3 O 4 @PDA HR, these two materials have no significant performance difference for the elimination of preformed S. aureus biofilms. The elimination of preformed S. aureus biofilms using Fe 3 O 4 @PDA HR via NIR laser irradiation was further assessed by CLSM. A significant contrast between the untreated area and the area that received NIR treatment within biofilm could be observed ( Figure 9D). The heat conversion capability of Fe 3 O 4 @PDA HR was triggered via laser treatment, and induced hyperthermia to bacterial cells, resulting in cell death as indicated by red signals, which represent the staining of nuclei of dead cells by propidium iodide [29,30], whereas virtually no effect on the growth of bacterial cells in the untreated area was observed, as indicated by strong green fluorescent signals, that had arisen from the metabolism of the SYTO 9 dye by the living cells [31].
Compared to previous studies on the use of Fe 3 O 4 for biofilm eradication [18,32], the utilization of PDA HR to encapsulate and deliver monodisperse Fe 3 O 4 NPs has several advantages as it could simultaneously co-deliver therapeutic drugs in a controllable manner. In addition, PDA based material could facilitate the localization of Fe 3 O 4 NPs and optimize their quantity being uptaken by a particular cell of interest through the binding of catechol and amine groups distributed on PDA surface to the dopamine receptor family, D2DR as previously suggested by Liu's group [33]. Even though the use of Fe 3 O 4 filled PDA HR for biofilm elimination in this study was just carried out on gram positive bacteria as a proof of concept, the obtained results were quite promising. Hence, it is note-worthy for further study to focus on its potential application for different biofilms formed by different pathogenic bacteria or even cancerous cell lines in vitro and in vivo.

Synthesis of Mesoporous Silica Rods (MSR) and Its Carboxylic Functionalization
The fabrication of MSR followed a procedure which was previously employed by Yang et al. [34]. Briefly, an amount of 0.615 g Pluronic F127 and 0.8925 g of CTAB were dissolved in a mixture of 80 mL H 2 O and 75 mL NH 4 OH (2.5 wt%) under stirring. Upon dissolution, 3 mL of TEOS were added to the as-prepared mixture. The reaction mixture was vigorously mixed for 2 min, followed by aging for 3 h under static conditions. The MSR was collected and washed several times prior to calcination at 550 • C for 2 h.
Grafting of the carboxylic acid group onto the surface of the silica was performed following a procedure, which was presented by An et al. [35] with some modifications. In brief, 200 mg of bare MSR was refluxed in 10 mL of toluene and 1 mL of APTES under stirring at 70 • C for 18 h to obtain amine functionalized MSR. After that, the amine functionalized MSR was dissolved in 20 mL of dimethylformamide (DMF) with the aid of sonication, followed by the addition of 20 mL 0.1 M succinic acid in DMF. The reaction was refluxed at 70 • C overnight under stirring. The carboxylic MSR particles were then collected via centrifugation and washed with DMF and ethanol, respectively.

Synthesis of Amine Functionalized Monodisperse Fe 3 O 4 NPs
The synthesis of monodisperse Fe 3 O 4 NPs was achieved through thermal decomposition of an iron precursor, iron (III) oleate. At first, the iron (III) oleate complex was synthesized according to a published procedure described by Park et al. [36] with some modifications. In detail, 9.133 g of sodium oleate (30 mmol) and 2.7 g of FeCl 3 ·H 2 O (10 mmol) were dissolved completely in a solution containing 13.5 mL H 2 O, 20 mL EtOH and 32.8 mL hexane. The solution was then heated to 70 • C and blanketed with an argon atmosphere for 3 h. Upon the completion, the obtained waxy iron oleate that formed was washed three times with distilled water in a separatory funnel. The remaining hexane was volatilized naturally in the dish, and the final iron oleate complex was dried at 60 • C under vacuum.
After that, the synthesis of monodisperse Fe 3 O 4 NPs followed a procedure which was previously described by Zhao et al. [37] with some minor modifications. In brief, 1.6 g iron oleate, 220 µL oleic acid, and 120 µL distilled H 2 O were mixed with 20 mL of 1-octadecene. The resulting solution was degassed and heated up to 320 • C for 2 h under inert atmosphere. After a predetermined time of 30 min, the solution was cooled down to room temperature and a volume of 60 mL i-PrOH was added to precipitate the nanoparticles. The nanoparticles were separated by centrifugation and washed 3 times with ethanol. To

Loading Fosfomycin to Fe 3 O 4 @PDA HR
Typically, 100 mg of as-prepared Fe 3 O 4 @PDA HR was dispersed in 20 mL of fosfomycin solution (1 mg/mL) in sterilized H 2 O and left stirring overnight. After that, the drug loaded PDA HR were magnetically separated and washed several times with sterile H 2 O to remove unbound fosfomycin molecules. A liquid chromatography setup fitted with a mass spectrometer (LC-MS) was used to quantify the loading of fosfomycin. The experimental detection of fosfomycin via LC-MS was conducted using the direct injection mode with a mobile phase containing 20 mM ammonium acetate and methanol at a volumetric ratio of 50:50. The flow rate was fixed at 5 µL/min for all samples.

NIR Laser Irradiation Mediated Drug Release Profile
In a typical experiment, an aliquot of 3 mL of 20 mM ammonium acetate at pH 7 containing fosfomycin loaded Fe 3 O 4 @PDA HR (1 mg/mL) was employed. After 15-min intervals the cuvette was irradiated with a NIR laser light at a wavelength of 795 nm with a power density of 0.13 W/cm 2 for 15 min. Before and after laser treatment, 1 mL of release media was collected and subjected to the LC-MS measurement. The release of fosfomycin from Fe 3 O 4 @PDA HR under the same conditions without laser treatment was used as a control.

Release of Fosfomycin under Exposure to Rotational Magnetic Field
An aliquot of 1.0 mL of fosfomycin loaded Fe 3 O 4 @PDA HR (10 mg/mL) was dispersed into 20 mL of 20 mM ammonium acetate at pH 7 and exposed to a rotational magnetic field at a rotational speed of 2000 rpm. After a predetermined time, 1 mL of release media was collected for LC-MS measurement along with the addition of 1 mL of fresh buffer to the release media. The release of fosfomycin from Fe 3 O 4 @PDA HR under the same conditions without exposure to the magnetic field was used as a control.

Determination of Minimal Inhibitory Concentration of Free Fosfomycin and Fosfomycin Loaded Fe 3 O 4 @PDA HR against S. aureus
An overnight cultured suspension of Staphylococcus aureus bacteria was diffused in fresh media of Mueller-Hinton broth (MHB) and the absorbance was adjusted to be within the range of 0.08-0.13 at a wavelength of 600 nm. Then, 100 µL of adjusted bacterial suspen-sion was diluted in tubes containing 10 mL of fresh MHB to obtain bacterial suspensions of 10 6 CFU/mL. A serial dilution of fosfomycin ranging from 512 µg/mL to 1 µg/mL was set-up in a 96-well microtiter plate (Corning, NY, USA) with the initial concentration being at 1 mg/mL, followed by the addition of 75 µL of diluted bacterial suspension into each well. For the growth control, 100 µL of diluted bacterial suspension was employed and wells containing only fresh medium broth were used as a negative control. In case of fosfomycin loaded Fe 3 O 4 @PDA HR, a serial dilution of a stock of fosfomycin loaded Fe 3 O 4 @PDA HR particles ranging from 5 mg/mL to 0.15 mg/mL was prepared by using the initial concentration of fosfomycin loaded Fe 3 O 4 @PDA HR of 10 mg/mL. After 24 h incubation, the absorbance of each well after incubation was measured at the wavelength of 595 nm using a microplate reader (POLARstar, Omega, BMG LABTECH, Ortenberg, Germany) to determine the MIC value of free fosfomycin and fosfomycin loaded Fe 3 O 4 @PDA HR against S. aureus.
Evaluating Anti-Biofilm Formation Activity of Fosfomycin Loaded Fe 3 O 4 @PDA HR via Metabolic Assay An aliquot of 100 µL of fosfomycin loaded PDA suspension was pipetted to final concentrations of 0.25, 0.5 or 1 × MIC on a microtiter plate, followed by the addition of 100 µL of bacterial suspension at a concentration of 10 6 CFU per mL, which was supplemented with glucose 1% beforehand. The microtiter plates were incubated at 37 • C for 24 h to allow the biofilm formation. After that, the culture medium was carefully removed, and the bacterial biofilm was left on the plate and dried out at 60 • C for 30 min. Subsequently, 100 µL of phosphate buffer saline (PBS) was added to each well, followed by the addition of 5 µL of prepared 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide tetrazolium reduction assay (MTT) solution. Then, the plate was incubated at 37 • C for another 3 h before measuring ODs at the wavelength of 570 nm. The percentage of inhibition of peptide was calculated by: In brief, S. aureus biofilm was firstly created by adding 100 µL of bacterial suspension at a concentration of 10 6 CFU per mL supplemented with glucose 1% on microtiter plate with 24 h incubation at 37 • C. The culture medium was later carefully removed, and bacterial biofilm was left on the plate. At this stage, 100 µL of Fe 3 O 4 @ PDA HR suspension (1 mg/mL) was incubated with the preformed biofilm for 1 h for the biofilm uptake of nanoparticles. After that, the supernatant was removed, and the biofilms were washed with PBS to remove the excess nanoparticles. As-prepared samples were primarily exposed to a rotating neodymium magnet for 10-20 min, and the rotational speed of the magnetic field was maintained at 2000 rpm prior to the addition of 100 µL of fresh PBS and MTT reagent. A preformed biofilm of S. aureus treated with Fe 3 O 4 @ PDA HR suspension without undergoing further treatment was employed as a control. The biomass of remaining preformed biofilm after treatment was calculated by: Evaluating the Effect of Fe 3 O 4 @ PDA HR against Pre-Formed Biofilm upon Laser Irradiation The detailed procedure was as above, but instead of exposing to rotational magnetic field, a laser source at the wavelength of 795 nm with the power intensity at 0.5 W/cm 2 was applied to the treated biofilm for 5 or 10 min prior the addition of 100 µL of fresh PBS and MTT reagent.

Live/Dead Cell Staining for Antibiofilm Effect
Biofilms of S. aureus were first grown on a glass-bottom petri dish (FluoroDisk). The viability of biofilms after being exposed to different stimuli was then evaluated via the LIVE/DEAD™ BacLight™ Bacterial Viability Kit (Invitrogen TM , ThermoFisher, Melbourne, Australia). Specifically, 200 µL of stain solution containing 3 µL of SYTO ® 9 stain and 3 µL of propidium iodide stain predispersed in 1 mL of sterilized water was added to the treated biofilms, followed by a 30-min incubation in the dark. Excess staining on the biofilm was then removed and rinsed with water. Then, the live/dead stained biofilm was observed under confocal laser scanning microscopy (Olympus Fluoview 1000 IX81, Olympus Corporation, Melbourne, Australia).

Material Characterization
Scanning Electron Microscope (SEM) images and energy dispersive X-ray (EDX) spectroscopy were obtained on a field emission scanning electron microscope (FeSEM, ZEISS SUPRA 40VP, Carl Zeiss NTS GmbH, Oberkochen, Germany) at an acceleration voltage of 3 kV. Transmission electron microscopy (TEM) was performed using a JEM 2100F (JEOL Ltd., Akishima, Japan) at room temperature at an accelerating voltage of 200 kV. The crystal information of the samples was collected via an X-ray diffractometer (Bruker, D8-Advance, Karlsruhe, Germany) with Cu Ka radiation. The quantification of fosfomycin was carried out using the Agilent 1290 Infinity II UHPLC apparatus coupled with a 6545 Q-TOF MS (Agilent Technologies, Santa Clara, CA, USA). The optical characterizations were performed using a UV-Vis spectrophotometer (Halo RB-10, Dynamica Pty Ltd., Victoria, Australia). Confocal Laser Scanning Microscopy (CLSM) images were performed using Olympus Fluoview 1000 IX81 confocal laser scanning microscope, operated using 100× oil-immersion objective combined with 5× optical zoom.

Conclusions
The successful fabrication of Fe 3 O 4 filled PDA HR was confirmed by various characterization techniques including SEM, EDX, XRD and TEM. The capability of using obtained Fe 3 O 4 @PDA HR for fosfomycin loading and controlled release under various stimulations was fully examined. Of note, the incorporation of Fe 3 O 4 NPs within PDA HR not only offers magnetic guidance property for the drug carrier but also facilitates the release of the payload upon exposure to the rotational magnetic field. More importantly, for the first time we demonstrated that the encapsulation of monodisperse Fe 3 O 4 NPs within PDA HR could serve as a mechanical means to disrupt the structure of preformed biofilms upon exposure to the rotational magnetic field.
Undoubtedly, Fe 3 O 4 @PDA HR can be utilized efficiently either as a carrier platform for chemotherapeutic agents or as a mechanical means to disrupt the structure of preformed bacterial biofilms. This study offers an alternative way of making use of the drug carrier platform as a weaponized material rather than its conventional purpose for drug delivery to fight against bacterial biofilm in the era of antibiotic-resistant bacteria.

Conflicts of Interest:
The authors declare no conflict of interest.