# Treatment of MRSA-infected osteomyelitis using bacterial capturing, magnetically targeted composites with microwave-assisted bacterial killing

Sep 7, 2020

### Materials for experiments

All the starting materials were purchased from commercial suppliers. Pure multi-walled CNTs were purchased from Aiweixin Chemical Technology Co., Ltd. (Tianjin, China). FeCl3·6H2O and sodium acetate were purchased from Yuanye Biological Technology Co., Ltd. (Shanghai, China). The 1-tetradecanol and gentamicin used were purchased from Aladdin (Shanghai, China). Hexadecyltrimethylammonium bromide (CTAB) was purchased from Solomon Biotechnology Co., Ltd. (Tianjin, China). Sulfuric acid (H2SO4, 98%) and nitric acid (HNO3, 65–68%) were purchased from Jiangtian Chemical Technology Co., Ltd. (Tianjin, China). The rabbit Interleukin-6 ELISA Kit (CSB-E06903Rb) was purchased from Huamei Biological Engineering Co., Ltd. (Wuhan, China). All chemical solvents and salts used were of analytical grade. Starting materials were used without further purification unless otherwise noted.

### Processing of pure CNT

The functionalized CNTs powder samples were prepared in an H2SO4 and HNO3 soak. Briefly, (1) The pure CNTs were soaked in H2SO4 overnight and then stirred for 72 h; (2) the residual acid was washed off the CNTs using deionized water (dH2O) and the CNTs vacuum-dried; (3) the treated carbon tubes were dispersed in H2SO4 and stirred overnight under N2 airflow; the same volume of 68% HNO3 (diluted with dH2O) was added and stirred the mixture comprising at 65 °C for 2 h; (4) the CNTs were filtered and washed with methanol; and the CNTs were vacuum-dried to obtain the functionalized CNTs.

### Synthesis of Fe3O4 and Fe3O4/CNT

The Fe3O4/CNT was prepared as follows: FeCl3·6H2O (0.54 g) and CNTs (0.15 g) were dissolved in ethylene glycol (16 mL) and stirred for 30 min at room temperature. Polyethylene glycol (PEG 8000, 0.4 g) and CTAB (1.2 g) were added to the resulting solution, which was then stirred vigorously for 30 min. Finally, sodium acetate (1.44 g) was added and the solution continued to be stirred until it was homogeneous. It was then transferred to a 20 mL Teflon-lined stainless-steel autoclave and heated at 200 °C for 12 h. After natural cooling, the obtained product was collected by magnetic separation and washed with ethanol and dH2O several times. Finally, the obtained product was re-dispersed in acetone and isopropanol (3:1, v/v) and refluxed at 85 °C for 6 h to remove extra CTAB. Samples of Fe3O4/CNT with various doping levels were also synthesized by regulating the dosages of CNT (12.5, 35, 150, and 225 mg). The same preparation process was used to obtain Fe3O4 nanoparticles, except that no CNTs were added during the solvothermal process.

### Synthesis of the Fe3O4/CNT/Gent nanocapturer

Gent (50 mg) was dissolved in dH2O (5 mL), 1-tetradecanol (25 mg) was dissolved in ethanol (1 mL), and Fe3O4/CNT (50 mg) nanocomposites were dissolved in ethanol (4 mL). These solutions were all added to a 25 mL conical flask. After 5 min of ultrasonic dispersion, the solution was agitated in a shaker for 24 h under 50 °C. Finally, the precipitate was collected by magnetic separation, washed three times with dH2O, and vacuum-dried.

### Characterization of Fe3O4/CNT/Gent

The morphology of Fe3O4/CNT and Fe3O4/CNT/Gent was observed using SEM (S4800, Japan) and TEM (JEOL JEM-2100F, Japan). Elemental mapping images were obtained using a TEM (JEOL JEM-2100F, Japan). The surface compositions, chemical structure, and crystal structures were collected with XPS (250Xi, United States), FTIR (Nicolet IS10, United States), and XRD (D8 Advanced, Germany) utilizing CuKα radiation. The magnetic properties of the prepared samples were using vibrating sample magnetometer (SK-300, Japan).

### Microwaveocalortic effect measurements

The Fe3O4, CNT, Fe3O4/CNT, and Fe3O4/CNT/Gent in concentrations of 1 mg mL−1 (dissolved in physiological saline) were subjected to 15 min of ultrasonic dispersion and then microwaved (2.45 GHz, Schneider Medical Equipment Co., Ltd., China) for 5 min in a 2 mL Eppendorf (EP) tube. The temperature of each solution was recorded every minute using an FLIR thermal camera (FLIR E64501, Estonia). Similarly, for the microwaveocalortic on–off curve of Fe3O4/CNT/Gent (1 mg mL−1), after 20 min of irradiation, the cooling-stage temperature was also recorded every minute until it reached nearly room temperature for three cycles.

### Drug release of Fe3O4/CNT/Gent

To study the controlled release characteristics of Gent in Fe3O4/CNT/Gent the Fe3O4/CNT/Gent nanocomposites (1 mg) were dispersed in a sample tube filled with 1 mL of PBS (pH = 5.5). The Fe3O4/CNT/Gent was agitated in a shaker at 37 °C and then radiated by MW (20 min, 0.1 W cm−1) at preset periods. The concentration of Gent in the supernatant of nanocomposites was determined using a UV–vis spectrophotometer (333 nm)39. The release of Gent at different time points was calculated from the standard curve. The standard curve was calculated as Supplementary Fig. 25:

$${Y},=,0.01476,{X},-,0.00134,left( {{R}^2,=,0.999} right),$$

(2)

where, Y represents the absorbance of Gent at 333 nm and X represents the corresponding calculated Gent concentration (μg mL−1). The cumulative release was defined as mass of released Gent/mass of loaded Gent × 100%.

### MV absorption test

The electromagnetic parameters of Fe3O4, CNT, Fe3O4/CNT, and Fe3O4/CNT/Gent were prepared by uniformly mixing the absorbents with paraffin matrix according to a same mass fraction of 36% and compacting it into a columnar ring of with a 7.00-mm outer diameter and a 3.04-mm inner diameter. This was measured using a vector network analyzer (Agilent PNA-N5244A, USA). The MV absorption performances of Fe3O4, CNT, Fe3O4/CNT, and Fe3O4/CNT/Gent were evaluated by calculating the RL based on the transmission line theory40.

$$z_{mathrm{in}},=,z_0sqrt {frac{{mu _r}}{{varepsilon _r}}} tanhleft[ {jleft( {frac{{2,fpi t}}{c}} right)} right]sqrt {mu _rvarepsilon _r},$$

(3)

$${rm{RL}},=,20,{rm{log}}leftvertfrac{{z_{mathrm{in}},-,z_{mathrm{0}}}}{{z_{mathrm{in}},+,z_0}}rightvert,$$

(4)

where Zin is the input impedance at the absorber surface, Z0 is the impedance of the air, f is the MV frequency, t is the thickness of the absorber, and c is the velocity of light in free space.

The attenuation constant α was calculated using Eq. (5):

$$alpha,=,frac{{sqrt 2 pi f}}{c},times,sqrt {left( {mu ^{primeprime} varepsilon^{primeprime},-,mu ^{prime} varepsilon ^{prime} } right),+,sqrt {left( {mu ^{primeprime} varepsilon ^{primeprime},-,mu ^{prime} varepsilon ^{prime} } right)^2,+,left( {mu ^{prime} varepsilon ^{primeprime} + mu ^{primeprime} varepsilon ^{prime} } right)^2} }.$$

(5)

### Bacteria-capturing activity

Overnight bacterial cultures were adjusted to OD600 = 1.0–1.1 using physiological saline. Eight hundred μL of overnight bacterial cultures (OD600 = 1.1) were mixed with 200 μL of physiological saline (negative control), Fe3O4 solution (5 mg mL−1), CNT solution (5 mg mL−1), Fe3O4/CNT solution (5 mg mL−1), Fe3O4/CNT/Gent solution (5 mg mL−1), and Gent solution (5 × drug loading dose), respectively. Each culture was added to a 2 mL EP tube and incubated in a shaker (150 r.p.m. with a rotational radius of 10 cm) for 20 min at room temperature. The resulting solutions were placed in optical cuvettes, which were then placed on top of or alongside a magnet (neodymium rare earth permanent magnet with grade N38 magnetic energy, around 30 mm × 20 mm × 10 mm in length, width, and height, respectively) for 4 min to collect the magnetic nanoparticles. The bacteria-capturing activity was determined by the OD600 of the supernatants. Similarly, bacteria cells (OD600 = 1.1) were prepared and added to different concentrations of the Fe3O4/CNT/Gent solution (the final concentrations were 0, 0.25, 0.5, and 1 mg mL−1, respectively). After being shaken for 0, 10, 20, 60, 120, and 180 min at room temperature the samples were placed on a magnet to collect the Fe3O4/CNT/Gent-captured bacteria and the supernatant was collected for OD value testing with a microplate reader (SpectraMax I3MD, USA). The same method was used to determine the MRSA-capturing effect of different CNT-doped Fe3O4/CNT hybrids: 800 μL of bacterial culture (OD600 = 1.0) was mixed with 200 μL of physiological saline (shaken for 20 min; after magnet-induced separation) as the Ctrl group; 800 μL of bacterial culture (OD600 = 1.0) was mixed with 200 μL of the hybrid (1 mg mL−1, shaken for 20 min, after magnet-induced separation) as the experimental group. In order to directly observe the different bacteria-capturing capacity of Fe3O4/CNT/Gent (1 mg mL−1, shaken for 20 min) to both MRSA (CCTCC 16465) and E. coli (ATCC 8099), 20 μL of the suspensions were removed before and after magnetic induced separation, appropriately diluted in a liquid culture medium, plated on the Luria–Bertani (LB) agar plates, and cultured at 37 °C for 20 h.

### Antibacterial test

The in vitro antibacterial activity of Fe3O4, CNT, Gent, Fe3O4/CNT, and Fe3O4/CNT/Gent against MRSA and E. coli was quantitatively using the spread-plate method. After being cultured for ~24 h, the MRSA and E. coli (≈109 CFU mL−1) were diluted to ≈107 CFU mL−1 in LB medium for subsequent experiments. The bacterial suspensions were then incubated with physiological saline (control), Fe3O4 (1 mg mL−1), CNT (1 mg mL−1), Gent (7.31 μg mL−1, drug loading dose), Fe3O4/CNT (1 mg mL−1), and Fe3O4/CNT/Gent (1 mg mL−1) in 2 mL EP tubes with and without MV irradiation at a power density of 0.1 W cm−2 for 20 min. Specifically, the temperature of irradiated Fe3O4/CNT and Fe3O4/CNT/Gent (1 mg mL−1) was more than 50 °C after 5 min and <55 °C after 15 min. After different treatments, the diluted bacterial solution (20 μL) was smeared on LB agar plate evenly and cultured at 37 °C. The colonies were counted after cultured 20 h to calculate the antibacterial ratio, which was then assessed using Eq. (6):

$${mathrm{Antibacterial}},{mathrm{ratio}},{mathrm{(% )}},=,frac{{A,-,B}}{A}{mathrm{100% }},$$

(6)

where A and B represent the numbers of bacteria in the control group and experimental group, respectively.

The morphologies of the bacteria (both MRSA and E. coli) interacting with Fe3O4, CNT, Gent, Fe3O4/CNT, and Fe3O4/CNT/Gent were evaluated using SEM. The treated bacteria were soaked with glutaraldehyde (2.5%) solution for 4 h and washed with PBS (pH = 7.0). The samples were then dehydrated in different concentrations (30, 50, 70, 90, and 100%, v/v) of ethanol for 15 min and air-dried before observation.

### In vitro cytotoxicity evaluation

The MC3T3-E1 (ATCC CRL-2593) and NIH-3T3 (ATCC CRL-1658) were cultured in Dulbecco’s modified eagle medium and minimum essential medium alpha (α-MEM) respectively, including supplemented with 10% (v/v) fetal bovine serum, 1% amphotericin and 1% penicillin–streptomycin and incubated at 37 °C in 95% humidity and an atmosphere containing 5% CO2.

For the fluorescence morphology, MC3T3-E1 cells were first incubated in six-well plates at 37 °C for 24 h and further incubated with Fe3O4, CNT, Gent, Fe3O4/CNT, and Fe3O4/CNT/Gent (concentrations are consistent with those used in antibacterial experiments). The treated cells were cultured for another 24 h. After incubation, the samples were washed with PBS, soaked in formaldehyde (4%) for 10 min, and rinsed with sterile PBS. These samples were then stained with fluorescein isothiocyanate-phalloidin for 30 min (avoid light during this process), washed with PBS, stained again with 4,6-Diamidino-2-phenylindole for 20 s, rinsed with PBS three times, and then photographed using laser scanning confocal microscopy (Nikon A1R+, Japan).

MTT assay of MC3T3-E1. The various samples (Fe3O4, CNT, Gent, Fe3O4/CNT, and Fe3O4/CNT/Gent) and various concentrations of Fe3O4/CNT/Gent were co-cultured with MC3T3-E1 cells (105 cells/well) for 3 days (n = 5 independent samples). The control group was physiological saline. After incubation certain periods (3 days), the medium was removed and 3-(4,5-dimethylthiazol-2-yl)−2,5-diphenyltetrazolium bromide (MTT, 0.5 mg mL−1) was added, this resulting solution incubation for 4 h at 37 °C. The solution was then removed and 200 μL dimethyl sulfoxide was added (shook for 15 min) and centrifuged. Finally, the absorption of the supernatant at 570 nm was determined. The cell viability (%) was calculated by comparing the absorbance values of these samples with the control.

Evaluation hemolysis of Fe3O4/CNT/Gent nanocapturer. First, 5 mL of New Zealand rabbit blood was diluted with 50 mL of PBS solution and centrifuged in the mixture solution at 176 × g for 6 min to obtain red blood cells (RBCs). The RBCs were then washed three times with PBS and resuspended in 20 mL of the PBS solution. Different concentrations of Fe3O4/CNT/Gent nanocapturer were incubated with diluted RBCs for 4 h at 37 °C (n =3 independent samples). After centrifugation for 6 min at 176 × g, the absorbance of the supernatant was measured at 405 nm. A positive control (100% lysis) was prepared by treating RBCs with 1% Triton X-100. The hemolysis percentage was calculated by comparing the absorbance values of these samples with the positive control.

### In vivo safety

In brief, male New Zealand rabbits (6–8 weeks old, ~2.0–2.5 kg in weight) were first shaved to remove the hair on the hind legs. Subsequently, 200 μL of 5 mg mL−1 of Fe3O4/CNT/Gent nanocapturer was injected in situ, using rabbits that had not been subjected to surgery as the control. At different time points post-injection, the complete blood was collected for blood routine tests (WBC, RBC, HGB, MCH, PLT, MPV, MCV, Gran, HCT, Mon, and RDW; 1 day, 7 days, and 14 days post-injection) using an automatic animal blood cell analyzer (BC-2800vet, China). The concentrations of Fe ions (Fe2+ and Fe3+) were then measured using an ICP-MS (Agilent 7800, USA). The internal organs (heart, liver, spleen, lung, and kidney) were collected 14 days post-injection for histological analysis by H&E staining.

### Rabbit osteomyelitis model and treatment

The study was carried out in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The ethical aspects of the animal experiment were approved by the Animal Ethical and Welfare Committee (AEWC) of the Institute of Radiation Medicine, Chinese Academy of Medical Sciences (Approval No. IRM-DWLL-2019087). The male New Zealand white rabbits (6–8 weeks old, ~2.0–2.5 kg in weight) were randomly divided into four groups (n = 4 per group) at every preset time point: the control group (physiological saline), the MV treatment group, the Fe3O4/CNT/Gent + MV treatment group, and the dual-targeting (Fe3O4/CNT/Gent + MV + MF) group.

The rabbits were under a 25 ± 2 °C and 60–70% (humidity) conditions for 3 days. They were anaesthetized with an intramuscular injection of pentobarbital (30 mg kg−1) prior to surgery. After anesthesia, the hind legs of the rabbits were shaved and disinfected. A hole was drilled in the tibial plateau of the hind legs using medical electric drill 2 mm in diameter; via this hole, the legs were injected with 106 CFU (50 μL) of MRSA bacterial suspension in situ to establish the osteomyelitis model. Physiological saline or Fe3O4/CNT/Gent was added for different treatment options. Finally, the holes were sealed with bone wax and the muscles and skin were sutured carefully. The rabbits were subjected to different treatments: The control group was injected with physiological saline (200 μL). The MV group was injected with physiological saline (200 μL), and then irradiated MV for 20 min (2.45 GHz, 0.1 W cm−2, keeping the temperature between 50 and 55 °C once a temperature of 55 °C was reached); and the Fe3O4/CNT/Gent + MV group was injected with 200 μL of Fe3O4/CNT/Gent (5 mg mL−1) and irradiated with MV for 20 min (2.45 GHz, 0.1 W cm−2, keeping the temperature between 50 and 55 °C once a temperature of 55 °C was reached). The dual-targeting group was treated simply by adding a strong magnet (neodymium rare earth permanent magnet with grade N38 magnetic energy, around 20 mm × 10 mm × 10 mm in length, width, and height, respectively) to the Fe3O4/CNT/Gent + MV group experiment.

To evaluate the therapeutic effect of different treatments for MRSA-infected osteomyelitis, the infected femur and tibia specimens of the rabbit models were observed, photographed, and was investigated using 3.0 T MR scanner (GE, MR750, USA) after 14 days of treatment. At a preset time, the rabbits were sacrificed, their blood was collected for blood tests (IL-6 and WBC tests), their bone marrow was collected for Wright’s staining (Days 14) and H&E staining (Days 2, 7, and 14).

After 14 days of treatment, the heart, liver, spleen, lung, and kidney were stained with H&E. The bone marrow and the rest of the spleen and kidney tissues were weighed separately and quantitative of LB medium was added for every 1 g of tissue and homogenization. Homogenates (20 μL), properly diluted, were smeared on LB agar plate evenly. The number of bacteria were counted after incubation 20 h at 37 °C and was expressed as 1 g CFU g−1 for tissues.

### Statistics and reproducibility

All the quantitative data in each experiment were evaluated and analysed by one-way or two-way analysis of variance and expressed as the mean values ± standard deviations from at least three independent experiments, followed by Tukey’s multiple comparisons post hoc test to evaluate the statistical significance of the variance. The n.s. present P > 0.05 and ****P < 0.0001 were considered statistically significant.

### Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.