Nanomedicine, which involves the use of magnetic nanoparticles such as Fe3O4, has provided novel technical solutions for cancer diagnosis and treatment. Most studies in nanomedicine have focused on the use of nanoparticles with magnetic resonance imaging and hyperthermia. However, to achieve optimum anticancer effects, it is important to understand the physicochemical properties of magnetic nanoparticles and their interactions with biological entities. In this study, we synthesized Fe3O4 particles of various sizes and conjugated them with hematoporphyrin (HP) molecules by using a simple surface-modification method. HP molecules were covalently bound to the surface of Fe3O4 particles by a wet chemical process, resulting in Fe3O4@HPs particles that were uniform in size, were nontoxic, and exhibited strong anticancer effects on human prostate cancer (PC-3) and breast cancer (MDA-MB-231) cell lines. The Fe3O4@HPs particles showed remarkable and efficient photodynamic anticancer activity, depending on their particle size. These results indicate that all size of Fe3O4@HPs particles can be useful for photodynamic anticancer therapy, although the smaller size is better than the larger size and further studies will be needed to confirm the potential for clinical anticancer treatment.

1. Introduction

Recent progress in nanomedicine has led to the development of magnetic nanoparticles that show great potential for application in early diagnosis, targeted therapy, and personalized medicine [15]. In particular, efforts have been focused on early and accurate diagnosis and effective treatment. However, the results from preclinical and clinical studies of magnetic nanoparticles indicate several serious limitations such as inability to evade the immune system, anticipate toxicity via interactions with biological entities such as proteins and cell membranes, and achieve optimal bioperformance [6, 7].

To overcome these limitations, various approaches based on the functionalized surface coating of nanoparticles have been attempted to improve clinical outcome. Saito et al. [8] prepared dextran that is, ferucarbotran-coated magnetic nanoparticles, for use as enhanced contrast reagents in magnetic resonance imaging. Kievit et al. [9] proposed a facial approach to prepare specific, functional PEI-PEG-chitosan modified magnetic nanoparticles for a nonviral nanoparticle gene carrier system. Moreover, Mahmoudi et al. [10] fabricated polyethylene glycol fumarate-coated magnetic nanoparticles for bioimaging and drug delivery. In addition to polymeric coating, inorganic coating of magnetic nanoparticles with silica or gold has attracted attention, with regard to further surface derivatization of nanoparticles and because of their influence on colloidal stability and the biological behavior of magnetic nanoparticles in biomedical applications [11, 12]. Particularly, these coated nanoparticles can be easily surface-functionalized for bioconjugation to antibodies and for targeted delivery with particle localization in a specific area [1316]. Although various studies have reported a more efficient surface modification of magnetic nanoparticles, little work has been undertaken in this regard to systematically compare the size-dependent photodynamic anticancer efficiency among photosensitizer-coated magnetic nanoparticles. Among various photosensitizers, HP, a major biomolecule in erythrocytes, is enriched in tumor regions and can kill cancer cells by generating reactive oxygen species [17]. Therefore, HP-coated magnetic particles may be quite promising for applications in versatile imaging diagnosis and photodynamic therapy (PDT).

In this study, we primarily focused on comparing the photodynamic anticancer efficiencies of Fe3O4@HPs with various particle sizes. For this purpose, biocompatible Fe3O4@HPs of well-defined sizes were fabricated by a simple surface-modification process that functionalizes Fe3O4 particles by coating them with HP to allow the killing of cancer cells. The photodynamic anticancer activities of Fe3O4@HPs were evaluated with human prostate cancer (PC-3 cell) and breast cancer (MDA-MB-231 cell) cell lines in vitro to confirm the anticancer efficacy for clinical application.

2. Materials and Methods

All chemical reagents were of analytical grade and used as received without further purification.

2.1. Synthesis of Fe3O4 Particles of Different Sizes

Fe3O4 particles of different sizes were prepared as previously described [18]. Briefly, 0.54 g FeCl36H2O and 1.5 g NaOAc were dissolved in 20 mL ethylene glycol (EG) and diethylene glycol (DEG) mixture solvent, and the mixture was vigorously stirred for 30 min. The solution was transferred to an 80 mL, Teflon-lined autoclave, which was sealed, and the temperature was maintained at 200°C for 10 h, after which the solution was cooled to room temperature (RT) naturally. The black precipitate formed was collected by magnetic decantation, washed with deionized water and absolute alcohol several times, and then dried in a vacuum oven at 60°C for 6 h. The volume ratio of EG/DEG (/) determined the size of the Fe3O4 particles; thus, we obtained Fe3O4 particles with average sizes of 92.7, 171.9, 259.9, and 404.5 nm by using / ratios of 5/15, 10/10, 15/5, and 20/0, respectively.

2.2. Synthesis of Fe3O4@HPs Particles

To conjugate photosensitizer (PS) molecules, the surface of Fe3O4 particles of each size was treated with dielectric barrier discharge plasma for 30 min, as previously reported [19].

Fe3O4@HPs particles were prepared by a wet chemical process, as previously described [20]. The resulting Fe3O4 particles in tetrahydrofuran (THF) (20 mg/mL, 5 mL THF) were mixed with a solution of HP/THF (2.02 × 10−4 M). The mixture was agitated at RT for 24 h. After the reaction was complete, the product was washed with THF solvent several times. After the final wash, the residual THF solvent was removed and the particles were dried at 60°C for 6 h.

2.3. Characterization of Fe3O4@HPs Particles

Field emission scanning electron microscopy (FESEM) was performed using a Hitachi SU-70 scanning electron microscope equipped with an energy-dispersive X-ray spectroscope (EDS). X-ray diffraction (XRD) pattern of the product was obtained using a PANalytical Pert Pro MPD X-ray diffractometer with a Cu Kα radiation source ( nm) operated at 40 kV and 150 mA in a 2 range of 20° to 80°. The magnetic properties of the particles were investigated using a vibrating sample magnetometer (VSM; PPMS-9T, Quantum Design). Photoluminescence (PL) and photoluminescence excitation (PLE) spectra were measured using a spectrophotometer (F-4500, Hitachi).

2.4. Detection of Singlet Oxygen

We used 1,3-diphenylisobenzofuran (DPBF), a sensitive probe of reactive oxygen species (ROS), to detect singlet oxygen (1O2), via oxidative degradation of DPBF by 1O2 to produce nonabsorbent o-dibenzoylbenzene [21, 22], which resulted in a continual decrease in the absorption intensity of DPBF, as light irradiation continued. In the photochemical experiment, a 3.0075 mL aliquot of THF solution containing Fe3O4@HPs particles and DPBF (4.61 × 10−8 M) was introduced into a 1 cm quartz cell in the dark. The experiments were carried out by irradiating the samples with a Xe lamp (150 W, Abet Technologies, USA). A 480 nm, glass cutoff filter was used to filter out ultraviolet light. Photodegradation of DPBF was monitored by recording optical density (OD) of the absorption peak at 424 nm. After every 10 min of irradiation, the absorption spectrum of the samples was monitored using a UV-Vis spectrophotometer. A comparison was made between DPBF photodegradation of Fe3O4@HPs particles of different sizes to evaluate the relative capacity for 1O2 production.

2.5. Biocompatibility Assessment

To confirm the biocompatibility of Fe3O4@HPs particles, cytotoxicity tests were performed with fibroblasts (L-929 cells), prostate cancer (PC-3), and breast cancer (MDA-MB-231) cells, as previously described [19, 20, 23, 24]. Briefly, precultured cells were plated in a 24-well plate at 2.0 × 105 cells/mL for L-929, at 1.0 × 105 cells/mL for PC-3, and MDA-MB-231 cells. The cells were cultured at 37°C in an atmosphere containing 5% CO2 for 24 h and then incubated with different concentrations (0, 6.25, 12.5, 25, and 50 μg/mL) of Fe3O4@HPs particles at 37°C in an atmosphere containing 5% CO2 for another 24 h in the dark. Viability of each cell line was evaluated using cell counting kit-8 (CCK-8, Dojindo Laboratories, Kumamoto, Japan) to confirm the biocompatibility of Fe3O4@HPs particles. The relative cell viability was calculated as percentage survival in relation to untreated control cells.

2.6. Photodynamic Anticancer Activity Assessment

In vitro photodynamic anticancer activities of Fe3O4@HPs particles of different sizes were evaluated on PC-3 and MDA-MB-231 cells, as previously reported [19, 20, 23, 24]. Each cell line was incubated with different concentrations (0, 6.25, 12.5, 25, 50, and 100 μg/mL) and different sizes of Fe3O4@HPs particles at 37°C in an atmosphere containing 5% CO2 for 2 h in the dark. After replenishing the medium, the cells were irradiated by a conventional, green-light-emitting diode (LED). Twenty-four LEDs of type FD-32G-N1-1 (Shenzhen Fedy Technology Co., China) were used for building of the PDT-mode of the system developed, with total electrical power 3 W at 505 nm maximum wavelength and power density up to 7 mW/cm2 (maximum). After irradiation, the cells were incubated for another 24 h and, on the following day, cell viability was determined using a CCK-8 kit to measure photokilling activity on the cancer cells.

2.7. Morphological Detection of Apoptotic Cell Death

To evaluate apoptotic cell death, cell membrane translocation of PC-3 cells was analyzed using an EzWay Annexin V-fluorescein isothiocyanate (Annexin V-FITC) apoptosis detection kit (K29100, Komabiotech Inc., Seoul, Republic of Korea), according to the manufacturer instructions and as previously described [19]. Briefly, the PC-3 cells were incubated for 2 h after LED irradiation, rinsed with phosphate-buffered saline (PBS) and binding buffer, and stained with Annexin V-FITC reagent for 15 min at RT in the dark to stain the cell membrane. After incubation, the cells were washed with cold binding buffer and stained with propidium iodide (PI) to stain the nuclei.

To confirm nuclear fragmentation, PC-3 cells were irradiated with LED, fixed in ice-cold 70% ethanol for 15 min, and dried at RT. Cell membrane was stained with Texas Red C2-maleimide (30 ng/mL in PBS) for 2 h, and cell nuclei were counterstained with Hoechst 33342 (1 μg/mL in PBS) for 10 min.

Images of cells stained with fluorescent dyes were captured using a laser scanning confocal microscope (LSM 700) with a 20x objective lens and fluorescence optics (excitation at 488 nm for FITC, 530 nm for PI, 595 nm for Texas Red C2-maleimide, and 352 nm for Hoechst 33342; emission at 518 nm for FITC, 615 nm for Texas Red C2-maleimide, and 620 nm for PI). All images were analyzed using ZEN imaging software (ZEN 2009, Carl Zeiss MicroImaging GmbH).

2.8. Statistical Analysis

Data were obtained from three independent experiments (). Quantitative data are expressed as mean ± standard deviation (SD), and statistical comparisons were carried out using Student’s -test. Significant differences were indicated by .

3. Results and Discussion

EG is an effective reaction medium for the fabrication of fine metal or metal oxide particles owing to its high solubility and reducibility [25, 26]. When DEG is introduced together with EG in the reaction system, the size of the Fe3O4 particles can be successfully reduced by only varying the / ratio [27]. Similarly, we obtained Fe3O4 particles of different sizes by adjusting the / ratio of the synthetic solvent. FE-SEM images were captured to examine the morphology and size of the resulting spherical particles, as shown in Figures 1(a)1(d). All particles were spherical and remarkably uniform. In general, the size of as-prepared external particles can be precisely adjusted to obtain particles ranging between 93 and 405 nm in size. When the / ratios in this study were varied from 5/15 to 10/10, 15/5, and 7/0, the diameters of the resulting Fe3O4 particles were 92.7, 171.9, 259.9, and 404.5 nm, respectively. These results indicate that the mean diameter of the prepared Fe3O4 particles is proportional to the / ratios used during the solvothermal reaction. As shown in Figure 1 inset, the FE-SEM image at high magnification shows that the individual spheres were composed of irregular nanograins, measuring 10 to 25 nm in size. Figures 1(e)1(h) show size histograms of the Fe3O4 particles with various sizes, which were estimated by sampling 300 particles in different regions of the SEM image. These data show the mean particle size and size distribution, where the average diameters , 171.9, 259.9, 404.5 with , 16.3, 18.6, 11.8 were obtained, respectively, using a Gaussian fit.

The crystalline structures of the as-prepared Fe3O4 particles were characterized by XRD analysis, as shown in Figure 2. The position and relative intensity of all diffraction peaks matched well with the face-center cubic spinel structure of Fe3O4 (JCPDS card number: 19-0629). These results indicate that the peaks become sharper with increase in crystallite size. According to the Scherrer formula, the average crystallite sizes calculated were approximately 8.5, 11.7, 15.7, and 26.2 nm, based on the strongest peaks (311), and were much smaller than the external diameters (93 to 405 nm) of the Fe3O4 particles. These results suggest that the grain size of Fe3O4 particle shows a gradual increase with an increasing volume ratio of EG. Therefore, the volume ratio of EG/DEG plays an important role in determining the particle size and grain size.

The size-dependent magnetic properties of Fe3O4 particles were measured at RT, and the magnetization curves are shown in Figure 3. All the magnetization curves display relatively high saturation of magnetization (). The values of Fe3O4 particles with sizes of 92.7, 171.9, 259.9, and 404.5 nm were 60.1, 69.9, 79.8, and 89.5 emu/g, respectively. All the as-prepared samples possessed an value relatively smaller than those of bulk Fe3O4 particles. The magnetic remnant and coercivity showed a significant increase with increasing particle size, indicating the transformation from superparamagnetic to ferromagnetic nature. This result can be ascribed to the sequentially increased primary nanocrystal size.

The size-controlled Fe3O4 particles can be easily bonded with HP molecules after plasma treatment by means of the surface-modification process. The photoluminescence excitation (PLE) spectrum of Fe3O4@HPs showed the same characteristics as that of pure HP in THF solvent, showing an absorbance peak at 400 nm, which is the Soret band of PS, and the Q bands were located at 500, 532, and 574 nm, as shown in Figure 4. At the excitation wavelength of 500 nm, pure HP produced two strong emission peaks at 631 and 695 nm, whereas Fe3O4@HPs provided slightly red-shifted peaks at 649 and 697 nm. This result implies that the HP molecules were immobilized onto the size-controlled Fe3O4 particle surface.

The HP loading content of the hybrid Fe3O4@HPs particles depends on the weight and size of the Fe3O4 particles as a substrate. As the size of the particle decreases, the number of HP molecules bonded to the surfaces of the Fe3O4 particle increases at the same weight condition, owing to the increase in the surface area of the whole particles. Under the same weight condition (50 μg Fe3O4@HPs particles), the HP contents observed were 2.03, 1.88, 1.81, and 1.46 μg, according to the various particle sizes (from 92.7 to 404.5 nm). These results were confirmed by the analysis of UV-VIS absorption spectra of the size-controlled Fe3O4@HPs particles compared with the standard calibration. Therefore, the smallest nanoparticle complex is expected to represent the highest generation efficiency of singlet oxygen, with increase in dose.

Singlet oxygen (1O2) generation from the Fe3O4@HPs was confirmed by indirect detection of DPBF photodegradation. When the Fe3O4@HPs particles are excited by light, they emit phosphorescence as a form of deexcitation and they can achieve an intersystem cross, producing 1O2 (excited state of O2) by transferring energy to the surrounding oxygen molecules, as another form of deexcitation. DPBF, as a 1O2 probe, has a typical absorption band at 424 nm. As shown in Figure 5, the absorption intensity of DPBF in all Fe3O4@HPs solutions decreased with the light irradiation time and the particle size of Fe3O4@HPs. The degradation rate was proportional to the 1O2 yield. These results indicate that the photodegradation efficiency of DPBF was increased as the particle size of Fe3O4@HPs decreased to 404.5, 259.9, 171.9, and 92.7 nm, respectively. These results are consistent with the number of HP molecules conjugated on each particle with decrease in particle size.

To confirm the biocompatibility of each size of Fe3O4@HPs, we used fibroblasts (L-929) and two cancer cell lines (PC-3 and MDA-MB-231) to check cytotoxic effects of the nanoparticles under dark conditions, using the method recommended by the International Organization for Standardization (ISO) 10993-5 [28].

In the cytotoxicity test, cell viability with all particle sizes was > 90%, as shown in Figure 6. This result indicates that the Fe3O4@HPs are biocompatible, without exhibiting any cytotoxicity, and they can be safely used clinically as anticancer therapy.

To confirm the photodynamic anticancer activities of Fe3O4@HPs depending on the particle size, we used two cancer cell lines (prostate cancer PC-3 cells and breast cancer MDA-MB-231 cells). Photodynamic anticancer activities were evaluated by the CCK-8 method after irradiation with LED, as shown in Figure 7.

Cell viability of prostate cancer cells with 95 nm Fe3O4@HPs was 0.6% () for 100 μg/mL, 0.8% () for 50 μg/mL, and 13.2% () for 25 μg/mL. On the other hand, values for PC-3 cells with 175 nm, 280 nm, and 420 nm Fe3O4@HPs particle sizes were 2.0%, 1.8%, and 1.9% for 100 μg/mL; 8.9%, 9.2%, and 15.7% for 50 μg/mL; and 31.4%, 37.9%, and 61.4% for 25 μg/mL, respectively. These data indicate that the values for photodynamic anticancer activity for PC-3 cells were 98.1%, 99.2%, and 86.8% with 95 nm Fe3O4@HPs; 98.0%, 89.1%, and 68.4% with 175 nm Fe3O4@HPs; 98.2%, 90.8%, and 62.1% with 280 nm Fe3O4@HPs; and 90.1%, 84.3%, and 38.6% with 420 nm Fe3O4@HPs, respectively.

Additionally, the photodynamic anticancer activity values for breast cancer (MDA-MB-231) cells were 97.3%, 93.0%, and 73.0% with 95 nm Fe3O4@HPs; 93.9%, 87.9%, and 75.7% with 175 nm Fe3O4@HPs; 85.7%, 84.2%, and 78.0% with 280 nm Fe3O4@HPs; and 75.5%, 66.3%, and 43.6% with 420 nm Fe3O4@HPs, respectively.

These results are consistent with the amount of HP on the surface of each size of the Fe3O4@HPs, as described above. A large quantity of HP on the Fe3O4@HPs surface can lead to the generation of a large quantity of singlet oxygen, which plays a key role as a mediator of cell death during light irradiation, and the singlet oxygen generated from Fe3O4@HPs could determine not only the photodynamic anticancer effects but also the type of cell death [19, 29].

From these results, we confirmed that various sizes of the Fe3O4@HPs nanoparticles exerted a dose-dependent photodynamic anticancer activity in both cancer cells tested. Additionally, the Fe3O4@HPs showed slightly greater photokilling efficacy against prostate cancer cells as compared to breast cancer cells, suggesting that there is a correlation in photokilling efficacy between particle sizes and cell types.

We also observed cell membrane translocation and nuclear fragmentation of cancer cells while confirming apoptotic cell death, using an Annexin V-FITC apoptosis detection kit and a fluorescence dye, Hoechst 33342, as cell membrane translocation and nuclear fragmentation of cells are hallmarks of apoptotic cell death. These methods can be used to easily detect apoptotic cell death, as Annexin V-FITC binds to the translocated membrane phosphatidylserine, while nuclear deformation and fragmentation is visualized by staining with PI and Hoechst 33342.

Figure 8 shows cell membrane translocation and nuclear fragmentation of PC-3 cells. In Figure 8(a), the images treated with the Fe3O4@HPs show early and late-stage apoptotic cell death after irradiation by emitting green fluorescence indicating cell membrane translocation and red fluorescence marking nuclear material. However, the control cells stained by both dyes, Annexin V-FITC and PI, did not show fluorescence, demonstrating that cancer cell death by light irradiation after treatment of Fe3O4@HPs may occur by means of apoptosis for all sizes of Fe3O4@HPs nanoparticles.

Furthermore, we also confirmed nuclear fragmentation in PC-3 cells using Hoechst 33342 fluorescence dye as shown in Figure 8(b). The images of Fe3O4@HPs-treated PC-3 cells show fragmented nuclear material with a reduced and granular nuclear body (line arrows) regardless of particle size, as compared with the control.

From these results, we confirmed that the photodynamic anticancer activities of Fe3O4@HPs with various particle sizes may occur via apoptotic cell death after light irradiation and the Fe3O4@HPs with various particle sizes can be useful for PDT after confirming the efficacy of anticancer activity in vivo.

4. Conclusions

In this study, multifunctional Fe3O4 of various particle sizes were fabricated and used as a platform for surface conjugation of HP, which exhibited photodynamic anticancer activities. These multifunctional Fe3O4@HPs particles showed good chemical stability and biocompatibility, as well as effective photodynamic anticancer activity when tested on prostate cancer (PC-3) and breast cancer (MDA-MB-231) cell lines. Additionally, these multifunctional Fe3O4@HPs particles could load varying quantities of HP molecules depending on their size.

These results indicate that all size of Fe3O4@HPs particles can be useful for PDT, although the smaller size is better than the larger size and further studies will be needed to confirm the potential for clinical anticancer treatment.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

Authors’ Contribution

Ki Chang Nam, Kyong-Hoon Choi, Kyu-Dong Lee, and Jung Hyun Kim contributed equally to this work.


This study was financially supported by a grant from the National Research Foundation of Korea (NRF-2015M3A9E2066855).