Green Synthesis of Fe₃O₄ Nanoparticles Stabilized by a Garcinia mangostana Fruit Peel Extract for Hyperthermia and Anticancer Activities

Abstract

Introduction

Fe₃O₄ nanoparticles (Fe₃O₄ NPs) with multiple functionalities are intriguing candidates for various biomedical applications.

Materials and Methods

This study introduced a simple and green synthesis of Fe₃O₄ NPs using a low-cost stabilizer of plant waste extract rich in polyphenols content with a well-known antioxidant property as well as anticancer ability to eliminate colon cancer cells. Herein, Fe₃O₄ NPs were fabricated via a facile co-precipitation method using the crude extract of Garcinia mangostana fruit peel as a green stabilizer at different weight percentages (1, 2, 5, and 10 wt.%). The samples were analyzed for magnetic hyperthermia and then in vitro cytotoxicity assay was performed.

Results

The XRD planes of the samples were corresponding to the standard magnetite Fe₃O₄ with high crystallinity. From TEM analysis, the green synthesized NPs were spherical with an average size of 13.42±1.58 nm and displayed diffraction rings of the Fe₃O₄ phase, which was in good agreement with the obtained XRD results. FESEM images showed that the extract covered the surface of the Fe₃O₄ NPs well. The magnetization values for the magnetite samples were ranging from 49.80 emu/g to 69.42 emu/g. FTIR analysis verified the functional groups of the extract compounds and their interactions with the NPs. Based on DLS results, the hydrodynamic sizes of the Fe₃O₄ nanofluids were below 177 nm. Furthermore, the nanofluids indicated the zeta potential values up to −34.92±1.26 mV and remained stable during four weeks of storage, showing that the extract favorably improved the colloidal stability of the Fe₃O₄ NPs. In the hyperthermia experiment, the magnetic nanofluids showed the acceptable specific absorption rate (SAR) values and thermosensitive performances under exposure of various alternating magnetic fields. From results of in vitro cytotoxicity assay, the killing effects of the synthesized samples against HCT116 colon cancer cells were mostly higher compared to those against CCD112 colon normal cells. Remarkably, the Fe₃O₄ NPs containing 10 wt.% of the extract showed a lower IC₅₀ value (99.80 µg/mL) in HCT116 colon cancer cell line than in CCD112 colon normal cell line (140.80 µg/mL).

Discussion

This research, therefore, introduced a new stabilizer of Garcinia mangostana fruit peel extract for the biosynthesis of Fe₃O₄ NPs with desirable physiochemical properties for potential magnetic hyperthermia and colon cancer treatment.

Keywords:

• green synthesis
• Fe₃O₄ nanoparticles
• Garcinia mangostana
• magnetic hyperthermia
• cytotoxicity assay

Introduction

Fe₃O₄ nanoparticles (Fe₃O₄ NPs) can play an important role in cancer therapy. Fabrication of Fe₃O₄ NPs can be carried out by different procedures such as co-precipitation, thermal decomposition, and green synthesis.^1–3 Co-precipitation is a conventional method withthe capability to produce large-scale of Fe₃O₄ NPs.^4–6 This method still requires improvement particularly in controlling the particle size and composition of the NPs. In synthesis of Fe₃O₄ NPs, different types of natural stabilizers such as plant extracts and bioactive molecules can increase the colloidal stability and physiochemical characteristic of Fe₃O₄ NPs.¹,^7–9 It has been reported that coating the NPs could be costly, laborious, time-consuming, and toxic to the environment when chemicals are used.¹⁰ To overcome these issues, preparation of NPs by a green approach has drawn considerable attention.¹¹ During the green synthesis process, natural-based stabilizers may hydrolyze iron salt solution for ferric hydroxide formation, which is then reduced by biomolecules to form NPs.¹² As a green stabilizer, plant extracts and its polyphenolic compounds could be presented into/onto NPs to decrease particle interactions as well as improve thermodynamic stability.¹,¹³ Therefore, Fe₃O₄ NPs stabilized with the plant extracts could potentially show better water permeability, antioxidant activity, biocompatibility, biodegradability, and tolerable toxicity for various biomedical applications compared to using other traditional methods.^14–16

Compared to the normal amount consumed with a low phenolic content, the plant extract might induce a greater amount of a metabolite compound of polyphenol or flavonoid subclass and distribute at the targeted site for stronger antioxidant and anticancer activities.¹⁷ Polyphenols, especially anthocyanins, undergo a major structural changes during the in vitro or in vivo evaluations, improving both bioavailability and biological properties.¹⁸ It is worth mentioning that the peel of some fruits has higher antioxidant activities than the pulp.¹⁹ The extract of fruit peels such as Garcinia mangostana (G. mangostana)¹⁹ and mango²⁰ are rich sources of antioxidants. Even though there are many studies related to the green synthesis of Fe₃O₄ NPs using different plant extracts such as Kappaphycus alvarezii²¹ and Punica granatum peel;⁶ however, applications of the green synthesized Fe₃O₄ NPs with anticancer properties are still less focused.¹

Over the years, using agro-waste has been intriguing for researchers to fabricate environmental-friendly materials to tackle problems of toxicity and the reduction in landfill space.²² G. mangostana is known as mangosteen and considered as a tropical fruit from Guttiferea family. It contains 17% outer pericarp, 48% inner pericarp, 31% juicy flesh, and 4% cap.²³ The pulp extract of G. mangostana has been famously used for a food supplement and herbal medicine.¹⁹ Peel of G. mangostana is considered as a waste material, albeit, its crude extract may contain benzophenones, flavonoids, and anthocyanins compounds with antioxidant properties.²⁴ For example, xanthones in flavonoid compounds with a high composition of α-mangostin and γ-mangostin have been reported for anticancer treatments.²⁴ Xanthones have been used to treat different types of cancer cells.²⁵ Despite that there have been some studies on G. mangostana fruit peels in synthesis of Au NPs²⁶ and Ag NPs,²⁷ there have been no precise studies on the ability of G. mangostana to synthesize Fe₃O₄ NPs.

Magnetic hyperthermia therapy (MHT) is currently considered as an attractive application of Fe₃O₄ NPs in cancer treatment, however, it is still under the clinical trial.²⁸ For this purpose, the ferrofluid samples are exposed to an external alternating magnetic field (AMF) in order to assess their heating capabilities known as specific absorption rate (SAR). In MHT, the AMF strengths can control the hyperthermia temperature (T_H) and properties of Fe₃O₄ NPs.²⁹ It has been stated that a secure T_H is ranged from 42°C to 47°C to cause hydrolyzing the tissue’s proteins and acute necrotic cancer cell death over 30–60 min, whereas the normal cells can be unaffected due to heat transfer phenomena.²⁸,³⁰ It is worth mentioning that Fe₃O₄ NPs in a low-viscous medium may trigger a magnetic response of anisotropic particles to be physically rotated by Brownian heating loss mechanism for increments of T_H and SAR values.³¹ This may be achieved by developing the green synthesis of Fe₃O₄ NPs stabilized with a plant extract.

Based on World Health Organization (WHO) report, there have been approximately 18 million cancer cases and nearly 9.6 million people have died of cancer in the year 2018 alone.³² In nanomedicine, NPs smaller than 10 nm are likely to be removed through renal clearance, whereas NPs bigger than 100 nm can be trapped rapidly in spleen and liver via macrophage phagocytosis.³³ Fe₃O₄ NPs with an appropriate size and surface chemistry can be applied for various therapeutic applications, including drug delivery, magnetic resonance imaging (MRI) contrast agent and MHT.¹,^34–37 There have been studies using co-precipitation method to synthesize Fe₃O₄ NPs for colon cancer treatment.⁶,¹¹,¹⁶,²¹ For instance, dextran-coated Fe₃O₄ NPs were synthesized by using a facile co-precipitation method with a mean size of ~10 nm.³⁸

To our best knowledge, this current study is probably the first report to use the crude extract of G. mangostana fruit peel as a novel and green stabilizer in the biosynthesis of Fe₃O₄ nanofluids with high quality and low-cost for potential hyperthermia and colon cancer treatments. The Fe₃O₄ NPs samples were stabilized with various weight percentages of the peel extract and they were evaluated by X-ray powder diffraction (XRD), transmission electron microscopy (TEM), field emission electron microscope (FESEM), energy-dispersive X-ray spectroscopy (EDS), vibrating sample magnetometer (VSM), and Fourier-transform infrared spectroscopy (FTIR). Dynamic light scattering (DLS) was also used to examine the stability of the samples solutions over four weeks of storage. The green synthesized Fe₃O₄ NPs were assessed by magnetic hyperthermia analysis. In addition, the in vitro cytotoxicity assays of the samples were evaluated using the human colon cancer cell line (HCT116) and colon normal cell line (CCD112).

Materials and Methods

Materials

G. mangostana fruit was obtained from Terengganu, Malaysia. Iron (II) chloride tetrahydrate (FeCl₂.4H₂O ≥ 99%) and iron (III) chloride hexahydrate (FeCl₃. 6H₂O, 97%) were purchased from Sigma-Aldrich. Sodium hydroxide (NaOH) was purchased from R&M Chemicals. All aqueous solutions were prepared using double distilled water. The chemicals were used without further purification. All glassware used was washed with distilled water and dried before used.

Preparation of Garcinia mangostana Fruit Peel Extract

The extract of G. mangostana fruit peel was obtained using a modified method from our group.⁶,²⁶ Briefly, the fruit peels were washed several times to remove dust, followed by drying at an ambient condition. A total of 10 g of the fruit peel was ground and mixed with 100 mL double deionized water at 80°C, using an oil bath under a constant stirring at 250 rpm for 1 h. The crude extract solution (10 m/V) was filtered with filter paper (Fioroni 601) and oven-dried at 45°C for 24 h. The dried extract powder was termed as S5 and stored at 4°C for further processing.

Green Synthesis of Fe₃O₄ NPs Stabilized by the Extract of Garcinia mangostana Fruit Peel

A simple co-precipitation method and the extract of G. mangostana fruit peel as a novel stabilizing and capping agent were used to synthesize Fe₃O₄ NPs. The crude extract of G. mangostana fruit peel, iron salts, and sodium hydroxide acted as a stabilizer, iron sources, and a reducing agent, respectively.⁶ Four different extract solutions (wt.%) were prepared. In total, 1, 2, 5, and 10 g of the dried extract powder were added into four different beakers containing respectively 99, 98, 95, and 90 g of double deionized water and stirred for 15 min at room temperature. FeCl₃. 6H₂O, 97% (2.53 g) and FeCl₂.4H₂O ≥ 99% (0.99 g) at a molar ratio of 2:1 were respectively added into each solution. After that, NaOH (1 M) was dropwise added into the respective solutions to adjust the pH to ~11 followed by vigorous stirring for another 30 min. Finally, the samples were centrifuged three times at 14,000 rpm for 12 min and the collected precipitates were oven-dried at 70°C. The same procedure without using the extract was carried out for the preparation of bare Fe₃O₄ NPs. The synthesized Fe₃O₄ NPs with 0, 1, 2, 5, and 10 wt.% concentrations of the peel extract were termed as S0, S1, S2, S3, and S4, respectively, and the peel extract powder alone was termed as S5.

Characterization

The structural characteristics of the synthesized samples (~3 g) were analyzed by using a PANalytical X’Pert PRO XRD with Cu Kα radiation (λ= 0.15406nm). An applied current of 20 mA and an accelerating voltage of 45 kV in a 2θ (from 5° to 80°) with a scanning rate of 2θ/min were used. TEM (JEM-2100F-Japan) determined morphology, particle size, and structure of the sample as the solution of the sample (~1.5 mg in 10 mL distilled water) was dropped onto 300-mesh copper grids and also a TEM image was analyzed under the microscope. FESEM was used to determine the size distribution and morphology of the samples (~2 mg) by using a JSM-7800F Prime Schottky, then the FESEM attached to EDS analysis for chemical characterization of the samples. An accelerating voltage of the microscope was set at 5.0 kV and the standard magnification was selected at 40 kX. Moreover, VSM (Model 7400, Tokyo, Japan) was used to evaluate the magnetic properties of the samples (~2 g) at room temperature. The functional groups of the samples were identified by using IR Tracer-100 FTIR (Thermo Nicolet, USA) within the wavelength range of 400–4000 cm⁻¹. Each sample (~5 mg) was mixed with powder of potassium bromide (KBr) (500 mg) at a ratio of 1:100 w/w to produce a pellet.

Dynamic Light Scattering (DLS) Analysis and Impact of Storage Time on Fe₃O₄ Nanofluids

For DLS analysis, Anton Paar instrument was used to potentially measure stability of the synthesized samples in the distilled water solution. In order to evaluate the effect of storage time on the Fe₃O₄ NPs dispersions, each sample solution (80 mL; 100 μg/mL) was distributed into four glass bottles (15 mL each), covered with aluminum foil and stored at 4°C.^39–41 Zeta potential, hydrodynamic particle size, and polydispersity index were measured immediately post-synthesis (as prepared) and subsequently after 1, 2, 3, and 4 weeks of storage. Independent experiments were repeated at least three times, and the data were expressed as mean ± standard deviation for all triplicates within an individual experiment.

Magnetic Hyperthermia Studies

To investigate the magnetic hyperthermia efficiency for the colloidal magnetic NPs (S0–S4), 1 mg of the sample powder was dispersed in 1 mL distilled water (1 mg/mL) in accordance with the human ‎body structure, then was sonicated for 2 min at a frequency of 1.15 MHz with acoustic energy and power of 500 J and 25 W, respectively. The hyperthermia conditions were provided by using an induction heating ‎device (Easyheat 3542LI, 4 kW, Ambrell, Gloucestershire, UK)⁴² with three varied currents of 75, 100, and 125 A, which respectively corresponded to frequencies of 318, 313, and 312 kHz. To reduce the temperature variation with the environmental condition, the Fe₃O₄ NPs dispersions were put inside a 2 mL glass vial surrounded by a polystyrene box.²⁹,⁴³ The dispersed S0–S4 samples in a vial vertically were placed in the center of a short helical coil with 8-turned loops and 2.54, 2.92 and 5.10 cm in ‎inner diameter, outer diameter, and ‎length, respectively, and were cooled by water circulation to ‎ensure constant temperature and impedance. Temperature changes over time (dT/dt) during the hyperthermia experiments were measured with a thermocouple connected to a data processing system to obtain T_H.

The generated magnetic field strengths (H) at different currents are ‎determined using Equation 1(1) :⁴²,⁴⁴,⁴⁵‎

(1)

Where n, i, and L represent the number of the coil turns, applied current (A), and inner coil diameter (m), respectively. The calculated values of AMF strength at ‎75, 100, and 125 A,‎ were 23.6, 31.5, and 39.37 kA.m⁻¹, respectively. Heating efficiency of the Fe₃O₄ NPs ‎samples are quantitatively estimated through ‎ SAR‎ under the external AMF exposure for 3600 sec using Equation 2(2) :⁴⁵

(2)

Where C_m is the specific heat capacity of the colloidal dispersion, including Fe₃O₄ and water medium with values of 0.65 J.g⁻¹.K⁻¹ and 4.18 J.g⁻¹.K⁻¹, respectively. The m_m was defined as the total mass of the colloidal dispersion ‎comprised the medium and the Fe₃O₄ NPs sample ‎‎(iron oxide and the extract as a stabilizer) at a ‎concentration of 1 mg/mL, whereas C_m m_m = (C_medium × ‎m_medium) + (C_sample× m_sample). Further, m_Fe shows the iron mass per unit of the Fe₃O₄ NPs ‎sample (based on the Fe ratio),⁴⁴ and the dT/dt signifies ‎the initial slope of the temperature profile.

Cell Lines and Culture Reagents

HCT116 colon cancer (ATCC CCL-247) and CCD112 colon normal (ATCC CRL-1541) cell lines were purchased from American Type Culture Collection (ATCC) and cultured according to ATCC’s recommendation.⁶,⁴⁶ All cell lines were cultured in high-glucose Dulbecco’s Modified Eagle’s medium (DMEM) (#12800, Thermo Fisher Scientific) supplemented with 10% foetal bovine serum (FBS) (#10270-106, Thermo Fisher Scientific), and 1% penicillin/streptomycin (#15140-122, Thermo Fisher Scientific).

In vitro Cytotoxicity Assay

Cytotoxicity assays were conducted to verify the cellular killing effect of all the samples (S0–S5) by using CellTiter-Glo 2.0 Luminescent Cell Viability Assay (#G9241, Promega), according to the manufacturer’s instruction with a slight modification.⁶,⁴⁶ Briefly, 5000 HCT116 and CCD112 cells per well (100 µL/well) were seeded onto a 96-well plate and incubated overnight (12–16 h) at 37°C in a 5% CO₂ and 95% humidified incubator. After that, 2-fold serially diluted samples at concentrations of 0, 15.62, 31.25, 62.53, 125, 250, 500 and 1000 (100 µL/well) were added into the wells and the plate was incubated for 72 h at 37°C in a 5% CO₂ humidified incubator.¹⁶,⁴⁷ Then, 100 µL of the reagent per well was added into the plate and incubated for 1 h at 37°C in 5% CO₂ incubator before the plate was read using a multimode microplate reader (Tecan). The dose–response graph was plotted by calculating percent cell viability using the equation below (Equation 3):

(3)

In addition, the inhibitory concentration, which caused 50% growth inhibition (IC₅₀) was determined, using an online calculator (https://www.aatbio.com/tools/ic50-calculator) as previously described.¹⁶,⁴⁶

Independent experiments were repeated at least three times, and the data were expressed as mean ± standard deviation for all triplicates within an individual experiment. Data were analyzed using a Student’s t-test. p < 0.05 was considered significant.

Results and Discussion

Reaction

Figure 1 indicates a schematic diagram of the possible chemical compounds in the crude extract of G. mangostana fruit peel, which served as a stabilizing agent and reacted with a mixture solution of Fe³⁺/Fe²⁺ ions by using a facile co-precipitation method to produce Fe₃O₄ NPs. The formation of the green synthesized Fe₃O₄ NPs was observed through color change of the solution from purple to black. The observed response of the synthesized Fe₃O₄ NPs to an external magnet confirmed the magnetic properties of the NPs. The physiochemical properties of the green synthesized magnetic samples were studied, and the samples were evaluated by magnetic hyperthermia analysis and also their cytotoxicity assays were analyzed towards colon cancer cell line (HCT116) and normal cell line (CCD112).

Figure 1 A schematic of the chemical compounds in the crude extract of G. mangostana fruit peel which served as a green stabilizer to synthesize Fe₃O₄ NPs for magnetic hyperthermia study and elimination of the colon cancer cells.

X-Ray Diffraction (XRD) Analysis

Figure 2 shows the XRD spectra for the synthesized Fe₃O₄ NPs. The NPs present a similar pattern and diffraction peaks at 2θ = 30.44°, 35.82°, 43.47°, 53.92°, 57.45°, 63.01°, and 74.69°, which respectively are corresponding to (220), (311), (400), (422), (511), (440), and (533) crystal planes of the pure cubic spinel crystal structure phase of Fe₃O₄ based on the literature data (JCPDS file no: 00-003-0863).⁶,¹⁶ Debye-Scherrer equation (Equation 4(4) ) was used to measure the crystallite size of the synthesized Fe₃O₄ NPs.⁴⁸

(4)

Figure 2 XRD spectra of the synthesized Fe₃O₄ NPs (S0–S4).

Where hkl is the Miller indices of the lattice planes being examined, D_hkl is the size of crystallite in direction perpendicular to the lattice planes, K is the crystallite-shape factor with a Scherrer constant = 0.9 for an absence information of crystallite-shape, λ is a wavelength of X-rays = 0.154059 nm, β_hkl is a full width at half maximum (FWHM) of the XRD diffraction peak in radians in 2θ scale, and θ is the half diffraction angle of the peak.⁴⁹ By using the equation above, crystallite size of the strongest reflection at 35.82° (311) peak for S0, S1, S2, S3, and S4 samples was 12.84 nm, 12.03 nm, 11.52 nm, 9.34 nm, and 7.81 nm, respectively. From the XRD patterns of all the samples, purity of the crystalline phase was acceptable due to low impurity peaks. Therefore, the above results could indicate the successful use of the extract as a green stabilizer in the synthesis of Fe₃O₄ NPs (magnetite phase).

Transmission Electron Microscopy (TEM)

Figure 3A–D illustrate TEM image, particle size distribution, selected area electron diffraction (SAED) pattern, and electron diffraction pattern of the S2 sample, respectively. As seen in Figure 3A, most of the particles were nearly spherical with potentially good dispersity and minor agglomeration. The presence of the agglomeration could be due to the van der Waals forces for binding particles together and also shear forces that can be applied on the nanoscale.⁵⁰ In addition, the presence of the hydroxyl groups in the peel extract could possibly led to the agglomeration. A histogram of the NP size distribution was plotted with 220 counts as shown in Figure 3B. The size distribution was in a range of 6 to 20 nm with a mean size of 13.42±1.58 nm, and a standard deviation of 2.74±0.16 nm. Based on the SAED pattern in Figure 3C, diffraction rings of the Fe₃O₄ phase were indexed as (311), (220), (400), (422), (511), (440) and (533), which was in good agreement with the XRD results.⁶ Figure 3D shows electron diffraction pattern of S2, indicating a regular and uniform crystallinity with lattice spacing on TEM image around 0.28 nm. Therefore, TEM study displayed that the biosynthesized Fe₃O₄ NPs possessed nearly spherical shapes with good crystallinity.

Figure 3 (A) TEM image, (B) histogram of the Fe₃O₄ NPs distribution, (C) SAED pattern, and (D) electron diffraction pattern of the synthesized Fe₃O₄ NPs of S2.

Field Emission Scanning Electron Microscopy (FESEM) and Energy Dispersive X-Ray Spectroscopy (EDS)

Figure 4A and B demonstrate FESEM images and EDS of S0–S5, respectively. As seen in the images, the nano-sized particles were nearly spherical. In addition, the peel extract potentially rolled as a protective agent to cover the surface area of the Fe₃O₄ NPs. Thus, increasing the coating ratio around the NPs was in line with the enhancement of the extract concentration. However, this coating layer may not be seen for S0 as a naked Fe₃O₄ NPs. Irregular shapes were detected because of the agglomeration process possibly related to the strong inter-particles and Van der Waals forces, high surface area to volume ratio, and magnetic attraction among the Fe₃O₄ NPs.⁶,⁵¹ The S5 sample illustrated the image of the peel extract with an unorganized structure. Figure 4B indicates EDS spectra of the samples. The extract powder (S5) presented 64.70 wt.% and 35.80 wt.% of carbon and oxygen, respectively. These components were potentially deposited in the green synthesized Fe₃O₄ NPs. Thus, the samples containing higher concentration of the peel extract possessed higher proportion of carbon and oxygen. In this manner, the ratio of carbon was higher in S4 than S1, showing the successful use of the extract as a stabilizer in the green synthesis of Fe₃O₄ NPs. Therefore, EDS spectrum of the Fe₃O₄ samples exhibited good purity of the compounds, owing to the presence of iron, carbon, and oxygen elements.

Figure 4 (A) FESEM image (scale bar = 200 nm ^____) and (B) EDS spectra of the synthesized Fe₃O₄ NPs (S0–S4) and the crude extract of G. mangostana fruit peel (S5).

Vibrating-Sample Magnetometer (VSM)

The VSM study of the synthesized Fe₃O₄ samples is illustrated in Figure 5A and B. The saturation magnetization of S0–S4 was found to be 73.15, 69.42, 64.35, 59.50, and 49.80 emu/g, respectively, at room temperature. The samples with the higher extract ratio showed lower magnetic properties, indicating the contribution of the extract stabilizer for the biosynthesis of Fe₃O₄ NPs. When the magnetic field was removed, an increase in the applied field rose the magnetic moment/mass. Hysteresis loops occurred when the external magnetic field was applied to the NPs,⁵² and magnetization decreased from a plateau value to zero. The increase of the extract ratio decreased and increased the saturation magnetization and the coercivity of the samples, respectively. Of this, among the Fe₃O₄ samples, S0 as a bare magnetic NPs indicated the lowest coercivity of 10.80 Oe and the highest saturation magnetization of 73.15 emu/g. However, S4 containing the highest extract ratio (10 wt.%) showed the highest coercivity (87.21 Oe) and the lowest saturation magnetization (49.80 emu/g). This is due to the non-magnetic nature of the extract as coated the magnetic NPs. It can be understood from the VSM results, the green synthesized Fe₃O₄ NPs could be conveyed by an external magnetic field and potentially directed to a specific target inside the body for biomedical applications, although future studies would need to focus on.

Figure 5 (A) Saturation magnetization and (B) coercivity of the synthesized Fe₃O₄ NPs (S0–S4).

Fourier-Transform Infrared Spectroscopy (FTIR)

Figure 6 shows FTIR spectra of S0–S5. The samples with higher concentration of the extract stabilizer displayed transmittance peaks closer to the crude extract alone in a similar wavenumber range. The intensity of C=O stretching vibration and C-O stretch increased due to the presence of the carbonyl group in the anthocyanin structure and alcohol group, respectively.⁵³ In the green synthesized Fe₃O₄ NPs samples (S1-S4), a broad band between 1621 cm⁻¹ and 1649 cm⁻¹ could be due to C=O stretching vibration.⁶ Furthermore, the peaks in a range from 3442 cm⁻¹ to 3596 cm⁻¹ were assigned to the hydroxyl group, representing the O-H stretching vibration, which was higher for the Fe₃O₄ NPs samples with the higher extract ratio. The functional groups of O-H, C=O, and C-O could possibly indicate the presence of gartanin compound in the samples. Gartanin is a chemical compound in xanthone isolated from G. mangostana fruit peel.⁵⁴ In all spectra of the synthesized Fe₃O₄ NPs, peaks appeared at a range from 636 cm⁻¹ to 721 cm⁻¹, showing characteristic of metal-oxygen band, which was the Fe-O stretching vibration of Fe₃O₄ NPs.⁵⁵ These stretching vibration bands could be linked to the metal in the tetrahedral and octahedral site.⁵⁶ Based on FTIR analysis of Fe₃O₄ NPs stabilized with Lagenaria siceraria extract, the stretching vibration at 3354 cm⁻¹, 1701.55 cm⁻¹, and 624 cm⁻¹ represented the presence of O-H bond stretching, C=O stretching, and Fe-O stretching, respectively.⁶ From the FTIR spectrum of S5, a peak at 3598 cm⁻¹ was assigned to the presence of O-H stretch and H-bonding and a peak at 2911 cm⁻¹ indicated the C-H stretching vibration band.⁵³ The characteristic of the carbonyl group is the C=O stretching vibration at 1652 cm⁻¹, whereas, the presence of the aromatic group was indicated by the band at 1460 cm⁻¹. The peaks between 1300 cm⁻¹ to 1000 cm⁻¹ were attributed to the functional group of C-O as it potentially related to the alcohols, ethers, ester, carboxylic acids, and amides in the crude extract of G. mangostana fruit peel.⁵³ Compared to other green synthesized Fe₃O₄ NPs, the extract showed the highest impact on the S4 sample, which contained the highest extract ratio (10 wt.%). In this manner, the O-H stretching peak of the extract’s spectrum shifted from 3598 cm⁻¹ to 3470 cm⁻¹ after the biosynthesis of Fe₃O₄ NPs. Meanwhile, the peak for the C=O and C-O stretching vibration of S5, respectively, shifted from 1652 cm⁻¹ to 1628 cm⁻¹ and from 1015 cm⁻¹ to 1031 cm⁻¹ after capping with Fe₃O₄ NPs in the S4 sample. Thus, the FTIR results could indicate that the extract successfully served as both stabilizing and capping agents during the green synthesis process of the Fe₃O₄ NPs.

Figure 6 FTIR spectra of the synthesized Fe₃O₄ NPs (S0–S4) and the crude extract of G. mangostana (S5).

Dynamic Light Scattering (DLS) Analysis and Impact of Storage Time on Fe₃O₄ Nanofluids

Table 1 depicts DLS results with values of zeta potential, hydrodynamic size, and polydispersity index of the samples, respectively. Adding the extract stabilizer increased the zeta potential value and also the hydrodynamic size. Among the synthesized magnetite samples, S0 as a bare Fe₃O₄ NPs sample had the lowest hydrodynamic size and the zeta potential value of 95.93±2.2 nm and −20.72±1.7 mV, respectively. Although, S4 with the highest stabilizer concentration (10 wt.%) showed the highest hydrodynamic size and zeta potential value of 176.15±1.46 nm and −34.92±1.26 mV, respectively. This may exhibit that coating the NPs with the plant extract provided a repulsive force between the NPs.⁵⁷ Compared to the TEM size, the DLS size was larger for S2. This can be explained that DLS indicates the combination of particle size and surrounding diffuse layer of the particle; however, TEM analysis is attributed to the particle size alone. In the aqueous media, the presence of the particle−particle interactions affected hydrophobic attraction energy between Fe₃O₄ NPs to attract each other, albeit, hydrophilic extract potentially governed hydrophilicity of the NPs to improve their thermodynamic stabilities.⁵⁸ The extract may also act as a tailoring agent to improve the surface features of the synthesized Fe₃O₄ NPs in the aqueous media. As value of zeta potential increased, polydispersity index decreased, showing that zeta potential was inversely attributed to polydispersity index. All the samples displayed the polydispersity index values below 0.7. This indicates that the extract of G. mangostana fruit peel as a new stabilizer potentially caused the nucleation of the Fe₃O₄ with a narrow dispersity.⁶ In a different study, Sathishkumar et al⁵⁷ synthesized Fe₃O₄ NPs using a stabilizer of Couroupita guianensis Aubl. fruit extract, that the zeta potential of the fabricated NPs was found to be −26 mV.

Table 1 Zeta Potential, Hydrodynamic Particle Size, and Polydispersity Index of the Synthesized Sample Suspensions During Four Weeks of Storage

	As-Prepared	After 1 Week of Storage	After 2 Weeks of Storage	After 3 Weeks of Storage	After 4 Weeks of Storage
Sample	Zeta potential (mV)
S0	−20.72±1.7	−16.96±1.1	−15.26±2.2	−14.58±1.6	−12.00±1.6
S1	−25.61±1.8	−24.32±0.5	−23.77±1.3	−22.24±1.4	−20.22±1.9
S2	−28.64±1.3	−27.44±1.0	−26.92±1.2	−25.14±0.2	−22.71±1.5
S3	−29.58±1.1	−27.87±1.4	−25.45±1.8	−25.17±1.3	−24.36±1.4
S4	−34.92±1.2	−34.25±0.9	−33.14±1.1	−31.83±2.2	−30.15±2.7
S5	−15.66±1.2	−11.54±2.0	−8.23±2.3	−7.65±1.5	−6.82±0.5
Hydrodynamic particle size (nm)
S0	95.93±2.2	97.23±1.6	109.98±1.3	125.00±1.2	134.00±2.6
S1	107.90±0.2	110.09±1.3	115.27±0.5	121.20±1.1	128.20±2.1
S2	120.34±1.7	125.47±1.0	127.54±1.2	130.90±1.0	134.84±1.5
S3	145.82±1.0	150.63±1.4	150.81±1.8	158.28±1.4	165.61±2.1
S4	176.15±1.4	180.35±0.9	184.00±1.1	190.44±1.8	198.27±2.0
Polydispersity index
S0	0.31±0.03	0.40±0.06	0.42±0.03	0.45±0.02	0.50±0.06
S1	0.29±0.02	0.35±0.04	0.38±0.05	0.40±0.01	0.41±0.07
S2	0.27±0.02	0.31±0.06	0.33±0.02	0.35±0.04	0.39±0.09
S3	0.25±0.04	0.29±0.09	0.33±0.06	0.34±0.04	0.35±0.02
S4	0.23±0.01	0.25±0.03	0.28±0.04	0.30±0.08	0.32±0.04

As provided in Table1, the zeta potential values of S0–S4 after the storage of 1, 2, 3, and 4 weeks were slightly and gradually decreased. This can indicate the low particle agglomeration and aggregation are in the line with time, which is similar to different reports.⁴¹,⁵⁹ The zeta potential values exhibited that the Fe₃O₄ NPs suspensions with the higher extract ratio could present better colloidal stability than those NPs with the lower extract ratio after 4 weeks of storage. Particle size of the magnetic nanofluid samples at different storage time was obtained from DLS measurement (Table 1). Figure 7A and B show the hydrodynamic size distributions of S0–S4 as prepared and after 4 weeks of storage, respectively. An increase in the average particle size after the prolonged storage could be related to a minor agglomeration in the Fe₃O₄ NPs. Albeit, there was no major agglomeration visually noticed on the sample suspension even after 4 weeks of storage. In addition, the polydispersity index values were almost unchanged with the values below 0.7 even after 4 weeks of storage. Therefore, the Fe₃O₄ NPs suspensions displayed good stability during 4 weeks of storage.

Figure 7 Hydrodynamic size distributions of the nanofluid samples containing Fe₃O₄ NPs (A) as-synthesized and (B) after four weeks of storage.

Magnetic Hyperthermia Study

The hyperthermia performance of the Fe₃O₄ NPs samples was assessed through measuring their temperature rise under the exposure of different AMF strengths to determine the minimal AMF strength. Further, the T_H maintained within the secure hyperthermia range in order to prevent unpredictable disorders and subsequently perform a successful MHT. The thermal profiles of the nanofluid samples containing 1 mg/mL of Fe₃O₄ NPs under the exposure of varied AMF strengths of 23.60, 31.50, and 39.37 kA.m⁻¹ ‎over 3600 sec are illustrated in Figure 8. All the graphs illustrate almost a similar thermal behavior throughout the treatment time under different AMF exposure. Clearly shown that the temperature increased significantly until 360 sec (Phase 1), followed by a moderate temperature rise until heating intervals of approximately 1800 sec (phase 2). Then, the temperature raised negligibly in the aqueous medium (phase 3). The highest effect of the AMF strength was observed in phase 1, where the slope of the temperature rise (dT/dt) at this stage which was significantly higher than the other phases. Hence, the slope of the phase 1 for all of the thermal profiles was considered for the SAR value measurements. The SAR values of the nanofluid samples exposed to different AMF strengths are provided in Table 2. The SAR values increased considerably with increasing AMF strengths from 23.60 kA.m⁻¹ to 39.37 kA.m⁻¹ due to the heating loss mechanisms,⁶⁰ which is also in line with previous research reports such as bare superparamagnetic Fe₃O₄⁴⁴ and polycaprolactone-coated Fe₃O₄⁴⁵ and γ-Fe₂O_3.⁴² The SAR values reduced remarkably with increasing the extract concentrations from ‎0 to 10 wt.%, which might be ascribed to the non-magnetic nature, which acted as a heating barrier for the Fe₃O₄ NPs. Thereby, the highest and the lowest SAR values were respectively obtained for S0 and S4 at each AMF strength. Nevertheless, S1 with the lowest extract concentration could render significant SAR value as compared to other extracted counterparts with values of 165.74‎±0.5 W.g⁻¹, 212.06±0.4 W.g⁻¹, and‎ ‎246.86±0.2 W.g⁻¹ at 23.6 kA.m⁻¹, 31.5 kA.m⁻¹, and 39.37 kA.m⁻¹, respectively.

Table 2 The Achieved Hyperthermia Properties of the Nanofluid Samples Containing Fe₃O₄ NPs and Varied Extract Concentrations Exposed to Different AMF Strengths Over 360 Sec

Sample	Frequency (kHz)	Current (A)	AMF Strength (kA.m^−1)	TH (°C)	SAR (W.g^−1)
S0	318	75	23.6	49±0.2	188.67±0.5
	313	100	31.5	51±0.2	218.64±0.3
	312	125	39.37	54±0.6	260.96±0.4
S1	318	75	23.6	46±0.3	165.74±0.5
	313	100	31.5	49±0.4	212.06±0.4
	312	125	39.37	52±0.3	246.86±0.2
S2	318	75	23.6	45±0.4	144.59±0.5
	313	100	31.5	47±0.3	195.72±0.2
	312	125	39.37	49±0.6	239.80±0.4
S3	318	75	23.6	43±0.5	128.72±0.3
	313	100	31.5	45±0.3	178.79±0.2
	312	125	39.37	48±0.4	225.70±0.3
S4	318	75	23.6	42±0.2	98.74±0.6
	313	100	31.5	43±0.2	130.48±0.3
	312	125	39.37	44±0.4	179.85±0.1

Figure 8 The temperature profiles of the nanofluid samples containing the synthesized Fe₃O₄ NPs of S0–S4, which were obtained after exposure to the varied AMF strengths of 23.60 kA.m⁻¹, 31.50 kA.m⁻¹, and 39.37 kA.m⁻¹.

Although the S0 sample could show the highest SAR values over the range of AMF strengths, this sample was incapable of maintaining T_H at the secure hyperthermia range (42–47°C) even at the minimal applied AMF strength of 23.6 kA.m⁻¹. However, the extract could potentially control the medium temperature ‎of the nanofluid under the hyperthermia condition to produce sufficient ‎thermal energy within the secure hyperthermia range.⁵⁶ The data provided in Table 2 clearly implies that the S4 sample with the highest extract concentration (10 wt.%) could retain a medium temperature rise within the secure hyperthermia range which is appropriate for prospective hyperthermia treatment. The S4 solution indicated temperature rise (with the SAR values) of ‎42±0.2‎°C (98.74±0.6 W.g⁻¹),‎ 43±0.2‎°C (130.48±0.3 W.g⁻¹) and‎ 44±‎0.4°C (179.85±0.1 W.g⁻¹) at AMF strengths of 23.6 kA.m⁻¹, 31.5 kA.m⁻¹, and 39.37 kA.m⁻¹, respectively. The extract of G. mangostana peel ‎as a novel stabilizer could trigger a positive effect on Fe₃O₄ NPs ‎preparation to perform a successful MHT. Therefore, according to the obtained magnetic hyperthermia results from Table 2, the synthesized Fe₃O₄ NPs ‎stabilized with the extract could be considered as a promising candidate for hyperthermia ‎therapy of cancer. ‎

In vitro Cytotoxicity Assay

Figure 9A and B demonstrate the cytotoxic effects of S0–S5 against colon cancer (HCT116) and normal cell lines (CCD112) after 72 h of treatment, respectively. Dose-dependent killing effect was seen in all the tested samples in both cell lines. Table 3 indicates the IC₅₀ of S0–S5 against the colon cancer and normal cell lines. The crude extract of G. mangostana fruit peel showed higher anticancer activity compared to the extracts of different plants such as Punica granatum fruit peel,⁶ Juglans regia green husk,¹⁶ and seaweed (Kappaphycus alvarezii).⁶¹ This could be due to the presence of xanthone and α-mangostin in the crude extract of G. mangostana peel.⁶² Among all the magnetic samples, S4 with the highest extract concentration (10 wt.%) showed the highest killing effect against the cancer cells with the lowest IC₅₀ (99.8 µg/mL). The IC₅₀ of S4 in the normal cells was found to be 140.80 µg/mL, which was slightly higher than that of the cancer cells. Figure 10 summarizes the toxicity of S4 on HCT116 cancer and CCD112 normal cell lines. At 125 µg/mL and 250 µg/mL, S4 killed 52±0.54% and 63±1.68% of the cancer cells, respectively. At 500 µg/mL and 1000 µg/mL, S4 caused respectively, 72±1.15% and 80±0.61% killing effects against the cancer cells; however, it displayed 63±2.17% and 68±1.05% elimination of the normal cells, respectively. The anticancer activity of the magnetic samples possibly could be due to release of iron ions from the Fe₃O₄ NPs following its breakdown and the subsequent Fe reutilization process by cells for Fe metabolism, which could be one of the cancer cell killing mechanisms.⁶³ It is worth mentioning that the extract as a stabilizer and capping agent possibly tailored assemblies of Fe₃O₄ NPs with a temperature-responsive structure, and self-heating capacity to destroy the cancer cells. The nucleation of the NPs in the presence of the extract was translated into an efficient bonding between the phenolic compounds of the extract into/onto the NPs to strongly avoid deep oxidation, possibly suitable for various biomedical applications.¹³,¹⁵ As previously reported,⁶⁴ normal cells could decrease their heat-sensitivity five or six times than that of the growing situation at the conflux, but the cancer cells were unable. In this manner, cancer cells might indicate higher temperature (above 37°C) compared to normal cells (below 37°C). This slightly increased temperature of the cancer cells might trigger higher release of the phenolic compounds against the cancer cells. To our best understanding from these in vitro results, certain concentrations of the biosynthesized Fe₃O₄ NPs effectively eliminated the colon cancer cells; however, further improvements such as surface modifications and/or polymer coating on the NPs may decrease the toxicity of the G. mangostana/Fe₃O₄ samples on the normal cells.

Table 3 Inhibitory Concentration (IC₅₀) of S0–S5 on HCT116 and CCD112 Cell Lines

Sample	IC50 of HCT116 (µg/mL)	IC50 of CCD112 (µg/mL)
S0	>1000	750.1
S1	347.6	315.3
S2	698.9	542.6
S3	716.5	539.1
S4	99.8	140.8
S5	133	93.5

Figure 9 Relative viability of (A) CCD112 colon normal cell line and (B) HCT116 colon cancer cell line treated with the synthesized Fe₃O₄ NPs (S0–S4) and the crude extract of G. mangostana (S5). Data are expressed as mean ± standard deviation for triplicates within an individual experiment. Statistical significance was performed using a Student’s t-test (*p < 0.05).

Figure 10 Relative viability of CCD112 colon normal cell lines and HCT116 colon cancer cell lines treated with Fe₃O₄ NPs of S4 containing 10 wt.% of G. mangostana peel extract.

Conclusion

In this work, the green synthesis of Fe₃O₄ NPs was conducted by using a novel stabilizer of the crude extract of G. mangostana fruit peel at various concentrations (1, 2, 5, and 10 wt.%). TEM analysis of the magnetic sample containing 2 wt.% of the extract showed the spherical NPs with an average size of 13.42±1.58 nm and its SAED pattern was related to the Fe₃O₄ magnetite phase, which was in well agreement with the XRD data. Based on the FESEM images, the green synthesized NPs were surrounded by the extract. The saturation magnetization values of the samples with the lowest and highest extract concentration were 49.80 emu/g and 69.42 emu/g, respectively. The FTIR data indicated that the green synthesized Fe₃O₄ NPs contained the spectroscopy peaks related to the iron and the extract with xanthone, carboxylic, alcohol, and aromatic groups. The Fe₃O₄ nanofluids containing the highest ratio of the extract stabilizer indicated an improved particle charge of −34.92±1.26 mV and hydrodynamic size of 176.15±1.46 nm. Furthermore, the DLS analysis exhibited that the Fe₃O₄ nanofluids showed the colloidal stability even after 4 weeks of storage. The Fe₃O₄ nanofluids can be potentially used in hyperthermia therapy, since they mostly caused the T_H increase at the secure hyperthermia range (42–47°C) and the acceptable SAR values with the thermosensitive performances under exposure to AMF. Cytotoxicity assays of the samples were evaluated against CCD112 normal and HCT116 colon cancer cells. Particularly, the Fe₃O₄ NPs containing 10 wt.% of the extract showed a lower IC₅₀ value (99.80 µg/mL) in HCT116 than in CCD112 (140.80 µg/mL). Consequently, this study indicated that the extract of G. mangostana fruit peel is a low-cost stabilizing and capping agent to improve the colloidal stability and physiochemical properties of the Fe₃O₄ NPs for potential hyperthermia therapy and anticancer treatments. Further investigations can lead to enhance the selectivity of the samples towards the cancer cells with the minimal toxicity on the normal cells. It can be considered to load an anticancer drug onto the green synthesized Fe₃O₄ NPs for the future study.

Acknowledgments

This research was funded by Takasago Thermal Engineering Co. Ltd. grant (#4B422) from the research management center (RMC) of Universiti Teknologi Malaysia (UTM) and Malaysia-Japan International Institute of Technology (MJIIT). Special thanks to School of Medical and Life Sciences, Sunway University Malaysia for the cell culture and laboratory facilities for the in vitro studies. This work was also supported by the UHK project VT2019-2021.

Disclosure

The authors declare no conflicts of interest for this work.