Tetrahydroxyquinone: A Suitable Coating for Ferrofluids Used in Magnetic Hyperthermia

Abstract

In this work, tetrahydroxyquinone (THQ) was used for the first time to coat iron oxide nanoparticles (IONPs) and to carry out in vitro experiments in magnetic hyperthermia. Synthesis by co-precipitation resulted in spherical IONPs with a core diameter of 13 ± 3 nm and covered by a 0.5 nm thick coat of THQ, which provided them with a reasonably good zeta potential of ζ = −28 ± 2 mV at pH = 7.3, and thus colloidal stability. The magnetic properties of the THQ-coated IONPs are promising: the low coercive field of H_c = 7 Oe, the high magnetic saturation of M_s = 70.5 emu/g and the low blocking temperature of T_b = 273 K indicate superparamagnetic characteristics at room temperature. Additionally, a high specific absorption rate SAR = 135 W/g (at 300 Oe and 530 kHz) was determined. Cell biological experiments using the human cell line HT-29 evidenced negligible cytotoxicity up to 2 mg/mL. Magnetic hyperthermia (MHT) assays demonstrated fast and reliable heating and reduced the metabolic activity of the cells to 42% upon reaching 42 °C within 15 min. The production of ROS by THQ-coated IONPs could not be detected, which may indicate a reduction in the undesired side effects caused by oxidative stress. Considering these good physicochemical and cell biological properties, this ferrofluid is a promising candidate for the initiation of in vivo experiments for cancer treatment by MHT in murine models.

1. Introduction

Tetrahydroxyquinone (THQ), or tetrahydroxy-p-benzoquinone C₆O₂(OH)_4, can be synthesised by oxidation of myo-inositol [1], which is present in many foods, such as vegetables and seeds. The crystalline structure was determined to be a centrosymmetric quinone with two water molecules associated by hydrogen bonds [2]. In the field of electrochemistry, THQ represents an interesting organic compound to be used as electrodes in lithium [3] or other eco-friendly and high-efficient rechargeable batteries [4,5,6], because it can form salts with very high conductivity when mixed with metal ions, such as Fe(II) or Cu [7,8]. Due to its redox activity, THQ can form stable complexes with various metal ions [9], and moreover, it has good chelating properties [10].

In the middle of the last century, THQ was used as an oral treatment of keloid scars, and its effect on connective tissue was analysed in several patients [11]. Souza-Pinto et al. (1996) discovered that cytotoxic effects of THQ on V79 Chinese hamster fibroblasts are caused by a disturbance of the intracellular Ca²⁺-homeostasis and the generation of H₂O₂, which impair mitochondrial function and DNA synthesis [12]. Moreover, in cell culture experiments, THQ increased the formation of reactive oxygen species (ROS), which, in combination with a reduced activity of the protein kinase B pathway, induced apoptosis in HL60 leukaemia cells [13]. Surprisingly, only a few scientific publications address nanometric structures involving THQ for biomedical or biotechnology applications. However, there are many recent scientific studies on ferrofluids of iron oxide nanoparticles (IONPs) with surface modifications using other phenol-derived compounds, such as dopamine or catechol [14,15,16,17]. These works focused on cancer treatment, applying magnetic hyperthermia on different cell lines or rodent models; thus, it was necessary to accurately determine several physicochemical properties, such as colloidal stability, size distribution of the IONPs, specific absorption rate (SAR) capability, magnetic properties, cytotoxicity and other biologically relevant properties, such as ROS production.

The aim of this work was the synthesis and characterisation of a new ferrofluid of THQ-coated IONPs, with the aim of evaluating its suitability and performance in magnetic hyperthermia. Additionally, its physicochemical properties were compared with similar ferrofluids of IONPs, which are coated with other phenol-derived compounds.

2. Materials and Methods

2.1. Synthesis of Uncoated and Coated IONPs

The uncoated IONPs were co-precipitated under an inert N₂ atmosphere, following the steps as described in [14,15,16]. First, 15 mL 0.1 M FeSO₄ × 7 H₂O (Fermont, Monterrey, Mexico) was added to 30 mL 0.1 M FeCl₃ (Sigma-Aldrich, St. Louis, MO, USA), and the mixture was stirred at 350 rpm for 5 min, maintaining a temperature of 30 °C. Then 3 mL of the precipitating reagent (5 M NH₄OH) was slowly added (with a flow rate of 1 mL/min). Afterwards, the suspension was centrifuged (45 min, 3000 rpm) obtaining a liquid phase and a pellet of precipitated IONPs. These IONPs were washed four times; this was achieved by retaining them using a strong magnet (1.5 T) and replacing the supernatant with 100 mL distilled water. After the last washing step, the first ferrofluid with uncoated IONPs was obtained.

The preparation of the ferrofluid with THQ-coated IONPs was, with small variations, carried out as described in [16] 40 mL 4.35 mM THQ (Sigma-Aldrich). It was solubilised for 24 h, mixed with 40 mL of the first ferrofluid previously synthesised with uncoated IONPs and sonicated for 35 min. Afterwards, the suspension was stirred at 400 rpm for 24 h at 30 °C and the resultant THQ-coated particles were magnetically separated: the precipitate contained clusters of coated beads, which were discharged, while the supernatant contained the homogeneously suspended THQ-coated IONPs. This supernatant was centrifuged (90 min, 6000 rpm) and the pellet resuspended in 10 mL distilled water.

2.2. Physicochemical Characterisation

The union between THQ and the surface of the IONPs was determined by infrared measurements between 450 cm⁻¹ and 4000 cm⁻¹ using the FTIR technique. The samples were dried in a vacuum oven (Lindberg/Blue M, Thermo Scientific, Waltham, MA, USA) at 2.7 kPa and 45 °C for 12 h. The powdered samples were analysed with an FTIR spectrometer (Nicolet iS5, Thermo Scientific) with the reflectance totally attenuated. The measurements for each ferrofluid were performed in duplicate. Additionally, pure THQ was analysed as a reference for the typical resonant peaks of the coating itself.

The crystalline structure of the IONPs was revealed by X-ray diffractometry using an Empyrean device from Malvern Panalytical (Malvern, UK). After depositing dried samples of uncoated and coated IONPs onto the sample holder, they were analysed using an angle of incidence covering the interval 5 ≤ 2θ ≤ 80° with 0.02° steps and a sampling time of 30 s.

The size distribution and the shapes of the uncoated and coated IONPs were determined by transmission electron microscopy (TEM). Aliquots from each ferrofluid were diluted 1:100, placed onto FCF-200-Cu grids and dried in a vacuum chamber at room temperature. Then, sequences of TEM micrographs with 200,000× magnification and 80 kV accelerating voltage were taken (JEM-2100, JEOL, Tokyo, Japan) for posterior offline analysis.

To evaluate the magnetic properties of dried uncoated and coated IONPs, the samples were thermalised in a silicone sample holder of the magnetometer (VSM VersaLab, Quantum Design, San Diego, CA, USA); then, the intensity of the magnetic field was varied from −30 to 30 kOe and the respective magnetisations were registered. In a second step, the magnetisation was quantified following the ZFC-FC protocol, i.e., at a constant magnetic field of H = 100 Oe, and the temperature was increased from 50 to 400 K.

The colloidal properties of the THQ-coated IONPs were measured using a Zetasizer meter (ZS90, Malvern, UK). Samples were filled into its special capillary cells (DTS1070), where the pH was adjusted in the range of pH 2.1 to 12.5 using 0.1 M HCl and NaOH, respectively. Then, the zeta potential, the hydrodynamic diameter and the polydispersity index were analysed in triplicate.

The amount of THQ attached to the surface of the IONPs was estimated thermogravimetrically. Samples of approximately 10 mg (uncoated and coated IONPs) were deposited onto a small platinum plate and heated, under a N₂ atmosphere, from 40 to 800 °C with a heating rate of 20 K/min while the changes of the mass were registered.

To quantify the specific absorption rate (SAR) and hence the absorbed power density of the THQ-coated IONPs, a home-made induction heating system was used [18] (MX Pats. 65,340, Mx/a/2018/002848, Ocotlán, Mexico). Eppendorf tubes, each containing a 1 mL sample, were placed in the cavity of the induction generator. The frequency of the magnetic field (f) was set to 530 kHz while the magnetic field (H) was increased from 0 to 301 Oe (in steps of 50 Oe) within 5 min; simultaneously, a fluoroptic sensor (Luxtron-One, LumaSense Technologies, Santa Clara, CA, USA) recorded the temperature of the ferrofluid.

2.3. Evaluation of ROS Formation

A solution of methylene blue (MB) becomes decolourised by radicals due to its oxidative decomposition. In order to evaluate the ROS formation by THQ-coated IONPs, six Eppendorf tubes were prepared, as summarised in

Table 1. Composition of the six labelled samples prepared to degrade methylene blue (MB).

Label	Content
	MB	H2O2	THQ	IONP-THQ
MB	+
MB H2O2	+	+
MB THQ	+		+
MB THQ H2O2	+	+	+
MB IONP-THQ	+			+
MB IONP-THQ H2O2	+	+		+

[16,19]. The final concentration of MB was 5 µg/mL, that of THQ was 0.2 mg/mL, that of THQ-coated IONPs (IONP-THQ) was 2 mg/mL and of H₂O₂ 0.9%. After having joined all required constituents, the samples were incubated at room temperature in the dark for 10 min to achieve the reaction with MB; subsequently, the samples were centrifuged at 10,000 rpm for 10 min (5415 D, Eppendorf, Hamburg, Germany) and the absorbance of all the supernatants was measured at a wavelength of 665 nm (Biopop, Mecasys, Daejeon, Korea). Additionally, the absorbance of the supernatant of the samples containing only IONP-THQ was subtracted from the absorbance of the corresponding samples MB H₂O₂ IONP-THQ and MB IONP-THQ. Values are expressed as means of 3 independent experiments, error bars indicate the standard deviation and the statistical significance (p < 0.05) was computed by a one-way ANOVA complemented with a Tukey’s post hoc test between multiple means, using the software Statistica 13.3 (StatSoft, Hamburg, Germany).

2.4. Cytotoxicity Analysis and Magnetic Hyperthermia (MHT)

The human cell line HT-29 (#HTB-38, ATCC, Manassas, VA, USA) was grown in DMEM (ATCC) supplemented with 10% foetal bovine serum (Gibco, Waltham, MA, USA) and 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) at 37 °C, 4% CO₂ and 95% of relative humidity. For the experiments, approximately 1.5 × 10⁵ cells were seeded onto a cover slide in 12-well-plates (cytotoxicity tests) or in single 2 mL-glass-wells (for MHT), containing 1 mL growth medium. Before starting the different experiments, the cells were incubated for 24 h to attach them on the cover slides.

For the cytotoxicity assays, 0.0 (control), 0.5, 1.0, 2.0 and 3.0 mg/mL of IONP-THQ or pure THQ were added in the range of 1 to 300 µg/mL, and the cells were incubated again for 24 h under standard conditions. Afterwards, the growth medium was removed, and the cells were washed three times with phosphate-buffered saline (PBS) to remove non-internalised IONPs. After that, 1 mL of fresh growth medium containing 2% water-soluble tetrazolium (WST-1; Clontech, Mountain View, CA, USA) was added into each well and the plate was incubated for 3 h. The metabolically active cells reduced the added tetrazolium to a formazan. At the end, the growth medium was centrifuged for 1 min at 10,000 rpm (5415 D, Eppendorf) and the absorbance of the formed formazan was measured at 440 nm (Biopop, Mecasys).

For the MHT assays, 2 mg/mL of IONP-THQ were added, and the cells were incubated for another 2 h to increase the endocytosis of the nanoparticles. Afterwards, the cells were subjected to MHT. In total, 20 min of irradiation was applied, using a fixed frequency of f = 530 ± 0.1 kHz; during the first 5 min of irradiation, each sample was heated up to a selected temperature (37, 39, 42, 45 or 48 °C), which was then manually maintained to be almost constant during the following 15 min. After the MHT, the cells were incubated for 24 h under standard conditions, then the old medium was removed, the cells were washed up to five times with PBS to remove non-internalised IONPs and 1 mL of growth medium containing 2% WST-1 was added. Then, the procedure to determine the metabolic activity was carried out as described above.

2.5. Measurement of Intracellular ROS Production

The possible ROS production by HT-29 cells was evaluated using the method suggested by Siddiqui et al. [20]. For this purpose, HT-29 cells were seeded on cover slides in 12-well plates (see Section 2.4), and their content and labels are described in

Table 2. Content and labels of the samples for ROS measurement.

Label	Content
	DCF	H2O2	THQ	IONP-THQ
NC	+
PC	+	+
DCF THQ	+		+
DCF THQ H2O2	+	+	+
DCF IONP-THQ	+			+
DCF IONP-THQ H2O2	+	+		+

. The final concentration of THQ was 0.2 mg/mL of IONP-THQ 2 mg/mL, and that of H₂O₂ was 0.0035%. Then the cells were incubated for 2 h under standard conditions (see Section 2.4), and afterwards washed repeatedly with PBS to remove the IONPs or THQ. Finally, the cells were kept in 1 mL growth medium without FBS, and 20 min before starting the measurements, 10 µL/mL 20 µM 2,7-dichlorodihydrofluorescein diacetate (DCFH₂-DA) was added to the cell culture to permeate freely into the cells. Both acetate moieties were hydrolysed by esterase activity to produce hydrophilic dichlorodihydrofluorescein (DCFH₂), which was trapped within the cells. Mainly due to the generated peroxides, DCFH₂ became oxidised to produce the green fluorescent dichlorofluorescein (DCF), which was observed using the epifluorescence microscope (Axioskop 40FL, Zeiss, Oberkochen, Germany).

3. Results and Discussion

The crystal structures of the uncoated and coated IONPs are shown in Figure 1a. In both spectra, the main peaks appear at the angles (2θ) 30.1°, 35.4°, 43.1°, 53.4°, 57.0° and 62.6°, which are associated with the respective Miller indices (2 2 0), (3 1 1), (4 0 0), (4 2 2), (5 1 1) and (4 4 0). Both X-ray diffraction patterns matched well with the standard patterns JCPDS 19-629 and JCPDS 39-1346 of magnetite or maghemite, but using XRD, it is not possible to determine their ratios. Although some authors affirm that magnetite is mainly formed by co-precipitation synthesis [21,22], the characteristic Miller indices (2 1 0) and (2 1 1) of maghemite (typically observed between 25° and 30°) may be hidden by the noise of the signals. Therefore, other types of tests, such as Mössbauer spectroscopy, may be necessary to elucidate the final structure [23]. The crystal size of the IONPs can be estimated by applying the Scherrer equation on the main peak of the plotted spectra [14,15,16,17,21,22], resulting in a diameter of 15.0 nm and 12.5 nm for the uncoated and coated samples, respectively.

The FTIR spectra of pure THQ and coated and uncoated IONPs are depicted in Figure 1b. The spectra of both types of IONPs exhibit the expected vibrations of the Fe–O bond of magnetite at 580 cm⁻⁷¹ [24]; additionally, the interaction between iron oxide and THQ modified the O–H stretching vibrations, as can be demonstrated by comparing the O–H signals at 2970 cm⁻¹. Additionally, the broad signal from 3000 to 3700 cm⁻¹ was caused by the interaction of deprotonated oxygen atoms of the THQ with the iron [25]. Further, a high electron delocalisation in O–H present in the segment HO–C=C–C=O caused its shift to a lower frequency. These stretching bands, located at 1500–1660 cm⁻¹ correspond to the bonds located at 1660 cm⁻¹ assigned to C=O stretching vibration, as can be seen in the spectra of THQ and THQ-coated IONPs. Finally, the signal from 900–1000 cm⁻¹ was associated with C–O stretching vibrations, as can be seen in the spectra of THQ and coated IONPs.

Under the transmission electron microscope, the IONPs could be observed as grey or dark round dots, which suggested in both cases a spherical shape (Figure 2a,b). The size distributions were estimated by analysing a series of micrographs using the software ImageJ (https://imagej.nih.gov/ij/ accessed on 1 May 2022). The uncoated IONPs had an average diameter of σ_IONP = 17 ± 5 nm, while the THQ-coated IONPs were with an average diameter of σ_IONP-THQ = 13 ± 3 nm, a little smaller (see bars plot and Gaussian fit of Figure 2c). This result coincided with the sizes estimated by XRD-analysis. That the coated IONPs were smaller than the uncoated IONPs was a consequence of the slightly different preparation. While the uncoated IONPs were concentrated by centrifugation (3000 rpm) and were then washed using magnetic separation, the procedure for the coated IONPs was inverted: first, they were magnetically separated and afterwards recovered from the supernatant by centrifugation (6000 rpm). Thus, the uncoated IONPs come from a precipitate, while the coated IONPs arose from a supernatant.

Thermogravimetric analysis revealed that uncoated IONPs lost only η = 3.5% of their initial mass when heated up to 400 °C, and the mass loss did not increase when heated up to 800 °C (Figure 3a). A similar behaviour was exhibited by the THQ-coated IONPs, although they lost η = 11.5% of their initial mass when heated up to 400 °C (Figure 3a). Analysing the speed of the mass loss, i.e., the derivates of Figure 1a, revealed in both cases a rapid mass loss in the range of 40–120 °C due to loosely bound surface water (Figure 3b). A second rapid mass lost could be observed, again in both cases, in the range of 160–240 °C, and was caused by the evaporation of by hydrogen-bound hydration water (Figure 3b). Finally, only for the THQ-coated IONPs, a third fast mass loss of about η = 6.5% in the range of 240–360 °C can be detected (Figure 3b), which can be attributed to the evaporation of THQ itself. Following the procedure described in [16], and considering σ_IONP-THQ = 13 ± 3 nm, ρ_THQ = 1.67 g/cm³ and ρ_Fe3O4 = 5.2 g/cm³, the thickness of the THQ-coat of the IONP-THQ can be estimated to τ = 0.5 nm. Similar values for σ and η are reported in [15,16], where IONPs, coated with other phenolic compounds, such as dopamine or catechol, were prepared by co-precipitation and were used in magnetic hyperthermia experiments.

The zeta potential (which describes the colloidal stability of a dispersion) for the THQ-coated IONPs for the pH range of 2.1 to 12.0 is shown in Figure 4a. At a pH = 7.3, which was close to the pH relevant for most biological systems, the zeta potential was ζ = −28 ± 2 mV, which indicated a reasonable colloidal stability. At a higher pH, the zeta potential increased slightly, while at a lower pH the zeta potential diminished rapidly. This coincided with the hydrodynamic diameter (HD) of the THQ-coated IONPs: at a low pH, soft clusters with a HD of >1000 nm were formed, while at pH 7.5 the HD reached a diameter of 250 nm (Figure 4b). Under these conditions, the polydispersity index PDI could be calculated to 0.24, which was almost an acceptable value for colloidal dispersions of nanoparticles [26]. Thus, at neutral pH, the THQ-coated IONPs had a negative electric charge and formed soft clusters. Similar values of ζ are reported in [15,16] using coatings of dopamine and catechol.

The respective magnetisations (M) of dried uncoated and coated IONPs were determined in the range of the magnetic field (H) from −30 to +30 kOe (Figure 5a). The magnetic saturation (M_s) was nearly 70.5 emu/g for the uncoated and 66.1 emu/g for the THQ-coated IONPs. That the coated IONPs had a 6.2% lower magnetic saturation than the uncoated IONPs was expected, because according to the TGA-measurements, they consisted of 6.5% diamagnetic THQ, which was magnetized in the opposite sense to H. Moreover, due to the bigger average size of the uncoated IONPs (Figure 2c), their coercive field (H_c) with 23 Oe was bigger than the one for the coated IONPs with 7 Oe (Figure 5a, inset). Furthermore, the dependencies of the magnetisation on the temperature (T) of both samples were obtained following the ZFC-FC protocol at 100 Oe (Figure 5b). While for the uncoated IONPs up to 400 K, no inflection point in the ZFC-curve can be observed, for the THQ-coated IONPs the inflection of the ZFC-curve occurs at the blocking temperature of T_b = 273 K. Thus, the low H_c of 7 Oe as well as the presence of a blocking temperature (preferably below room temperature T_b = 0 °C for biomedical applications) are strong evidence of a predominant superparamagnetic behaviour for the coated IONPs, and the weak coercivity can be associated with the presence of fewer ferrimagnetic grains. By contrast, the uncoated IONPs reached the blocked state of their magnetisation only because of their broad polydispersity (i.e., some particles were relatively big). Such superparamagnetic behaviour (coated IONPs) and blocked magnetisation (uncoated IONPs) have been observed for other IONPs of similar size coated with phenol-derived compounds [15,16]. Nevertheless, the THQ-coated IONPs exhibit even better magnetic properties, reaching a much higher M_s and a lower T_b.

The specific absorption rate (SAR) is an indicator of how well the IONPs can be heated in an alternating magnetic field. With the application of a constant frequency of 530 kHz and an increase in the amplitude of the magnetic field (H) from 0 to 301 Oe, the increase in the temperature (ΔT) of 1 mL sample containing c = 0.1% coated IONPs was registered over time (Δt). The averaged absorbed power density (p) can be described by the formula p = β(ΔT/Δt), with β = ρ c_V/c and ρ ≈ 1 g/cm³ and c_V ≈ 4.18 J/g·K (density and specific heat capacity of water, respectively). The dependence of p on H was determined following the procedure described in [14,15,16]. As shown in Figure 5c, the maximum absorbed power density was 135 W/cm³ at 301 Oe, which was nearly three times that of catechol-coated IONPs used in in vitro experiments for MHT [16] and 22% higher than for dopamine-coated IONPs previously used for MHT in in vivo trials [15].

The ROS-formation by Fenton reactions induced by THQ-coated IONPs was analysed by the degree of MB degradation (Figure 6). The absorbance of MB after H₂O₂ treatment (MB H₂O₂) was to some extent increased compared to the control (MB). THQ itself (MB THQ) did not reduce the absorbance of MB significantly, although according to the standard redox potentials, a reduction of MB to leuko-MB would be possible. In combination with H₂O₂ (MB THQ H₂O₂), the absorbance reduced slightly more; it is possible that H₂O₂ triggered the formation of THQ-radicals (semiquinone) [27], which then reduced MB to leuko-MB, or even induced a degradation of MB by ROS-formation. The THQ-coated IONPs (MB IONP-THQ) then increased the absorbance of MB slightly, nearly as much as THQ-coated IONPs spiked with H₂O₂ (MB IONP-THQ H₂O₂). This means that IONP-THQ neither induced ROS-formation nor exhibited a reducing effect. The redox capability of THQ may have diminished when bound to iron oxide, or the Fe²⁺/Fe³⁺-redox system may have prevented the one-electron transference and thus the formation of reactive THQ-radicals. Moreover, the redox potential for THQ is highly dependent on the pH of the system and even a slight acidification reduces the reducing power of THQ considerably [27,28].

The cytotoxic effects of THQ and the THQ-coated IONPs on the human cell line HT-29 were estimated by quantification of their metabolic activity (Figure 7a). While 3 mg/mL of coated IONPs reduced the relative metabolic activity to 65% and hence was considered to be cytotoxic, concentrations of c ≤ 2.0 mg/mL resulted in a relative metabolic activity of 105 ± 15% were therefore considered adequate for MHT. Moreover, concentrations up to 300 µg/mL THQ were not cytotoxic, because the cells maintained their metabolic activity at 100% (Figure 7a). According to the thermogravimetric results, THQ represented 6.5% of the weight of the THQ-coated IONPs, thus, 300 µg/mL THQ would be found for 4.6 mg/mL IONP-THQ. For MHT, only 2.0 mg/mL THQ-coated IONPs were used, and the cells were heated to 37, 39, 42, 45 or 48 °C for 15 min. Heating to just 39 °C reduced the relative metabolic activity to ~65%, while heating to even higher temperatures (42–48 °C) reduced the metabolic activity further to 32%–42% (Figure 7b). In contrast, with 2 mg/mL dopamine-coated IONPs, the metabolic activity was reduced to 25% only at 48 °C [15], while with catechol-coated IONPs, the metabolic activity was nearly eliminated at only 43 °C; however, 3 mg/mL was used in this case [16]. Thus, the THQ-coated IONPs are a promising candidate to kill the cancerous cells HT-29 in MHT treatments.

It has been shown that THQ induces ROS-production in vitro, which ultimately triggers apoptosis in HL60 cells [13]. Therefore, and for the first time, it was evaluated whether THQ-coated IONPs have a similar effect. The negative control exhibited almost no fluorescence, indicating the background ROS-production in healthy HT-29 cells (Figure 8a). In contrast, in the positive control, most cells had bright fluorescence, because the added DCFH₂ was oxidised by the added H₂O₂ (Figure 8b). THQ itself did not induce ROS-production, as can be seen in the very low fluorescence in Figure 8c, while THQ spiked with H₂O₂ exhibited a bright fluorescence and can be detected in nearly all cells (Figure 8d), although the average fluorescence might be less than in the positive control (Figure 8b). The THQ-coated IONPs did not oxidise DCFH₂ readily, either directly or indirectly by ROS-formation; thus, almost no fluorescence can be observed (Figure 8e), as with pure THQ (Figure 8c). On the other side, when the THQ-coated IONPs were spiked with H₂O₂, some fluorescence could be detected in only a few cells (Figure 8f), to a much lower degree than in the positive control (Figure 8b) and in THQ spiked with H₂O₂ (Figure 8d). This indicates that THQ bound to IONPs differed considerably in its redox activity from pure THQ, which also coincided with the reduced reactivity and lack of ROS-formation by H₂O₂-spiked IONP-THQ in the reduction or degradation of MB (Figure 6).

Finally,

Table 3. The main physicochemical characteristics of IONPs coated with different phenol-derived compounds used in MHT.

Coating	σ (nm)	Ms (emu/g)	p (W/cm3)	Ros Production
Catechol [16]	14.7	34	41	Yes
Dopamine [15]	15	24	110	-
Tetrahydroxyquinone	13.5	66	140	No

summarises the main characteristics of the THQ-coated IONPs in order to compare them with other IONPs coated with other phenol-derived compounds, previously reported and used in MHT.

4. Conclusions

THQ is a good alternative for coating iron oxide nanoparticles to perform in vitro experiments for cancer treatment by magnetic hyperthermia. Although the THQ coat was approximately 50% thinner than the phenol-derived coat used for other IONPs, they exhibited similar colloidal stability (ζ) and even better magnetic properties, such as a higher magnetic saturation (M_s), a lower blocking temperature (T_b) and a higher specific absorption rate (SAR). Indeed, the M_s of THQ-coated IONPs was up to 2.7 times higher than that observed when the magnetite was coated with catechol or dopamine; additionally, in colloidal suspension, their power-absorption capability (p) was approximately 22% greater than magnetite–dopamine and more than two times greater than magnetite–catechol. The magnitude of p is the most important parameter in destroying cancer cells via MHT. Additionally, it could be demonstrated that in vitro, the THQ-coated IONPs were not cytotoxic up to concentrations of 2 mg/mL. Thus, it would be interesting to conduct further experiments and procedures in order to count the number of THQ-coated IONPs promoting the temperature rise only from inside or attached to the surface of the cells. Moreover, the ROS formation by THQ-coated IONPs was considerably inhibited, which would reduce negative side effects of oxidative stress during cancer treatment in non-affected areas. Thus, considering its good physicochemical properties and low cytotoxicity, this ferrofluid is a promising candidate for the initiation of in vivo experiments for tumour eradication by MHT in murine models.