In vitro studies on curcumin-loaded multiwalled carbon nanotubes antioxidant activities and cytotoxicity against Hep G-2 liver cancer cell lines

Oxidation of MWCNTs

To oxidize MWCNTs, a mixture of sulfuric acid and nitric acid in a volume ratio of 1:3 was used. 0.2 g of MWCNTs were dispersed in this acid mixture and then refluxed for 24 hours at 70°C and 350 rpm. The resulting suspension was diluted with distilled water to a volume of 1,000 ml, filtered through a 0.45 μm cellulose membrane, and repeatedly washed to remove any remaining acid. The acid-free MWCNTs were then exfoliated from the membrane and dried in an oven at 70°C for 72 hours. [28,29].

Functionalization of MWCNT by chitosan

The chitosan-MWCNT composite was prepared using two methods: a solution processing method by mixing a dispersion of oxidized MWCNT with chitosan solution according to Sobh et al. [1], and by chemical functionalization by oxalyl chloride under nitrogen gas according to Bozeya et al. [28]. The drug loading was achieved by combining the required amount of drug (curcumin) with ethanolic dispersions of both chitosan-MWCNT and oxidized-MWCNT at room temperature, each in dark conditions and in light conditions.

MWCNT functionalization by solution processing method

Low molecular weight chitosan (5 mg/ml) was dissolved in 1% (v/v) glacial acetic acid, mixed with 2.5 g/ml of oxidized MWCNTs, sonicated for 20 minutes, stirred for 48 hours. at room temperature, then 1% (w/v) NaOH was used to collect chitosan bounded with MWCNTs (Chitosan-MWCNT), washed several times with deionized water, centrifuged at 4,000 rpm and 4°C to remove any unbounded chitosan, and then allowed to dry at room temperature [1].

Chemical functionalization of MWCNT by oxalyl chloride

A mixture of 120 mg of previously prepared oxidized MWCNT and 80 ml of DMF was sonicated for 20 minutes, then 1 ml oxalyl chloride was added drop-wise to the suspension under nitrogen gas. This mixture was stirred at 350 rpm in an ice bath for 2 hours, then it was stirred at room temperature for another 2 hours, followed by overnight stirring at 70°C to remove excess oxalyl chloride.

After that, a mixture composed of 490 mg of low molecular weight chitosan and 40 ml of DMF was sonicated for 20 minutes, then added to the previous mixture (MWCNT, DMF, and oxalyl chloride) and stirred for 48 hours at 110°C. The resulting mixture was then filtrated and washed until no chitosan residues remained in the filtrate; finally, the particles were vacuum dried for 4 hours [28].

Loading of curcumin into MWCNTs

Forty milligrams of oxidized MWCNT, chitosan-MWCNT, and chitosan-MWCNT-oxalyl-Cl each were solely dispersed in 20 ml of absolute ethanol, sonicated for 30 minutes. This step was repeated for two different batches each, as it was applied under dark and light conditions. Forty milligrams of curcumin powder were weighed in 6 tubes, each one was dispersed in 20 ml of absolute ethanol, and sonicated for 30 minutes. Then, the two solutions (curcumin and MWCNT) were mixed, sonicated for 30 minutes, and stirred at 350 rpm for at least 16 hours on either dark or light to load the curcumin on the surface of MWCNTs (Table 1).

Each solution was centrifuged at 4°C and 3,000 rpm for 10 minutes and the supernatant was used to measure the loading efficiency with the aid of a spectrophotometer at 428 nm, whereas the resulting combinations were listed in Table 1. Samples were washed with deionized water and dried at room temperature for further use [1].

Determination of curcumin loading efficiency onto MWCNT

The supernatant resulting from the previous step was diluted to 1:1,000 and the absorbance of the diluted sample was measured at 428 nm, using a spectrophotometer. The loading efficiency was calculated according to Equation (1) [30].

$\mathrm{Loading}\mathrm{efficiency}=\frac{\mathrm{Ct}\mathrm{-}\mathrm{Cs}}{\mathrm{Ct}}\times 100\%(1)$

where Ct is the total amount of curcumin used and Cs is the amount of curcumin in the supernatant.

Determination of curcumin entrapment efficiency onto MWCNT

To measure the entrapment efficiency of curcumin on MWCNTs, ethanolic dispersions with a concentration of 1 mg/ml were prepared for each treatment previously listed in Table 1, each dispersion was prepared, sonicated for 20 minutes, and then centrifuged in 4°C at 4,000 rpm for 30 minutes. The absorbance of curcumin in the supernatant (free drug) was measured by spectrophotometer at 428 nm according to Figure 1 which shows the curcumin spectrum. Then, the entrapment efficiency of curcumin in each drug was measured using Equation 2 [30].

$\mathrm{Entrapment}\mathrm{efficiency}=\frac{\mathrm{WCi}\mathrm{-}\mathrm{WCf}}{\mathrm{WCi}-\mathrm{WCf}}\times 100\%(2)$

where WC_i is the initial weight of curcumin and WC_f is the weight of free curcumin.

Fourier transform-infrared spectroscopy (FTIR)

FTIR was performed to check for oxidation using a Thermo Nicolet NEXUS 670 (Gaithersburg, MD, USA), using 32 scans over the range 4,000–400 cm⁻¹ with 4.0 cm⁻¹ resolution. Samples were characterized using KBr [28].

Scanning electronic microscope

Sample morphology and the examination of nanotube alignment were investigated using Scanning electron microscope (SEM). The analysis was performed using SEM QUANTA FEG 450 instruments. Samples were placed on aluminum stubs, sputtered with gold (5 nm thickness), and analyzed with appropriate magnification using Inspect™ F50 apparatus.

Thermogravimetric analysis (TGA)

To confirm the attachment of biomolecules to the surface of CNTs A METTLER TOLEDO TGA/SDTA851^e was used to determine the thermal stability of pristine MWCNT, oxidized-MWCNT, chitosan, curcumin, chitosan-MWCNT, curcumin-chitosan-MWCNT, and curcumin-MWCNTs by heating (2–3) mg of each sample from 25°C to 1,000°C at a rate of 10°C/minute in a nitrogen atmosphere [31].

Brunauer–emmett–teller (BET) surface area analysis

The surface area of samples was measured by Quantachrome Nova 2200e series device. Before analysis, samples of oxidized-MWCNT, curcumin-chitosan-MWCNT, and curcumin-MWCNTs were degassed at 105°C for 20 hours. Measurements of surface area and pore size distributions of the degassed samples were determined by using pure nitrogen gas as adsorbed in liquid nitrogen (77 K) [29].

Table 1. Loading conditions of curcumin on either MWCNT alone or chitosan-MWCNT. [Click here to view]

Figure 1. Curcumin spectrum using UV-Vis spectrophotometer. [Click here to view]

Antioxidant activity

Determination of antioxidant activity using the DPPH radical scavenging assay. The antioxidant activity of the oxidized-MWCNT, chitosan, curcumin, chitosan-MWCNT, curcumin-chitosan-MWCNT, and curcumin-MWCNT against DPPH was determined using the previously described method [32–34]. A methanolic-DPPH solution with a concentration of 0.6 mM was mixed with 8 dilutions of each sample starting from 1,000 until 7.8 μg/ml as follows: 120 μl sample was added to 240 μl methanol along with 120 μl of 0.6 mM DPPH solution so the concentrations were divided by 4 to reach the final concentration loaded in the 96 well plates. Three replicates per sample and concentration were used. The mixture was kept in the dark for 30 minutes at room temperature and the absorbance was measured at 517 nm using an ELIZA plate reader. The methanolic dilution of DPPH was used as a negative control, while the Ascorbic acid was used as a positive control. DPPH radical scavenging activity of each sample and concentration was measured by Equation 3:

$\mathrm{DPPH}\mathrm{scavenging}\%=\frac{\mathrm{Absorbance}\mathrm{of}\mathrm{control}\mathrm{-}\mathrm{Absorbance}\mathrm{of}\mathrm{sample}}{\mathrm{Absorbance}\mathrm{of}\mathrm{control}}\times 100\%(3)$

The reducing ability of an antioxidant compound was quantitatively determined by decolorizing the violet color of the DPPH solution which was recorded as absorbance change at 517 nm using an ELIZA plate reader. This feature presets to reduce the DPPH in solution by an antioxidant substance as it gives an electron or hydrogen, causing the DPPH lighter in color, measured in terms of its absorbance. The higher the antioxidant concentration the lesser the absorbance of reagent.

Determination of antioxidant activity using the ABTS⁺ radical scavenging assay. The antioxidant assay of oxidized-MWCNT, chitosan, curcumin, chitosan-MWCNT, curcumin-chitosan-MWCNT, and curcumin-MWCNT was determined by using ABTS⁺. Radical ABTS⁺ was prepared through the oxidation of ABTS⁺ by potassium persulfate (K₂S₂O₈).

A mixture (1:1 v/v) of ABTS⁺ (7 mM) and potassium persulfate (2.45 mM) was prepared and kept in the dark for 16 hours at room temperature. 16 Serial dilutions were made for all previous 6 combinations starting from 10,000 ppm. 600 μl from various concentrations of oxidized-MWCNT, chitosan, curcumin, chitosan-MWCNT, curcumin-chitosan-MWCNT, and curcumin-MWCNT solutions was mixed with 60 μl ABTS⁺ solution, and absorbance was measured at 734 nm against a blank (PBS) [35]

Each concentration was analyzed three times. The average radical scavenging assay of samples was calculated through Equation 4:

$\mathrm{RS}\%=\frac{\mathrm{Abs}\mathrm{control}\mathrm{-}\mathrm{Abs}\mathrm{sample}}{\mathrm{Abs}\mathrm{control}\mathrm{-}\mathrm{Abs}\mathrm{sample}}\times 100\%(4)$

where Abs control is the absorbance of control and Abs sample is the absorbance of the sample.

ABTS⁺ is considered the best free radical reagent to be used to measure the electron/hydrogen donating potential of natural substances and is also an appropriate substrate for enzymes like peroxidases [36].

Anti-cancer activity

Cytotoxicity test. The effects of oxidized MWCNT, chitosan, curcumin, chitosan-MWCNT, curcumin-chitosan-MWCNT, and curcumin-MWCNT on the viability of fibroblast cells (control), and HepG-2 (liver cancer cell line), were evaluated using viability/cytotoxicity (MTT colorimetric assay) kit that analyzes the integrity of mitochondrial function by producing formazan crystals, which are proportionally related to the cell viability as the higher crystal production, exhibit higher cell viability [37,38]. Each cell line was cultured in 96-well plates in its appropriate media as follows: (fibroblast in Iscove’s media, while HepG-1 in EMEM media) which were supplemented with 10% (v/v) fetal bovine serum and antibiotics (penicillin-streptomycin). The cells were cultured at 37°C in a humidified atmosphere of 5% CO₂/95% air. When the cells reached 80% confluency, they were trypsinized, counted, then calculated to be 8,000 cells /well, and plated in 96-well culture plates, where they were incubated overnight. Stock solutions (1 mg/ml) of oxidized-MWCNT, chitosan, curcumin, chitosan-MWCNT, curcumin-chitosan-MWCNT, and curcumin-MWCNTs were prepared by dissolving the formulations in DMSO. Afterward, it was diluted in the appropriate media to different concentrations (400, 200, 100, 50, 25, and 12.5 μg/ml), the cells were treated with a 100 μl aliquot of each concentration for the next 72 hours. At the end of this incubation period, the media in each well was replaced by 100 μl of RPMI and 10 μl of MTT solution. After 3 hours 100 μl of the solubilization stop solution was added to each well. Then, after at least 1 hour the absorbance was measured using an ELIZA microplate reader at 570 nm. Results were quantified as relative values to that of the negative control, where the negative control (untreated cells) was set to 100% viability. All experiments were performed in triplicate.

From the absorbance data, the IC50 (concentration that induces half the maximum effect) was calculated using graph pad prism software. The percentage of cytotoxicity was measured according to Equation 5:

$\mathrm{Cell}\mathrm{viability}\mathrm{\%}=\frac{\mathrm{Absorbance}\mathrm{of}\mathrm{treated}\mathrm{cells}}{\mathrm{Absorbance}\mathrm{of}\mathrm{untreated}\mathrm{cells}}\times 100\%(5)$

Fourier transform-infrared spectroscopy

FT-IR spectra were used to assess the chemical structure and the exposed functional group on the surface of pristine MWCNT, the oxidized-MWCNT, chitosan, curcumin, chitosan-MWCNT prepared in both solution processing method and chemical functionalization using oxalyl-chloride, curcumin-chitosan-MWCNT, and curcumin-MWCNT using 32 scans over the range 4,000–400 cm⁻¹ with 4.0 cm⁻¹ resolution. Samples were characterized using KBr (Fig. 2). FT-IR spectra of pristine MWCNTs represented six distinct peaks for the functional groups on the pristine-MWCNT surface which were O–H, C–H, O=C=O, C=C, C-N stretching, and C-H bending. The absorption bands at 3,543, 3,466, and 3,417 cm⁻¹ refer to O-H stretching. The presence of the hydroxyl group could be due to the surrounding atmospheric moisture or to manufacturing methods that can introduce the hydroxyl group through the purification process [11], while the absorption band at 2,918 cm⁻¹ is associated with C-H stretching, which agreed with Abdolmaleki et al. [45] who get a peak on 2,922 cm⁻¹ for MWCNT defects. All peaks located between 2,840–3,000 cm⁻¹ are attributed to C-H stretching, whereas the absorption bands at 2,360, 1,618, and 1,381 cm⁻¹ are associated with O=C=O stretching, C=C stretching, and C-H bending, respectively, as shown in Figure 2.

Figure 2. The FTIR spectra of (a) pristine-MWCNTs and oxidized-MWCNTs, (b) chitosan and chitosan-MWCNT, and (c) curcumin, curcumin-MWCNT, and curcumin-chitosan-MWCNT. [Click here to view]

Table 2. Curcumin loading and entrapment efficiency results. [Click here to view]

For the oxidized-MWCNT FT-IR spectra (Fig. 2), the peaks at 3,548, 3,473, 3,413, and 3,234 cm⁻¹ correspond to O–H stretching, while the peaks at 2,918 and 2,852 cm⁻¹ were related to C-H stretching. The stretch peak located at 2,362 and 2,852 cm⁻¹ correspond to O=C=O stretching. As the gas phase product of MWCNTs was validated to be CO₂, the peak located at 1,722 corresponds to the presence of C=O stretching which is considered due to oxidation forming the COOH groups on the surface of MWCNTs. The peak at 1,618 cm⁻¹ corresponds to C=C stretching which could be assigned as the stretching of the CNT backbone [45], the peak at 1,383 cm⁻¹ is related to C-H bending, at 1,132 a new peak related to C-O stretching of a secondary alcohol, and a peak of C-N stretching located at 1,020 cm⁻¹ is also observed. The FTIR spectra demonstrated that the pristine-MWCNTs had been oxidized to the point where they had alcohol and carboxylic acid groups on their surface before being functionalized with chitosan. The oxidation of reactive defects produced after exposing the pristine MWCNTs to a mixture of nitric acid and sulfuric acid may result in the formation of these functionalities [46].

The bands located between 3,570 and 3,200 cm⁻¹ demonstrated the presence of a hydroxyl group, while the peaks between 2,935 and 2,840 cm⁻¹ were attributed to methylene C-H stretching. The peaks between 2,360 and 2,335 cm⁻¹ in FT-IR spectra of pristine-MWCNT, and oxidized MWCNT, were corresponding to O=C=O stretching, since CO₂ is considered as the gas phase product of MWCNT; therefore, CO₂ is the product of pristine and modified MWCNTs after oxidation while the presence of vibrational peaks at CO₂ location in the spectra of chitosan, solution-processed chitosan-MWCNT, curcumin, and curcumin-MWCNT appeared due to environmental factors since the samples should not contain CO₂ [47,48]. The emerged peak at 2,079 cm⁻¹ in the FT-IR spectra of chemically functionalized chitosan-MWCNT attributed to the presence of N=C=S isothiocyanate. The peaks between 1,760 and 1,700 cm⁻¹ were attributed to C=O, while the bands located between 1,650–1,600 cm⁻¹ and 1,690–1,675 cm⁻¹ represented the presence of conjugated alkene C=C stretching for the MWCNT backbone [45,48,49]. An aldehyde C-H bending was demonstrated for the peaks located between 1,390 and 1,380 cm⁻¹, while between 680 and 610 cm⁻¹ an alkyne C-H bending appeared, in addition, a secondary amine C-N stretching peaks also existed at 1,090–1,020 cm⁻¹ [45,48,50].

The IR Spectra data clearly indicated the presence of both curcumin and chitosan on the oxidized MWCNT. It is noted that the main characteristic peaks for functional groups in both MWCNT and the produced combination appeared in the Spectra after loading. This is in agreement with Sobh et al. [1], who noted the presence of main peaks characteristic for functional groups in both of free drug and the produced chitosan-MWCNT loaded with curcumin, when observing the presence of a peak at 3,220 cm⁻¹ for O-H vibration of a carboxyl group and at 1,040 cm⁻¹ is the stretching vibration of C-O. Whereas when combining chitosan with MWCNT, a stretching vibration of N-H appeared at 3,480 cm⁻¹ while a C=O stretching was vibrated at 1,040 cm⁻¹ in the CONH, which confirms that the chitosan was grafted onto MWCNT via amide linkage which comes in accordance with our results for oxidized MWCNT at a wavenumber of 3,230 and 1,020 cm⁻¹ stretching peaks of –OH vibration of the carboxyl group and C-O stretching was appeared, respectively, while for curcumin-chitosan-MWCNT, a vibration peak for N–H stretching was noted at 3,477 and at 1,018 cm⁻¹ a vibration of C=O stretching in the CONH appeared, which confirms the amide linkage between the curcumin and the chitosan-MWCNT.

Scanning electron microscopy

There are two clear aspects to be observed when examining the SEM image of pristine MWCNT (Fig. 3); the entanglement and the severe agglomeration of pristine MWCNTs forming cotton-like agglomerates [51–53]. The SEM image of pristine MWCNT shows hollow straight tubes with different size distributions and those tubes form a bundle of tangled tubes with smooth surfaces and compact sidewalls.

After oxidation of pristine MWCNT with a mixture of strong acids, the SEM micrograph (Fig. 3) shows clear intact single tubular threads-like MWCNT with a wider average diameter, rougher surface structure, less entangled MWCNTs, and agglomerates. It also maintains the whole structure with no collapse after the oxidation [53,54]. The roughness on the surface of oxidized MWCNT is due to the formation of defect sites because of oxygenated functional groups’ attachment on their surface after acid treatment [55]. The severe structural damage also associated with the used-strong oxidizing acids which attack the existing active sites on the walls of CNTs may also be due to the mixture scission effect and high treating temperature during refluxing [55]. There were no obvious impurities traces appearing on oxidized MWCNT as the used pristine MWCNT was of 95% purity and acid treatment was responsible for the removal of any impurities traces [55,56]

SEM observation of oxidized-MWCNT functionalized with chitosan and loaded with curcumin showed less structural damage on its surface, as they were shortened, and their surface became more layered. The curvy MWCNTs in various size distributions with some aggregations shown in SEM micrographs that is probably due to the presence of intermolecular forces sticking the tubes together [52,53].

Thermal degradation analysis of functionalized CNTs

Generally, CNT thermal deterioration involves three stages, the first stage is moisture release, the second stage involves the structural deterioration of functional groups on the CNT surface, and the third one involves the oxidation of remaining carbon [57]. Figures 4 illustrates the TGA graphs of pristine-MWCNTs, oxidized-MWCNT, chitosan, chitosan-MWCNT, curcumin, curcumin-MWCNT, and curcumin-chitosan-MWCNT. At 703°C, Pristine-MWCNTs start to lose some of their weight due to the decomposition of the associated organic groups [58], while the oxidized-MWCNTs start to lose some of their weight when reaching 800°C. There was a progressive weight loss above 700°C and 800°C for Pristine-MWCNTs and oxidized-MWCNTs, respectively. This loss could be attributed to the removal of oxygen-containing groups such as hydroxyl groups that arose during the CNT production and oxidation processes [9,11]. The low weight loss amount may be attributed to the purity of the sample used [59]. With increasing temperature, the gradual small amount of weight loss for both pristine-MWCNT and oxidized MWCNT is probably due to the small amount of moisture absorbed by the adsorbent [60]. Chitosan lost approximately 5% of its weight between 75°C and 260°C, due to water evaporation, while the major weight loss (approximately 50%) occurred between 260°C and 340°C, which was attributed to the breaking of the glycosidic linkages during chitosan degradation [61]. When comparing the thermal behavior of chitosan with chitosan-MWCNTs; between 300°C and 800°C, chitosan-MWCNT started to lose approximately 5% of its weight due to water evaporation while when reaching 996°C approximately 10% of its weight was lost due to depolymerization of chitosan and decomposition of the carboxylic group (−COOH) of MWCNT releasing oxygen into the TGA chamber [58,62]. This confirms the successful grafting of MWCNT into chitosan via the solution processing method as the presence of MWCNT resulted in the increase of thermal stability of chitosan-MWCNT [1,9]. This result comes in accordance with Sobh et al. [1] who reported that chitosan was the least thermal stable sample during studying the thermal stability of chitosan, and chitosan-MWCNT before loading the curcumin, or 5-fluorouracil, where the thermal stability increased with the addition of MWCNT [63].

Figure 3. SEM of pristine-MWCNT (magnification scale bar 1 μm), oxidized MWCNT (500 nm scale bar), chitosan-MWCNT (500 nm scale bar), curcumin-chitosan-MWCNT(500 nm scale bar), and curcumin-MWCNT(500 nm scale bar). [Click here to view]

Regarding Curcumin thermal stability, it has been shown to be stable up to 70°C for 10 minutes, while above 70°C, curcumin began degradation, and at 100°C its decomposition rate increased [64]. Loading of curcumin on MWCNT or on chitosan-MWCNT enhances the curcumin thermal stability, while binding of curcumin with MWCNT is slightly less thermally stable than its binding with chitosan-MWCNT which led to a slight increase in thermal stability, and this agreed with Sobh et al. [1] who studied the thermal stability of curcumin when bounded to chitosan functionalized MWCNT and they found that the combination between chitosan-MWCNT and curcumin is thermally stable.

BET: surface area and pore size analysis

The surface area, total pore volume, and average pore diameter of oxidized MWCNT, curcumin-MWCNT, and curcumin-chitosan-MWCNT are given in Table 3. As seen in Table 3, the increase in BET surface area of oxidized-MWCNT compared to curcumin-MWCNT and curcumin-chitosan-MWCNT may be due to the removal of impurities during the oxidation process and moisture blocking the pores during the carboxylation process [29,39]. From a different angle, the inclusion of curcumin and chitosan into MWCNT can be attributed to the decrease in surface area of curcumin-chitosan-MWCNT in contrast to oxidized-MWCNT. The MWCNT’s accessible surface area is decreased as a result of this integration, which is consistent with research done in 2020 by Aghababaei et al. [65]. In comparison to pure MWCNT, the curcumin-chitosan-MWCNT, and curcumin-MWCNT composites are observed to have larger average pores. This finding shows that the MWCNT structure has been deformed as a result of the modification process. This discovery is in line with work done by Do?an et al. [39], who combined modified hexagonal-boron nitride nanoparticles with modified MWCNTs for hydrogen storage and found comparable outcomes.

Figure 4. TGA thermographs of chitosan, chitosan-MWCNT, curcumin, curcumin-MWCNT, and curcumin-chitosan-MWCNT. [Click here to view]

Table 3. BET surface areas, total pore volume, the average pore diameter of oxidized MWCNT, curcumin-MWCNT, and curcumin-chitosan-MWCNT. [Click here to view]