Mesoporous silica coated SPIONs containing curcumin and silymarin intended for breast cancer therapy

Abstract

Introduction

Super-paramagnetic iron oxide nanoparticles (SPIONs) are known as promising theranostic nano-drug carriers with magnetic resonance imaging (MRI) properties. Applying the herbaceous components with cytotoxic effects as cargos can suggest a new approach in the field of cancer-therapy. In this study mesoporous silica coated SPIONs (mSiO2@SPIONs) containing curcumin (CUR) and silymarin (SIL) were prepared and evaluated on breast cancer cell line, MCF-7.

Methods

Nanoparticles (NPs) were formulated by reverse microemulsion method and characterized by DLS, SEM and VSM. The in vitro drug release, cellular cytotoxicity, and MRI properties of NPs were determined as well. The cellular uptake of NPs by MCF-7 cells was investigated through LysoTracker Red staining using confocal microscopy.

Results

The MTT results showed that the IC50 of CUR + SIL loaded mSiO2@SPIONs was reduced about 50% in comparison with that of the free drug mixture. The NPs indicated proper MRI features and cellular uptake through endocytosis.

Conclusion

In conclusion the prepared formulation may offer a novel theranostic system for breast cancer researches.

Graphical abstract

Introduction

Today cancer is among the most serious health concerns worldwide. Numerous efforts intended for the early diagnosis and treatment of various cancers have been made during the last century. In this regard nanotechnology has conferred iconic features to drug delivery systems. In recent years, superparamagnetic iron oxide nanoparticles (SPIONs), a magnetic resonance imaging (MRI) contrast agent, has indicated promising potentials not only in diagnosis but also in drug delivery which demonstrates its capabilities as a theranostic system. Coating SPIONs with specific materials gives them the ability to carry therapeutic agents [1]. Moreover, due to the SPION magnetic features active tumor targeting is achievable by applying an external magnetic field. Among the various coatings mesoporous silica has indicated proper potentials for drug delivery. Mesoporous silica with its large surface area and pore volumes which can potentially host molecules, e.g. drugs, and not to mention its biodegradability and biocompatibility is an interesting candidate for SPION coating [2–5]. Previous studies have reported the improved bioavailability of drugs with poor water solubility through delivery by mesoporous silica coated SPIONs (mSiO2@SPIONs) [6–8]. Accordingly, in this study mSiO2@SPIONs was chosen as the drug carrier.

The combination of nanotechnology and herbal derivatives has suggested innovative approaches in drug delivery researches. Curcumin (CUR), the most effective compound in Turmeric, cause of the yellow color of the plant [9], is well known for its role in human health as an antioxidant, anti-inflammatory and anticancer [10]. The blockage of NF-κB signaling, the factor which controls DNA transcription, production of cytokine and cell survival, is reported as the main mechanism of CUR cytotoxicity [11, 12]. The down-regulation of AP-1, COX-1, and COX-2 and some other factors, as well as the up-regulation of tissue inhibitor of metalloproteinase-1 (TIMP-1), and p21 and p27 are also among the proposed anticancer mechanisms of CUR [13]. Another feature that makes CUR a desirable anti-cancer agent is its safety and negligible side effects in comparison with common chemotherapeutic agents [8, 14]. Silymarin (SIL), extracted from Silybum Marinum, is a mixture of flavonolignans including silibinin as the major component [15]. SIL is mainly used in hepatic ailment however, has indicated anti-proliferative potentials, as well. Various studies have shown the SIL cytotoxic effect on breast cancerous cell line, MCF-7 [16, 17]. Various mechanisms have suggested explaining the anti-cancer effects of SIL and its derivatives including: interfering with P-glycoproteins [18], down-regulating of HER2/neu expression [19], and Bcl-xl [20], and up-regulating of Bak, p53 and p21 [20], impairing the ROS-dependent mitochondrial function [21].

Since both CUR and SIL are hydrophobic compounds with proved anticancer potentials and designing a formulation for co-delivery of two or more hydrophobic drugs has been a challenge so far in this study mSiO2@SPIONs containing CUR and SIL, as anticancer agents, were prepared for the first time (to the best of our knowledge) to enhance the drug water solubility and cytotoxic effects. Following the NP characterizations and in vitro drug release, the toxicity and cellular uptake of the prepared nano-carriers were investigated on the MCF-7 cell line. MRI was also employed to investigate the contrast effect of NPs on cells treated with drug loaded mSio2@SPIONs.

Materials and method

Materials

FeCl₃. 6H₂O, toluene, Hydrochloric acid 37%, ammonia, ethyl acetate, ammonium nitrate, disodium hydrogen phosphate, potassium chloride, potassium dihydrogen phosphate, polysorbate (tween 80), MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide), sodium chloride were supplied from Merck, Germany. FeCl₂. 4H₂O, 1-Butanole, ammonium hydroxide, tetraethyl orthosilicate (TEOS) were from Sigma-Aldrich, USA. cetyltrimethylammonium bromide (CTAB) was from Calbiochem, USA-Canada and dimethyl sulfoxide (DMSO) was from Dae Ju, Korea. CUR and SIL were respectively supplied from Serva, USA and Qingdo BNP, China. All the materials including solvents were provided in analytical grades.

SPION synthesis

The reverse microemulsion method was employed to produce SPIONs in which nanodroplets of aqueous phase were dispersed in an oil phase through surfactant molecules surrounding these nanoreactors, makes them turn into SPION producing units [22, 23]. In this regard a two-step procedure was performed.

1. two microemulsions, No.1 and 2, were prepared. To prepare No.1 microemulsion FeCl₃.6H₂O (0.45 g) and FeCl₂.4H₂O (0.15 g) were solubilized in 2 ml of HCl [2 N] and added to 29 ml CTAB solution in toluene (6%) under stirring and an opaque reddish mixture was obtained. Microemulsion No.2 was prepared by slow addition of 2 ml ammonium hydroxide as a hydrolyzing agent to the CTAB solution in toluene (6%). The final mixture was white and coagulated. Each microemulsion was then tittered by approximately 1.5 ml of 1-butanol till either transparency or color change was achieved.

2. following the preparation of two microemulsions in step 1, they were mixed and stirred (1200 rpm) for an hour in a three-neck flask under continuous flow of nitrogen gas at 50 ${}^{\circ }\mathrm{C}$. The black-brown color of the final product indicated the ${\mathit{Fe}}_3{O}_4$ nanopaticle formation. Then 40 ml of ethanol was added and the mixture was homogenized for 4 extra minutes. Ultimately, the aqueous phase containing nanoparticles (NPs) was separated from the oil phase using a strong neodymium magnet. To eliminate extra CTAB and ions, the prepared NPs were washed four times with ethanol and acetone, respectively.

The collected NPs were dispersed in 60 ml deionized water and stored at 4 °C for further use.

Coating of SPION by silica

To prepare mesoporous silica coated SPIONs, CTAB and tetra ethyl ortho silicate (TEOS) as a silicate precursor, were employed.

The procedure was started by incorporation of 1 ml SPION nanosuspension (0.3% in DI water) into 10 ml of CTAB solution in DI water and kept under stirring for 3 h (hrs.) (1200 rpm). In order to evaluate the effect of surfactant amount on the size and polydispersity index (PDI) of silica coated SPIONs, 3 quantities of CTAB (100 mg, 50 mg, and 25 mg) were studied.

Ammonia (85 μl), as a catalyzer, and ethyl acetate (250 μl) [24, 25] were added afterwards. By drop wise addition of TEOS (25 μl) the hydrolysis and condensation procedure [26, 27] was started which lead to Si–O-Si groups formation that finally resulted in silica nanostructures. By removal of CTAB from the NP pores were generated inside the silica NPs. In brief NPs were first washed with distilled water and acetone twice and then 10 ml ammonium nitrate solution (13.5 mg/ml in ethanol) was added and the mixture was heated to 60 ${}^{\circ }\mathrm{C}$ and stirred overnight under reflux. An ion exchange procedure via ammonium nitrate leads to CTAB elimination and generation of nanocavities [28–31].

The mSio2@SPIONs were collected by centrifuging, dispersing in distilled water and finally freeze dried.

Loading the drugs

For incorporation of cargo molecules into the mSio2@SPIONs, 2 mg of NPs were suspended in drug solutions in DMSO (4 ml) and the mixture was stirred for 24 h at room temperature.

Different concentrations of CUR and SIL were investigated, 0.15, 0.25, 0.5 and 1.25 mg/ml. Drug Loading (DL) and entrapment efficiency (EE) [5], were indirectly assessed using dialysis sac (MWCO = 12KD, Sigma Aldrich, USA) in DMSO (50 ml), 15 min [32]. The free passage of both drugs through dialysis sac was evaluated during the pre-studies and no significant interference was observed. The un-entrapped drug was analyzed using ultraviolet spectrophotometer (UV, CE7500, Cecil, Cambridge). The calibration curve of CUR and SIL were plotted at 435.5 and 291 nm, respectively. DL% and EE% were calculated by formulas below:

$$\begin{array}{c}\mathrm{DL}\%=\frac{totaldrugwt-unentrappedteddrugwt}{\mathit{nanoparticlewt}}\times 100\\ EE\%=\frac{totaldrugwt-unentrappeteddrugwt}{\mathit{totaldrugwt}}\times 100\end{array}$$

Characterization of NPs

Size and zeta potential analysis

Hydrodynamic size and size distribution of NPs, poly-dispersity index (PDI), were measured by dynamic light scattering (DLS) method using Zetasizer Nano-ZS; (Malvern Instruments, UK). All measurements were performed at 25 °C with an angle detection of 90°. The zeta potential of freshly prepared NPs was determined by laser doppler electrophoresis method using the same apparatuses.

Morphology

Scanning electron microscopy (SEM, KYKY-EM 3200, China), operating at 26 kV was used for the morphology analysis. A small droplet of nano-suspension was gently located on a glass lamella and dried at room temperature before the scan.

In vitro drug release

In order to survey the release pattern of the prepared nano-formulations the formerly tested method by Milane et al. was applied [33]. An appropriate weight of CUR + SIL loaded mSiO₂@SPIONs equivalent to 1 mg of CUR and relatively 1 mg of SIL (based on their LE% values) was dispersed in 20 ml of PBS buffer containing 2% tween 80 at two pH 5.5 and 7.4 at 37 °C. In time intervals (0.5, 1, 2, 4, and 8 h) 0.8 ml samples were taken to analyze the drug concentrations. The taken volume was replaced by the same amount of PBS with similar pH and surfactant concentration to preserving the constant volume and sink condition [34–36].

Cellular study

Cell cytotoxicity investigation

The cytotoxicity of the NPs was assessed on MCF-7 breast cancer cell line via 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide (MTT) assay. MCF-7 cells were seeded in 96 well plate at a density of 15,000 cells/well in DMEM culture medium supplemented with 10% FBS and 10% penicillin–streptomycin. First to determine the IC50 of each drug and their mixture (1:19 w/w), the cells were divided into three groups treated with fresh medium containing samples of CUR, SIL, and the mixture of the two drugs (CUR + SIL) at concentrations of 6.25, 12.5, 50, and 100 µM and incubated for 24 and 48 h. at 37 °C. The mixture of CUR and SIL contains an equal amount of both drugs at the mentioned concentration. DMSO (0.5% v/v) was used to make soluble drugs in the culture medium. Previous studies have proven the safety of this concentration of DMSO on the cell line [37]. To evaluate the cytotoxicity of the prepared NPs during the pre-studies first the survival of the cells in presence of unloaded NPs (mSiO2@SPIONs) was investigated. Then the cells were treated with determined amount of CUR + SIL loaded mSiO2@SPIONs equal to the concentrations of 6.25, 12.5, 50, 100 µM for each drugs and incubated for 24 and 48 h. at 37 °C. In both tests after the incubation time, the medium was removed from each well and replaced with 50 µL of MTT solution (5 mg/mL in PBS). The plates were subsequently incubated for another 4 h. at 37 °C. Thereafter, 100 µL of DMSO was added to each well and the plate was shaken to make formazan dissolve. The intensity of the solubilized dye in each well was measured by the microplate ELISA reader (EL_800, BioTek, USA). The cytotoxicity of samples was measured by comparing the cell viability of each sample with the control cells which were treated with a drug-free culture medium.

Cellular uptake study

In order to investigate the cellular uptake of CUR + SIL loaded mSiO2@SPIONs MCF-7 cells were seeded on sterile glass slides in a 6 well plate. The cells were first treated with 1% w/v gelatin for better cell attachment and then with the samples consist of drug loaded mSiO2@SPIONs and the mixture of free drugs as control. Each sample was added at its IC50 concentration obtained from MTT assay. After a 4-h incubation the cells were washed with PBS (× 3), stained with Lyso Tracker Red (300 nM) and incubated for more 60 min to label the lysosomes. Subsequently, the cells were fixed with formaldehyde (4% v/v) and, exposed to DAPI for nuclei visualization. Cell uptake was evaluated via Nikon confocal microscope A1 (Nikon Inc., Switzerland) supplied with A1 scan head and detectors of 405 nm diode laser with DAPI filter and 543 nm diode laser with TRITC filter (Melles Griot, USA).

VSM analysis

Vibrating sample magnetometer (VSM) (model 155, PAR) was used to investigate the magnetic properties of the samples at room temperature.

In-vitro MRI analysis

The MCF-7 cells treated with the prepared NPs and Endorem,a commercial contrast (Guerbet, France), were scanned by a 3 Tesla MRI scanner (MAGNETOM Avanto – Siemens, Germany) at room temperature [38]. The equal Fe concentration of the samples was 10, 20, and 30 μg/ml based on the prepared SPION. The MRI T2 signal intensities were measured. T2 or T2*-weighted images were acquired under the fast spin-echo (TSE) or gradient-recalled echo (FFE) parameters. The results were compare with a commercial contrast agent, Endorm®, Guerbet, France.

Statistical analysis

Statistical software IBM SPSS Statistics Ver.20 was used to analyze the data. One-way analysis of variance (ANOVA) and t-test were employed and P < 0.05 was determined as significance. All the tests were carried out three times.

Results

NP characterizations

Size and Zeta potential

In this research NP sizes, PDIs and zeta potentials were evaluated in each step of preparation to determine any probable change.

According to the DLS results the prepared SPIONs showed an average hydrodynamic diameter of 25.50 ± 6.13 nm. After the coating procedure the size was elevated to 57.00 ± 11.11 nm (P < 0.05). The final CTAB removal made no significant shift in hydrodynamic size, 65.80 ± 15.20 nm (P > 0.05). A significant reduction was observed in PDI values from SPION, coated SPION, and CTAB removed ones (Fig. 1).

Fig. 1.

Size and PDI of NPs at different stages of NP synthesis procedure

Zeta potentials were also measured at each step. Bare SPION indicated a zeta potential of -11.25 ± 0.21 mV. With silica covering the zeta potential significantly grew to positive values (+ 42.65 ± 1.34 mV) and by excess CTAB elimination, remarkable negative values were obtained (-24.55 ± 5.86 mV). Incorporation of drugs showed negligible effect on the current negative values, -21.64 ± 4.33 mV (P > 0.05).

As can be seen in Fig. 2 following the coating process the sizes of NPs were increased in a CTAB concentration dependent manner. By the CTAB reduction from 100 to 25 mg, the size fell about 40 nm, (P < 0.05) (Fig. 2). In addition to size, PDI changes were surveyed as a function of the CTAB amount. As indicated in Fig. 2 the CTAB quantity increase led to slight PDI value growth, which was not statistically significant. Finally, 25 mg CTAB was selected as the surfactant dosage of choice to achieve an appropriate size and a monodispersed nanosuspension.

Fig. 2.

The impact of CTAB concentration on the size and PDI of silica coated NPs

Incorporation of the drugs into the NPs did not change the size and PDI significantly (P > 0.05); size: 69.32 ± 7.11 nm and PDI: 0.31 ± 0.05.

SEM images

Figure 3 depicts SEM images of the prepared NPs. As can be seen, NPs were spherical in shape and relatively mono-dispersed, and suggesting the size in the range of 60 to 80 nm which confirmed by DLS.

Fig. 3.

SEM image of $m{\mathit{SiO}}_4@SPION$ (left) and CUR + SIL loaded $m{\mathit{SiO}}_4@SPION$ (right)

Drug loading

According to the results CUR DL% in final NPs was increased by its concentration enhancement from 0.15 to 0.5 mg/ml, however, increasing from 0.5 to 1.25 showed the insignificant effect on its DL% (P > 0.05). In contrast, there were no predictable correlations between SIL concentrations and its DL% in final NPs. The maximum DL% of CUR was obtained at 0.5 and 1.25 mg/ml concentrations of drugs (1:1), while the maximum DL% of SIL was achieved at 1.25 mg/ml.

The highest EE%, more than 90%, of both therapeutic agents was attained at 0.25 mg/ml (Table1).

Table 1.

The average values of DL% and EE% in drug loaded NPs in different concentrations of CUR and SIL (mean ± SD*, n = 3)

CUR & SIL Concentration (mg/ml)	CUR DL (%)	SIL DL (%)	CUR EE (%)	SIL EE (%)
0.15	1.45 ± 0.15	0.02 ± 0.01	10 ± 1.59	0.2 ± 0.06
0.25	15.07 ± 1.63	16.02 ± 2.21	94.1 ± 9.53	98.62 ± 13.61
0.5	24.76 ± 3.65	8.47 ± 3.49	73.2 ± 10.89	27.62 ± 12.28
1.25	23.04 ± 1.85	16.61 ± 1.90	27.36 ± 2.20	± 19.01 ± 2.25

In vitro drug release

The amounts of released CUR and SIL from CUR + SIL loaded mSiO₂@SPIONs versus time in both pH of 5.5 and 7.4 are shown in Fig. 4.

Fig. 4.

The release profile of CUR and SIL from the NPs during 8 h.”a” and “b” represent the release profile of SIL and CUR at pH 5.5 and “c” and “d” show that at pH 7.4

Comparison of the CUR/SIL loaded mSiO2@SPION release profiles in pH 5.5 (Fig. 4b) and 7.4 buffers (Fig. 4d), revealed more amount of released CUR in acidic condition. In pH 5.5 almost 41% of CUR was released from the NPs within 8 h while the maximum amount of released CUR was about 20% in pH 7.4 and this pH dependent change in the released amount was statistically significant (P < 0.05). The released SIL reached to maximum concentration of 75% after 8 h and in spite of CUR, pH did not affect its release profile significantly (P > 0.05) (Fig. 4a, c). Since changes in pH condition did not considerably impact on the release behavior of SIL, the increase observed in CUR release can be related to the enhancement of CUR solubility rather than the NP structure. The current results were in in accordance with previous reports [39].

Cellular study

Cell cytotoxicity findings

According to the pre-studies no significant cytotoxicity was observed by unloaded NPs and more than 85% cell survivals was obtained in MCF-7 cells treated with mSiO₂@SPIONs. The results obtained from MTT assay after 24 and 48 h. Treating with free drugs are shown in Fig. 5. According to the results CUR (red lines) showed higher cytotoxicity on MCF-7 cells in comparison with SIL (yellow lines) in all test concentrations and times. As can be seen in Table 2 IC50 was lower in cells treated with SIL and CUR mixture (blue lines) compared to the cells exposed to each drug alone.

Fig. 5.

a) 24 h b) 48 h MTT result of SIL (yellow), CUR (red), and CUR and SIL mixture (blue) at different concentration

Table 2.

Mean IC50 values (µM) for CUR/SIL loaded mSiO₂@SPIONs, CUR and SIL mixture, CUR, and SIL (mean ± SD, n = 3)

	CUR/SIL loaded mSiO2@SPIONs	CUR & SIL mixture	CUR	SIL
After 24 h	5.28 ± 0.38	11.03 ± 0.17	33.13 ± 6.55	162.35 ± 1.27
After 48 h	5.55 ± 0.23	10.67 ± 0.92	26.52 ± 1.66	147.29 ± 11.49

As can be seen in Fig. 5 the viability in the cells treated with CUR + SIL was significantly lower in 6.25 and 12.5 µM concentrations compared with each drug alone (P < 0.05). However, in higher concentrations the cytotoxicity of the mixture was comparable with free CUR (P > 0.05).

As can be seen in Fig. 6, NPs containing both drugs (green columns) demonstrated remarkable more cytotoxicity in comparison with the mixture of drugs (blue columns) in all test concentrations and this reductions in cell viability were all statistically significant (P < 0.05).

Fig. 6.

MTT results of SIL (yellow column), CUR (red column), CUR and SIL mixture (blue column) and CUR/SIL loaded mSiO₂@SPIONs (green column) at different concentration on MCF-7 at a) 24 h b) 48 h (n = 3)

In Table 2 the IC50 of the drug loaded NPs, free mixture, CUR, and SIL are presented. IC50 of NPs was reduced 48 and 52% after 24 and 48 h., respectively in comparison with the mixture of the free components (P < 0.05).

Cellular uptake study

The uptake of the NPs by MCF-7 cells was studied using confocal microscopy. Lysotracker red is an acidophilic fluorophore that is used as a probe for acidic organelles such as lysosomes and endosomes in the cytoplasm. The common root of NP uptakes is through endocytosis. The remarkable color intensity of fluorophore observed in the image represents the endosomal uptake (Fig. 7).

Fig. 7.

Confocal microscopy images of MCF-7 cells exposed to a) CUR/SIL loaded mSiO₂@SPIONs and b) control

VSM results

VSM was used for evaluating the magnetization of the prepared NPs. Due to the VSM curves (Fig. 8) the saturation magnetization values of the bare SPION was around 67 emu/g at room temperature with very low coercivity content and remanence, which confirms the superparamagnetic behavior of Fe₃O₄ NPs. The saturation magnetization content of mSiO2@SPIONs was reduced to ~ 38 emu/g that due to reports is still acceptable for theranostic applications [40].

Fig. 8.

VSM curves of SPION (black) and mSiO2@SPIONs (red)

In-vitro MRI

As shown in Fig. 9, the MRI show that effect of all contrast agents in T2* and T2-weighted images, and the signal intensity reduced by increasing the iron concentration in the samples. In comparison with the Endorem, the cells treated with silica coated NPs showed the darker images at the same Fe concentration. When the T2 relaxation rate of the SPIONs increases, its ability to shorten the proton relaxation time elevates. Therefore, the magnetic resonance contrast between samples becomes more obvious, leading to an enhancement in both resolution and accuracy. In addition, the reduction of the T2 relaxation times was observed from the lowest to highest concentrations, in which the signal intensity of the T2-weighted images decreased. These MRI results suggested that the cells uptake level of NPs correlated with the MRI signal intensity. MRI results indicated that mSiO2@SPIONs has high T2-weighted MRI contrast and can be potentially utilized in breast cancer diagnosis, particularly in early stage detection under in vitro conditions.

Fig. 9.

A) T2-weighted imaging, B) the MR signal intensity of MCF-7 treated with (a) CUR/SIL loaded mSiO₂@SPIONs and (b) Endorem

Discussion

In the present study, a nano-assembly of silica equipped with the so-called magnetic NPs, SPIONs, were designed and employed for carrying two herbal components, SIL and CUR.

According to reports preparation of SPIONs through microemulsion method provides some advantages over the very common method, co-precipitation. As stated by Thomas et al. SPIONs resulted from microemulsion showed narrower size distribution in comparison with co-precipitation [41]. This can be taken to the mean that less aggregation would be taken place through microemulsion method [41]. Another cause for microemulsion preference is the lower size of acquired SPIONs. This phenomenon attributes to CTAB molecules encompassing nanoreactors, mentioned in SPIONs synthesis, that prevent overgrowing of finally fabricated SPIONs [23, 42] which was confirmed by our obtained results, as well.

Previous studies on SPIONs faced particle aggregation as a serious challenge due to its magnetic nature. Matson et al. reported a high agglomeration in their synthesized ultrafine iron oxide nano-powder exhibited sub-10 nm size [43] and thus surface modification seems necessary. Moreover, due to reports uncoated SPIONs demonstrate some degree of toxicity leading to undesirable effects [44]. To address this issue various coatings have been suggested which amongst silica coverage not only leads to more safety and biocompatibility [45], but also confers more stability on SPIONs.

Various methods have developed for silica coating using CTAB, however, the effect of its concentration on size and PDI has been rarely considered.

In the current study various concentration of CTAB was investigated during the coating process and as a result the more CTAB concentration the higher size and PDI values were obtained. It seems the more CTAB the more micelles surround the SPIONs that finally lead to more attracted TEOS molecules and thicker silica membrane [46].

Although, removing the surfactant following the coating procedure did not affect the average particle size significantly however, reduced the PDI values notably. This can be explained by NP aggregations caused by extra CTAB, not entrapped within the silica framework, before the removal procedure. As claimed by Liu et al. NP aggregations occurred at medium concentration of CTAB (~ 0.1–10 mM) owning to charge neutralization and/or hydrophobic interactions. However, in CTAB concentrations both at very low (< 0.01 mM) and high (100 mM) the charges of NPs were enough to prevent aggregations and stabilize the system [47]. A CTAB agglomeration phenomenon is generally due to attraction of hydrophobic tail of CTAB molecules surrounding the NPs in monolayer format. Since NPs aggregation considered as a size homogeneity disturbing factor, PDI values fell after CTAB elimination process.

Comparison of the release profiles of CUR + SIL loaded mSiO2@SPIONs in pH 5.5 (Fig. 4a) and 7.4 buffers (Fig. 4b), revealed the more amount of released CUR in acidic condition. In pH 5.5 almost 41% of CUR was released from the NPs within 8 h. While in pH 7.4 the maximum released CUR concentration was two-fold lesser. The released SIL reached to the maximum concentration of 75% after 8 h. And unlike CUR, pH did not affect the final SIL release significantly, however, the burst release of SIL was significantly elevated in acidic condition (Fig. 4c, d). MTT assay showed synergic death on MCF-7 cells treated with both drugs, CUR and SIL compared with each drug alone. Montgomery et al.showed similar results on colon cancer cells [48]. Another paper implies that simultaneous application of CUR and SIL along with two other herbal agents might serve as a bowel cancer chemo-preventive drug [49].

Since the IC50 obtained from cells treated with dual drug loaded NPs was lower than the drug mixture (Table 1), the co-delivery of SIL and CUR by mSiO2@SPIONs was able to elevate their cytotoxic activity which may be due to the enhanced drug solubility and cellular uptake.

Singh et al. implied that by exposing MCF-7cell with SPIONs at concentration of 62.5–1000 ppm of Fe, more than 85% of the cells remained alive In our study all concentrations of NPs, by which the cells were treated, were under this level and thereby no significant cytotoxicity was observed by blank mSiO2@SPIONs (Fig. 6) [50]. Additionally as mentioned before, the silica coat plays a key role in reducing the SPIONs toxicity.

In the study of NP cellular uptakes the significant red color in confocal microscopy images confirmed the preferable cellular uptake of NPs via the endosomal pathway. Similar results were obtained by previous studies [37, 51].

Conclusion

Altogether, it can be concluded that mSiO2@SPIONs not only can improve the physicochemical properties of drugs such as herbal cytotoxic compounds but also may be a powerful carrier for drug delivery. The increased cellular uptake of CUR and SIL can be explained by enhancement in water solubility which led to decreased IC50 on MCF-7 cells.. Therefore, the combination of cytotoxic herbals with mSiO2@SPIONs, which addressed in this study, may introduce a new approach in cancer diagnosis and treatment researches.

Declarations

Conflict of interest statement

On behalf of all authors, the corresponding author states that there is no conflict of interest.