Preparation of berberine magnetic nanoparticles and their inhibition of human gastric cancer BGC-823 cells Open Access

Gastric cancer is one of the most common gastrointestinal malignancies worldwide.^1–3 Currently, the treatment methods for gastric cancer include surgery, chemotherapy, and targeted therapy. Due to the significant harm that surgery and chemotherapy can cause to the body, targeted therapy can be used to target tumor cells and reduce the degree of damage to normal cells. Therefore, targeted therapy is currently the main direction of research in gastric cancer therapy.^4–7 

Among targeted therapeutic modalities, magnetic nanoparticles are widely used in the biomedical field due to their unique magnetic responsiveness and biocompatibility. Fe₃O₄ nanoparticles have attracted considerable attention due to their strong magnetic properties, large specific surface area, and easy separation. They can be used as functional materials, such as for catalysts, sensors, and biological probes and in additional applications in the biomedical field.^8–13 Research has found that Fe₃O₄ has certain unavoidable problems that affect its practical application, such as poor dispersion and instability. Therefore, developing appropriate protection strategies for future applications is of great significance. Currently, surface functional modifications are generally preferred to improve the stability of the magnetic nanoparticles.^14–18 Carboxymethyl chitosan (CMCS) exists widely in nature, is non-toxic and harmless, and has antioxidant, anti-inflammatory, and other physiological activities. When it is encapsulated with Fe₃O₄ nanoparticles, it can increase the biocompatibility of the magnetic nanoparticles, improve their magnetic responsiveness, and increase their dispersion.^19–24 

In recent years, with in-depth research on the pathogenesis of tumors, many drugs have been developed and applied in clinical practice, and decisive progress has been made, especially on the antitumor effects and mechanisms of action of traditional Chinese medicine monomer compounds. Berberine (BBR) is an isoquinoline alkaloid distributed in a variety of medicinal plants. It has a wide range of pharmacological effects and high biological safety. It has inhibitory effects on many tumors in vivo and in vitro, such as gastric cancer, colon cancer, liver cancer, and breast cancer. Its remarkable biological activity and strong immune regulation ability have attracted the attention and interest of researchers, and it is widely used in clinical chemotherapy and radiotherapy. Berberine has less cytotoxicity in normal cells in vitro than in tumor cells, and no significant toxic side effects have been observed in vivo. It has been confirmed that berberine exerts its anticancer activity through multiple pathways, such as inhibiting angiogenesis and inducing apoptosis, and also improves resistance to chemotherapy drugs.^25–31 However, the insufficient targeting, low solubility, and poor oral absorption of berberine limit its use in tumors. Therefore, it is necessary to develop a drug carrier for transporting berberine. This carrier should have the following characteristics: ① It should passively target, actively target, or physically and chemically target tumor cells. ② It should provide intracellular controlled drug release.

In this study, targeted treatment of gastric cancer was taken as the entry point, and the macromolecular polymer CMCS was used to modify the surface of the magnetic nanoparticles (Fe₃O₄) to improve their stability. First, based on single-factor experiments and using particle size as an evaluation index, the response surface methodology was used to optimize the preparation process of Fe₃O₄ magnetic nanocomposites and characterize their structural properties. Second, the prepared magnetic nanocomposites were further used to load the anticancer drug berberine (BBR), and the structural properties of the magnetic nanocarrier system were characterized. The drug loading and release properties of the drug delivery system loaded with the drug BBR on CMCS-coated magnetic nanoparticles (Fe₃O₄@CMCS-BBR) were then analyzed using a UV visible spectrophotometer. The inhibitory effect and biocompatibility of the drugs on the gastric cancer cell line BGC-823 were studied. The magnetic properties and targeting properties of the drug delivery systems were studied using fluorescence microscopy. Finally, the degree of cell apoptosis was detected using flow cytometry.

All the materials used in this study were of analytical grade and required no further purification. Chemicals used for the synthesis of CMCS-coated magnetite nanoparticles were ferrous chloride tetrahydrate (FeCl₂·4H₂O, Kermel), ferric chloride hexahydrate (FeCl₃·6H₂O, Kermel), sodium hydroxide (NaOH, Tianli) and CMCS (obtained from Macklin). Berberine (Yuanye) was used for the drug loading and release study.

Fourier transform-infrared (FT-IR) spectra were measured using a Bruker instrument. X-ray diffraction (XRD) patterns of the samples were taken with a Philips x-ray diffractometer over a 2θ range from 10° to 80°, using Cu Kα radiation (λ¼ 0.154 nm). The morphology and size of the particles were examined by transmission electron microscopy (TEM) (JEM 2100F). The magnetic properties of the samples were studied using a vibrating sample magnetometer (VSM) (Lakeshore). UV–Vis spectra were measured using a spectrophotometer. The inhibition rate of the drug on gastric cancer cells was studied using an enzyme-linked immunosorbent assay (Tecan). The magnetic targeting of the sample was studied using a fluorescence microscope (OLYMPUS). The degree of cell apoptosis was studied using flow cytometry.

A chemical co-precipitation method was used to prepare magnetic nanoparticles by the simultaneous hydrolysis reaction of trivalent iron ions with divalent iron ions in an aqueous solution.³² FeSO₄·7H₂O and FeCl₃·6H₂O were accurately weighed at a ratio of 1:2 to evenly disperse them in deionized water. The reaction principle can be simplified as follows: Fe²⁺ + 2Fe³⁺ + 8OH⁻ → Fe₃O₄ + 4H₂O. The precipitant used was a sodium hydroxide solution, which was stirred at an appropriate temperature and pH. The speed of the stirring reaction was fast, and the entire reaction process was carried out in an environment filled with nitrogen gas. After the reaction, the mixture was cooled to room temperature, separated with an external magnet, washed three times, and dried in a 45 °C vacuum oven for 24 h to obtain the final product, Fe₃O₄ magnetic nanoparticles, followed by characterization of their morphology and performance.^33,34

In reference to previous work,³⁵ 0.40 g of Fe₃O₄ was weighed and dissolved in 100 ml of deionized water, and 0.30 g of 1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDCI) and 0.16 g of N-Hydroxy Succinimide (NHS) were added under mechanical stirring. After the reactants were activated for 2 h, 200 ml of an 8 mg ml⁻¹ CMCS aqueous solution was added, and the reaction continued at a specific temperature range for 72 h. After the reaction was completed, the precipitate was separated through centrifugation, washed three times with deionized water and anhydrous ethanol, placed in a vacuum drying oven at 45 °C, and dried for 24 h for use.

The following parameters were used for the single-factor experiment: m_Fe3O4:m_CMCS ratios were 1:1, 1:2, 1:3, 1:4, and 1:5; the concentrations of CMCS were 2.0, 4.0, 6.0, 8.0, and 10.0 mg ml⁻¹; and the reaction temperatures were 25, 35, 45, 55, and 65 °C. Using the particle size of the magnetic nanocomposites as an evaluation indicator, the influence of single factors on the particle size of the magnetic nanocomposites was investigated within a reasonable range.

The single-factor results were analyzed (Table I), and three factors were selected—m_Fe3O4:m_CMCS (A), the concentration of CMCS (B), and reaction temperature (C)—for the response surface optimization experiments.

The single-factor experimental data were calculated using Origin, Design-Expert v. 10.0.3 software was used for the response surface experimental design, and a quadratic regression model was used for the analysis of variance on the experimental results.

According to the Chinese Pharmacopoeia 2020 version III UV–Vis spectrophotometric method,³⁶ 100 mg of berberine was weighed in a 100 ml volumetric flask, and a berberine solution at a concentration of 1 mg ml⁻¹ was prepared by adding Dimethyl Sulfoxide (DMSO) to the calibration line. The solution was sonicated for 1 h. Then 1, 1.5, 2, 3, and 4 ml of berberine solution were placed in 50 ml volumetric flasks numbered 1–5, and DMSO was added to the calibration line The absorbance at a wavelength of 424 nm was measured and fitted to obtain the equation of the standard curve.

The biocompatibility of the synthesized magnetic nanocomplexes in BGC-823 cells and the in vitro antitumor efficiency of the magnetic nanodrug delivery system were investigated using the CCK-8 method. BGC-823 cells in log phase growth were washed two to three times with PBS to remove the dead and floating cells as well as metabolites. The cells were digested with a digestion solution, collected in a centrifuge tube, and centrifuged for 5 min, the supernatant was discarded, new culture medium was added, and the cells were resuspended. The cell concentration was adjusted by adding culture medium, and the cells were counted using a cell counting plate. Then, 96-well plates were inoculated at a density of 5000 cells/well, ∼100 µl per well, and the plates were incubated for some time at 37 °C and 5% CO₂.

The main objective was to investigate the effect of the magnetic nanodrug carrier on tumor cells under the action of an applied magnetic field. First, BGC-823 gastric cancer cells were cultured, and cells cultured to the logarithmic phase were inoculated in cell culture dishes. After incubation in a CO₂ incubator, the magnetic nanocarriers loaded with berberine were added, and the dishes were placed in an incubator with a constant applied magnetic field on one side of the dishes and no magnetic field on the other side. After 24, 48, and 72 h of incubation under the applied magnetic field (3T), the cells in the magnetic field area and the non-magnetic field area of the culture dish were observed by light microscopy, and the distribution of magnetic carriers and the growth of cells in the two areas were recorded and compared.

Briefly, 10 mg of the magnetic coupling system was added to 100 µl of a 1 mg ml⁻¹ fluorescein isothiocyanate (FITC)/DMSO solution and stirred at room temperature without light for 24 h. The mixture was centrifuged and oven-dried for 24 h.

In brief, 10 mg of Dil was fully dissolved in 2.14 ml of anhydrous DMSO to obtain a 5 mM staining solution and then diluted with serum-free RPMI-1640 medium.

Cultured BGC-823 gastric cancer cells were inoculated in 24-well plates at 4 × 10⁴ cells/well, and 100 μl of the Dil cell membrane staining solution was added. The staining solution was gently shaken to cover the medium. 4',6-diamindino-2-phenylindole (DAPI) was added to the medium at 100 μl/well with a pipette gun, and the cells were incubated at 37 °C for 20 min after gently shaking until the staining solution covered the medium. After the cells were attached to the plates, the nanodrug was added to the cells in 1640 serum-free medium at 455 μg ml⁻¹, and the cells were incubated for 6 h in a fixed manner. After the drug was applied to the cells, the 24-well plates were removed at 2, 4, and 6 h; the culture medium was discarded, and cells were washed with PBS to remove the dead cells and the drug that did not enter the cells. The method of drug entry into the cells and its distribution were observed under a normal field of view and fluorescence excitation field of view using an orthogonal fluorescence microscope.

Using an AnnexinV-FITC/PI double-staining method, gastric cancer BGC-823 cells in a logarithmic growth phase were selected, and 5 ml of the drug solution was added to each of the drug-treated groups. The blank group received PBS. Then, 5 ml of DMEM culture medium containing 10% fetal bovine serum (1640 culture medium containing 10% fetal bovine serum) was added, and cells were placed in a CO₂ cell incubator. After incubating the drug with the cells for 72 h, 400 µl of the AnnexinV conjugate was added and repeatedly mixed to disperse into a single-cell suspension. Then, 5 µl of AnnexinV-FITC staining solution was added to the single-cell suspension, gently mixed, and protected from light at 2–8 °C for 15 min. Then 10 µl of PI staining solution was added, cells were gently mixed well while avoiding light for 5 min at 2–8 °C, and the cell regulation assay was performed on a flow cytometer.

The morphologies of Fe₃O₄, Fe₃O₄@CMCS, and Fe₃O₄@CMCS-BBR were characterized, and the transmission electron microscopy images are shown in Fig. 2(a). The synthesized Fe₃O₄@CMCS-BBR particles were approximately spherical, with a uniform particle distribution, and had significantly improved agglomeration dispersion compared with Fe₃O₄. The hydrodynamic diameter distributions of Fe₃O₄, Fe₃O₄@CMCS, and Fe₃O₄@CMCS-BBR particles in an aqueous solution are shown in Fig. 2(b). As can be seen from the figure, the average particle sizes of the three nanoparticles were 18.91, 39.26, and 73.75  nm. The particle sizes of all the particles were uniformly distributed and were present in a normal distribution, which is in accordance with our requirements. The dynamic light scattering (DLS) results were consistent with those obtained through transmission electron microscopy. The XRD patterns of Fe₃O₄, Fe₃O₄@CMCS, and Fe₃O₄@CMCS-BBR are shown in Fig. 2(c). In the spectrum of Fe₃O₄@CMCS, there was an obvious Fe–O absorption peak near 568.89 cm⁻¹, indicating the presence of Fe₃O₄. In addition, it was observed that in Fe₃O₄@CMCS, there were 1058.25 cm⁻¹ C–O–C absorption peaks and 1608.34 cm⁻¹ C=N absorption peaks in the spectrum, which correspond to 1064.99 and 1608.82 cm⁻¹ in CMCS, respectively. It was preliminarily determined that CMCS successfully modified Fe₃O₄. A distinct Fe–O characteristic peak appeared in the Fe₃O₄@CMCS-BBR spectrum at 572.75 cm⁻¹, which was consistent with the Fe₃O₄@CMCS peak corresponding to 568.89 cm⁻¹. Fe₃O₄@CMCS-BBR at 1062.58 cm⁻¹ was the strong absorption peak of C–O–C in CMCS, 1611.71 cm⁻¹ was the absorption peak of C=N, and Fe₃O₄@CMCS had strong peaks at 1608.34 and 1058.25 cm⁻¹. The most obvious feature was at 1510.47 cm⁻¹ in Fe₃O₄@CMCS-BBR, where the aromatic skeleton vibration absorption peak in the drug berberine appeared, corresponding to 1537.21 cm⁻¹ in the berberine spectrum, indicating the successful loading of berberine. The hysteresis line plots of Fe₃O₄@CMCS and Fe₃O₄@CMCS-BBR at room temperature are shown in Fig. 2(d). It can be seen that the patterns were not significantly different, with typical diffraction peaks appearing at 2θ of 30°, 35°, 43°, 57°, and 63°, which were consistent with the standard card of Fe₃O₄ and with the diffraction peaks of the crystalline surfaces (220), (311), (400), (511), and (440). This indicates the successful synthesis of the magnetic Fe₃O₄ nanoparticles. The FT-IR spectra of CMCS, Fe₃O₄, and Fe₃O₄@CMCS were analyzed and are shown in Fig. 2(e). The saturation magnetization intensity of Fe₃O₄ nanoparticles was as high as 84.68 emu/g, and in Fe₃O₄@CMCS, the saturation magnetization of the particles was reduced to 64.29 emu/g. The main reason is that the macromolecular polymers were non-magnetic substances, resulting in relatively weaker magnetic properties of the prepared magnetic nanocomplexes, but the magnetic properties still responded to applied magnetic fields, indicating that the magnetic nanocomplexes were present. The saturation magnetization of the particles was reduced to 50.46 emu/g, which was because the loaded drug berberine is a non-magnetic substance, and therefore, there was a non-magnetic component in the prepared magnetic nano-loaded system, which caused its magnetic properties to be relatively weaker and further indicated the successful binding of the loaded drug system. With the concentration of berberine solution as the horizontal coordinate and the absorbance value as the vertical coordinate, the standard curve equation was fitted: Y = 8.7978X + 0.027 97. The standard curve, in which the concentration range was from 0 to 0.10 mg/ml with R² = 0.996 13, indicated a good linear relationship [Fig. 2(f)]. Nanocomposites loaded with berberine were manipulated and deposited at the bottom of seven different pH solutions with an applied magnetic field. Samples were collected from the supernatant and analyzed using a UV spectrophotometer [Fig. 2(g)]. The maximum encapsulation rate was 51.23% when the pH was 5.5 (Table V). The release curve of berberine from a magnetic nano-loaded system at pH 7.4 is shown in Fig. 2(h). The drug release curve shows that Fe₃O₄@CMCS-BBR released ∼85.89% of berberine within 72 h. The release of berberine was faster in the first 2 h, and the release gradually slowed after 2 h. This is because the pH of berberine changed from 5.5 to 7.4 in the magnetic nanodrug delivery system, and the charge of berberine changed from positive to negative, which caused the electrostatic attraction between berberine and the surface macromolecular polymer to change, which then changed the electrostatic attraction between the two to electrostatic repulsion. This resulted in the rapid release of the drug in the short term, while the slow release in the later period was due to the magnetic nanodrug delivery system, in which the surface-encapsulated macromolecular polymer hindered the release of berberine molecules. According to the in vitro Qn, the correlation equations were fitted. The in vitro release pattern of the magnetic nanodrug delivery system was initially studied by comparing the correlation coefficients of each equation. The fitting results showed that the in vitro release equations of the magnetic nanodrug-loaded system were close to the first level release model, with the first level release > Higuchi model > zero level release model (Table VI).

The in vitro antitumor effects of the magnetic nanocomplex, berberine, and the magnetic nanocommunication system loaded with berberine were examined at different time points using the CCK-8 method [Fig. 3(a)]. After berberine treatment for 24, 48, and 72 h, the IC50 values of gastric cancer cells were 57.54, 28.91, and 28.83, respectively. The results showed that the survival rate of BGC-823 cells was more than 95%, even when cultured at high concentrations of 1000 µg/ml for 24, 48, and 72 h. This indicated that the magnetic nanocomplex had good biocompatibility and was an excellent nanocarrier. The in vitro antitumor activity of the magnetic nanodrug delivery system in gastric cancer BGC-823 cells was evaluated using the CCK-8 method. The cell survival rate decreased with an increasing drug concentration and longer incubation time [Fig. 3(b)]. The IC50 values of Fe₃O₄@CMCS-BBR after 24, 48, and 72 h of incubation were 538.3, 558.4, and 544.3 µg/ml, respectively. The IC50 of free berberine was lower than that of Fe₃O₄@CMCS-BBR, mainly because free berberine entered the cells by diffusion but Fe₃O₄@CMCS-BBR entered the cells through endocytosis. These studies suggest that more Fe₃O₄@CMCS-BBR entered the cancer cells and released berberine as the time increased. The distribution of Fe₃O₄@CMCS-BBR under a magnetic field and a non-magnetic field was observed. There was no change in the number and state of cells under a magnetic field at 0 h compared with the non-magnetic field [Fig. 3(c)]. After 6, 12, and 24 h of incubation, there was almost no magnetic nanoparticle aggregation without the magnetic field, but a large amount of magnetic nanoparticle accumulation was observed with the magnetic field, and the color changed to black. Therefore, this indicated the strong magnetic responsiveness of the prepared magnetic nanocarrier system. The uptake pattern and distribution of Fe₃O₄@CMCS-BBR in gastric cancer BGC-823 cells were studied with ortho-fluorescence microscopy. The fluorescence of the DAPI-stained cell nucleus, the fluorescence of the Dil-stained cell membrane, the fluorescence of the FITC-coupled magnetic nano-loading system, and the combined fluorescence are shown in Fig. 3(d). The drugs were incubated with the cancer cells for 2, 4, and 6 h. At 2 h, there was almost no distribution of the drug in the nucleus, and the fluorescence of the drug at this time was weak. At 4 h, drug fluorescence was already present in the nucleus. At 6 h, the drug fluorescence was very obvious, indicating that the uptake process of the cells was time-dependent. The experiments showed that Fe₃O₄@CMCS-BBR entered BGC-823 cells in an endocytic manner to release the drug and kill the cancer cells. The human gastric cancer BGC-823 cells treated with the magnetic nano-loaded system and the blank control solution were stained using an AnnexinV-FITC/PI double-staining kit and analyzed using a flow cytometer after staining. The scatter plot of the experimental results was obtained [Fig. 3(e)]. The apoptosis rate of the Fe₃O₄@CMCS-BBR-treated cells was 54.90%, indicating that the magnetic nano-loaded system promoted apoptosis of BGC-823 cells and therefore is an excellent targeted slow-release formulation.

In this study, targeted drug delivery was used as the entry point for the treatment of gastric cancer. A surface modification of the magnetic nanoparticles was carried out using macromolecular polymers, and the preparation process was optimized. Furthermore, the modified magnetic nanocomplex was used to load the drug berberine to improve the stability of the magnetic nanoparticles and the targeting of berberine, thus increasing the anti-gastric cancer effect of berberine.

In this study, Fe₃O₄ magnetic nanoparticles were first prepared through surface modification of Fe₃O₄ particles with macromolecular polymers. Second, the preparation process of three magnetic nanocomplexes was optimized using the response surface method and single-factor optimization test, resulting in an m_Fe3O4:m_CMCS ratio of 1:3.997, a CMCS concentration of 7.986 mg/ml, and a reaction temperature of 36.457 °C. The morphological features and structural properties of the synthesized magnetic nanocomplexes were characterized through TEM, DLS, XRD, FT-IR, and VSM, which confirmed that CMCS had successfully modified Fe₃O₄ to form magnetic nanocomplexes.

The drug loading properties of the magnetic nanocomplexes were investigated using the anticancer drug berberine as a carrier drug. The analysis of XRD, FT-IR, and VSM showed the successful loading of berberine. UV–Vis spectrophotometry was used to evaluate the drug loading performance and encapsulation rate of Fe₃O₄@CMCS-BBR, and the results showed that the maximum Fe₃O₄@CMCS-BBR EE% was 51.23% and that the maximum drug loading was 17.10% at a pH of 5.5. According to the drug release curve, within 72 h, Fe₃O₄@CMCS-BBR released ∼85.89% of berberine, and the in vitro release equation fitting by the correlation equation showed that the in vitro release equation of the magnetic nanocarrier system was close to a primary release model.

The inhibitory effect of Fe₃O₄@CMCS-BBR on gastric cancer BGC-823 cells was examined using the CCK-8 method. The results showed that the magnetic nanocomplexes were biocompatible with gastric cancer BGC-823 cells and that the magnetic nano-loaded drug system effectively inhibited the proliferation of cancer cells. The results of the magnetic property evaluation experiments showed that Fe₃O₄@CMCS-BBR had good responsiveness to an external magnetic field. In cell targeting evaluation experiments, the drug was co-incubated with BGC-823 cells for 2, 4, and 6 h. It was observed that the drug emitted obvious green fluorescence inside the cells and that the killing effect on cancer cells was greater than that without the drug, which also showed that the drug entered and acted on cancer cells in an endocytic manner. The degree of apoptosis was analyzed using flow cytometry, and the results indicated that Fe₃O₄@CMCS-BBR had a significant effect on promoting apoptosis of BGC-823 cells.

In summary, the magnetic nanocomplex prepared in this study has good bioactivity, magnetic properties, targeting properties, and good biocompatibility with gastric cancer BGC-823 cells due to their surface modification, which provides a foundation for further in vivo experiments.

This work was supported by the National Key R&D Program of China (Grant No. 2018YFC1706903), the National Natural Science Foundation of China (Grant No. 81403101), the Project of Shaanxi Key Laboratory of Basic and New Herbal Medicament Research (Grant No. 19JS019), and the Sci-Tech Innovation Talent System Construction Program of Shaanxi University of Chinese Medicine.

The authors have no conflicts to disclose.

X. L. and J. W. contributed equally to this paper as co-first authors.

Xianglong Liu: Methodology (equal); Validation (equal); Writing – review & editing (equal). Jiao Wang: Resources (equal); Writing – original draft (equal). Bodong Chen: Investigation (equal). Ben Niu: Project administration (equal). Jin Li: Conceptualization (equal); Project administration (equal).

The research results reported in this paper are the original research results of mine or the staff in this research group. Except where the author specifically notes and thanks, this paper does not contain research results that have been published or written by others but not yet published, and there is no infringement of the intellectual property rights of others or other copyright owners.