3.1. Assessment of the cytotoxicity of chalcones in vitro
During the study, 11 chalcones with different substituents were subjected to cytotoxicity tests against gastric adenocarcinoma (AGS) and breast cancer (MCF-7) cell lines in order to identify the most cytotoxic chalcone for subsequent incorporation into mesoporous silica nanoparticles. The results obtained showed that the chalcones exhibited significant cytotoxicity, with some showing better activity than cisplatin. Table 1 shows the IC50 values and selectivity of the chalcones against AGS and MCF-7 cells versus fibroblast cells (L-929).
It was observed that most of the chalcones were more active against AGS and MCF-7 cells than against fibroblast cells (L-929), which shows that chalcones have greater selectivity for cancer cells than for normal cells, an important characteristic for anticancer agents. Furthermore, some of the chalcones showed a cytotoxic effect close to that of cisplatin against AGS and MCF-7 cells, with greater selectivity against these cell types than cisplatin.
In general, the effects of the substituents on the cytotoxic activity of chalcones against AGS and MCF-7 cells were similar. Compound 1, the unsubstituted chalcone (benzalacetophenone), showed good cytotoxic activity, with IC50 ranging from 53.11 ± 1.63 to 57.93 ± 19.59 µM, which led us to investigate the influence of different substituents on the cytotoxicity of chalcones.
Methoxyl and hydroxyl groups are known to positively influence the cytotoxic activity of chalcones [5, 21]. The presence of the methoxy substituent in position 3 of ring B had a positive effect on the cytotoxic activity of chalcone 4 in both cancer cell lines (IC50 from 44.55 ± 0.09 to 48.12 ± 2.21 µM), while the methoxy group in position 2 of ring B (compound 3) was only beneficial for cytotoxicity against AGS cells (IC50 46.96 ± 3.38 µM), and its effect against MCF-7 cells (IC50 of 81.04 ± 16.52 µM) was negative compared with compound 1. Other tested chalcones with methoxy groups on the aromatic rings (compounds 2, 8 and 10) showed higher IC50 values, indicating that the position of the methoxy group on the aromatic ring significantly influences the activity of these compounds.
Compound 11 (3-hydroxychalcone) showed significant cytotoxic activity against AGS and MCF-7 cells (IC50 from 47.58 ± 0.16 to 47.97 ± 2.56 µM), showing that the presence of the hydroxyl group at position 3 of the ring has an important effect on the cytotoxicity of chalcones.
When analyzing chalcones with electron-withdrawing groups (chlorine, nitro) in their structure, it was found that chalcone 6 (3-chlorochalcone) showed good cytotoxic activity (IC50 from 44.54 ± 1.24 to 45.58 ± 4.66 µM), while chalcones 5 (3-chloro-4’-nitrochalcone), 7 (4’-nitrochalcone), 9 (4-nitrochalcone) and 10 (4-methoxy-4’-nitrochalcone) showed moderate cytotoxic activity. The presence of the chlorine group in position 3 of ring B exerts an important influence on the activity of chalcones, while the presence of the nitro group in position 4 of ring A and ring B appears to have a negative effect on the cytotoxic activity, as can be observed in compounds 5, 7, 9 and 10.
In general, the chalcones showed significant cytotoxic activity against AGS and MCF-7 cells, with chalcones 4 (3-methoxychalcone), 6 (3-chlorochalcone) and 11 (3-hydroxychalcone) standing out. Considering the cytotoxicity profile in both tumor cell lines and its physicochemical characteristics, 3-hydroxychalcone was selected for incorporation into mesoporous silica nanoparticles. It is worth highlighting that the hydroxyl group present in 3-hydroxychalcone could favor interaction with the silanol groups of silica though hydrogen interactions. The functionalization process was subsequently carried out with the purpose of adding amino groups to the silica surface. The presence of these groups on the silica surface could further favor the incorporation process by increasing the hydrogen interactions between the nanoparticles and 3-hydroxychalcone, with the possibility of incorporating a greater amount of chalcone into the silica.
3.2. Scanning electron microscopy (SEM)
Scanning electron microscopy (SEM) is an important technique for studying the morphology (texture, shape and size) of materials. From SEM images, it was possible to estimate the shape of the mesoporous silica nanoparticles. The MSN and MSN-APTES silica (Fig. 2) showed spherical particles, which is consistent with the results described in the literature [22]. However, it was possible to observe aggregates of different sizes, which are common in nanometer-sized particles.
3.3. Transmission electron microscopy (TEM)
Transmission electron microscopy (TEM) was used to characterize the pore structure of MSN and MSN-APTES mesoporous silica. The images show that there were pores, which appeared in a spherical shape and in a lighter tone due to the silica nanoparticles (Fig. 3). As observed in the SEM analysis, the silica particles were agglomerated, which can be considered a limiting factor in this type of analysis, as it makes it difficult to visualize the silica structures more clearly.
3.4. Fourier transform infrared spectroscopy (FTIR)
FTIR analysis was used to determine the chemical composition of the obtained mesoporous silica nanoparticles and the presence of chemical groups of the functionalizing agent on the silica surface. Figure 4 shows the FTIR spectra of MSN and MSN-APTES nanoparticles. The characteristic bands at 1080, 810 and 460 cm− 1 are related to the asymmetric stretching, bending (δ) and out-of-plane Si-O bonds, respectively. The infrared spectrum of the MSN sample showed peaks of hydroxyl stretching band for O-H axial deformation vibration mode of the Si-O-H group and water molecules at 3550 cm− 1 and 1640 cm− 1. The band at 960 cm− 1 is typical of the silanol groups (Si-OH) present in the silica structure [23]. After the functionalization process with APTES, this band showed a change in intensity. A decrease in the intensity of the band at 960 cm− 1 was observed, which can be attributed to the reduction in the number of silanol groups present in the silica structure due to the process of binding of the functionalizing agent to these functional groups.
For silica functionalized with APTES, a peak at 1560 cm− 1 corresponding to the NH2 deformation vibration was observed, indicating the presence of functionalizing amino groups on the surface of the mesoporous silica nanoparticles. Furthermore, the bands observed at 1460 − 1410 cm− 1 correspond to CH-CH2 bending [24]. These results are consistent with elemental analysis (CHN), in which an increase in the percentage of carbon and nitrogen was observed after the functionalization process.
3.5. Zeta potential analysis
Zeta potential analysis was used to evaluate the surface electric charge of particles of the MSN and MSN-APTES samples and of the samples incorporated with 3-hydroxychalcone (MSN-CHO and MSN-APTES-CHO). Table 2 shows the results.
Table 2
Sample | Zeta potential ± DP (mV) |
MSN | -6.65 ± 0.93 |
MSN-CHO | -3.37 ± 0.92 |
MSN-APTES | -2.93 ± 0.99 |
MSN-APTES-CHO | -2.89 ± 0.95 |
MSN silica has a negative charge (-6.65 mV) due to the silanol groups present on its surface. After the functionalization process with APTES, a decrease in the negative charge of the functionalized particles was observed due to the contribution of the amino groups of the functionalizing agent. The functionalization of MSN silica with APTES promotes the formation of covalent bonds between the molecules of the silanol groups present on the silica surface and the aminosilane groups of the functionalizing agent. As a result, the negative charge of the nanoparticles decreased from − 6.65 mV in the MSN sample to -2.93 mV in the MSN-APTES sample. These results show that the functionalization process was efficient and are in line with what was described by Andrade et al. [20]. For the samples incorporated with 3-hydroxychalcone, a decrease in the negative charge of the particles was observed. This result can be attributed to the presence of chemical groups in the 3-hydroxychalcone molecule, which contribute to the reduction of the negative charge of the sample.
3.6. Nitrogen adsorption
Nitrogen adsorption analysis was used to obtain information such as specific surface area, pore diameter, distribution and total volume by measuring the area occupied by nitrogen molecules adsorbed on the silica surface [25]. The specific surface area values of the MSN silica sample were calculated using the BET method, and the total volume and pore diameter values were calculated using the BJH method (Table 3).
Table 3
Nitrogen adsorption/desorption
Sample | Surface area m².g− 1 | Pore diameter BJH/DFT | Volume cc/g |
MSN | 115 | 3.5 nm | 0.391 |
MSNAPTES | 20 | 3.3 nm | 0.060 |
The values obtained are characteristic of mesoporous materials, confirmed by pore diameter values from 3.3 to 3.5 nm, which are within the expected standard for mesoporous materials (2 to 50 nm) [13]. The small variation in the pore diameters of the MSN (3.5 nm) and MSN-APTES (3.3 nm) samples evidences the occupation of the pores by the amino groups after the functionalization process.
The surface area and pore volume were 115 m2/g− 1 and 0.391 cc3.g− 1 for the MSN sample and 20 m2/g− 1 and 0.060 cc3.g− 1 for the MSN APTES sample, respectively, indicating that the functionalization process of MSN silica with APTES was successful, given the decrease in surface area and pore volume. Although smaller, the surface area of MSN-APTES silica is large enough to incorporate molecules.
Furthermore, the isotherms of the MSN and MSN-APTES samples showed adsorption and desorption branches that formed a type H1 hysteresis, characteristic of mesoporous materials (Fig. 5). A smaller volume of adsorbed nitrogen was found for the MSN-APTES sample than for the MSN silica in almost all relative pressures, indicating that the functionalization process was successful due to pore filling by the functionalizing agent, corroborating the results shown in Table 3, in which smaller pore diameter and volume values can be observed for the MSN-APTES sample compared with the MSN sample.
3.7. Thermogravimetric analysis (TGA) and elemental analysis (CHN)
Figure 6 shows the mass loss curves of the MSN, MSN-CHO, MSN-APTES and MSN-APTES-CHO samples. For the MSN sample, a mass loss of 3.7% was observed in the temperature range of 25°C to 150°C, which is related to the adsorption of water molecules on silica. From 150°C to 750°C, the mass loss was 9.5%, corresponding to the condensation of silanol groups [26]. An increase in mass loss was also observed for the MSN-CHO sample from 150°C to 750°C (11.3%), which can be attributed to the thermal degradation of the incorporated 3-hydroxychalcone. This 1.8% increase in the percentage of mass loss in the sample incorporated with 3-hydroxychalcone (MSN-CHO) compared with the unloaded sample (MSN) indicates the presence of chalcone in this sample and also shows that the incorporation process was successful.
For samples MSN-APTES and MSN-APTES-CHO, the mass loss in the temperature range of 150 to 750°C was 20.2% and 25.8%, respectively. The substantial mass loss for these samples in this temperature range can be attributed to the presence of amino groups of the functionalizing agent attached to the silica surface and the presence of 3-hydroxychalcone for the loaded sample, indicating that the functionalization and incorporation process was successful. The greater mass loss of MSN-APTES-CHO (5.6%) than MSN-CHO (1.8%) means that the incorporation process is more efficient in functionalized silica, due to the presence of amino groups on the silica surface, which favors the incorporation process by increasing the number of hydrogen interactions between silica and 3-hydroxychalcone.
The results obtained by thermogravimetric analysis are in line with those obtained using the elemental analysis technique (CHN). These results provided additional evidence that the silica samples were loaded with 3-hydroxychalcone and that the functionalization process was successful (Table 4).
Table 4
Mass loss and elemental analysis (CHN) of mesoporous silica nanoparticle samples
Sample | Weight loss (% m/m) 25–150ºC | Weight loss (% m/m) 150–750ºC | Residue (%m/m) > 800ºC | % Carbon | % Nitrogen |
MSN | 3.7 | 9.5 | 86.8 | 5.43 ± 0.04 | -0.18 ± 0.011 |
MSN-CHO | 3.4 | 11.3 | 85.3 | 5.91 ± 0.04 | -0.64 ± 0.0025 |
MSN-APTES | 7.6 | 20.2 | 72.2 | 13.02 ± 0.06 | 2.70 ± 0.055 |
MSN-APTES-CHO | 7.0 | 25.8 | 67.2 | 14.30 ± 0.06 | 2.55 ± 0.025 |
The MSN sample presented a carbon percentage of 5.43%, this percentage can be explained by the presence of the copolymer F127 that was not completely removed during the extraction process. After the incorporation process with 3-hydroxychalcone, the MSN sample (5.43%) presented an increase in the carbon percentage (5.91%), indicating that the incorporation process was successful. However, the carbon variations between the MSN and MSN-CHO samples were minimal, indicating that 0.48% (9.6 µg) of 3-hydroxychalcone was incorporated into the mesoporous silica nanoparticles at the tested concentration (2 mg/mL).
Furthermore, the MSN sample showed an increase in the percentage of carbon and nitrogen after the functionalization process (MSN-APTES), indicating that this process was successful and that the binding of the functionalizing agent 3-aminopropyltriethoxysilane occurred on the silica surface. An increase in the percentage of carbon was also observed in the functionalized sample incorporated with 3-hydroxychalcone (MSN-APTES-CHO). For APTES silica, the incorporation rate was found to be 1.28% or 25.6 µg—almost three times higher than the rate found for the MSN-CHO sample (0.48%). Functionalization with APTES favors the incorporation process due to hydrogen interactions between the amino groups of the functionalized silica and the 3-hydroxychalcone molecules, resulting in a higher incorporation rate.
3.8. 3-hydroxychalcone loading and release
After the incorporation process, the amount of 3-hydroxychalcone loaded was 0.48% or 9.6 µg for the MSN-CHO sample and 1.28% or 25.6 µg for the MSN-APTES-CHO. These values were obtained using the elemental analysis (CHN).
Figure 7 shows the release profile of the MSN-CHO and MSN-CHO-APTES samples. For MSN-CHO, an initial rapid release of 3-hydroxychalcone can be observed, around 30% (2.88 µg) in one hour, followed by a gradual and prolonged release, with 52.60% (5.05 µg) of chalcone released after 54 hours. These results can be explained by the interaction between the silanol groups of silica and the hydroxyl and carbonyl groups of 3-hydroxychalcone via hydrogen bonding, resulting in a slower release of 3-hydroxychalcone from the silica matrix.
The MSN-APTES-CHO sample showed a slower release, with 16.84% (4.31 µg) of 3-hydroxychalcone released in 54 hours. This characteristic can be explained by a possible more intense interaction between 3-hydroxychalcone and functionalized silica. The addition of amino groups to the silica surface can promote a greater number of hydrogen interactions between the nanoparticles and 3-hydroxychalcone, leading to slower release rates.
These findings show that mesoporous silica nanoparticles can be adapted to provide an initial release of an adequate dose of the bioactive compound in a short period of time, followed by a slow and gradual release of the drug over a long period of time, which can be useful for maintaining constant compound levels and action and reducing side effects.
3.9. Evaluation of the cytotoxicity of silica samples in vitro
Silica samples incorporated with 3-hydroxychalcone (MSN-CHO and MSN-APTES-CHO) and isolated silica samples (MSN and MSN-APTES) were subjected to cytotoxicity tests against gastric adenocarcinoma (AGS) and breast cancer (MCF-7) to verify the effectiveness of these systems in enhancing the cytotoxic activity of 3-hydroxychalcone.
Table 5 shows the IC50 values and selectivity of the silica samples against AGS, MCF-7 and fibroblast cells (L-929).
Table 5
IC50 (µM) and selectivity index (SI) of silica samples
Sample | IC50 L-929 | IC50 AGS | IC50 MCF-7 | IS AGS | IS MCF-7 |
3-hydroxychalcone | 275.03 ± 5.96* | 47.58 ± 0.16 | 47.97 ± 2.56 | 5.78* | 5.75* |
MSN | > 6000 | > 6000 | > 6000 | 1.00 | 1.00 |
MSN-CHO | 49.99 ± 1.11 | 12.93 ± 0.94 | 22.30 ± 4.15 | 4.16** | 2.06 |
MSN-APTES | 1619 ± 69.30 | > 7300 | > 7300 | 0.22 | 0.22 |
MSN-APTES-CHO | 73.67 ± 8.07 | 106.67 ± 7.80 | 80.98 ± 14.76 | 0.82 | 0.83 |
Cisplatin | 24.03 ± 7.42 | 39.59 ± 0.66 | 52.85 ± 4.80 | 0.61 | 0.46 |
IC50 and IS of chalcones and cisplatin (*: p < 0.0001; **: p < 0.001 compared with cisplatin).
The silica samples loaded with 3-hydroxychalcone showed significant cytotoxic activity (IC50 from 12.93 ± 0.94 to 106.67 ± 7.80 µM) against AGS and MCF-7 cells, with the MSN-CHO sample showing a better cytotoxic effect (IC50 from 12.93 ± 0.94 to 22.30 ± 4.15 µM) than free 3-hydroxychalcone (IC50 from 47.58 ± 0.16 to 47.97 ± 2.56 µM). These results indicate that the loaded mesoporous silica nanoparticles significantly influenced the interaction of chalcone with tumor cells, requiring lower concentrations of 3-hydroxychalcone to exert greater cytotoxic activity compared with free chalcone. While the mechanism of cellular uptake of 3-hydroxychalcone can occur by passive diffusion [27], for mesoporous silica nanoparticles, absorption occurs by endocytosis. Studies have reported that mesoporous silica nanoparticles have a high affinity for plasma membrane phospholipids, and the adsorption of these molecules on cell surfaces would lead to absorption via endocytosis [28].
As can be seen, the silica samples showed variations in cytotoxic activities. Even with higher incorporation rates, the MSN-APTES-CHO sample showed higher IC50 values (80.98 ± 14.76 to 106.67 ± 7.80 µM) than the MSN-CHO sample (12.93 ± 0.94 to 22.30 ± 4.15 µM), which had a low incorporation rate. These differences in cytotoxic activity can be explained by the different release profiles of the silica samples. As described in the release test, the MSN-CHO sample showed a better release profile, with a greater amount of 3-hydroxychalcone released over time than the MSN-APTES-CHO sample, due to more intense interactions between 3-hydroxychalcone and functionalized silica by hydrogen bonds. This type of interaction affects the release profile of the functionalized nanoparticles and consequently their cytotoxic activity, resulting in higher IC50 values.
In general, the silica samples showed greater cytotoxic activity against AGS and MCF-7 cells than against fibroblast cells (L-929), with the exception of the MSN-APTES and MSN-APTES-CHO samples, which showed greater cytotoxic activity against fibroblast cells than against cancer cells. These results can be explained by the characteristics of the cells tested. The APTES silica surface is normally positively charged under the conditions in which the tests were carried out, promoting greater interaction with the plasma membranes of normal cells, which are negatively charged and attract the functionalized nanoparticles via electrostatic interactions. These interactions may cause a destabilization of the plasma membrane, increasing its permeability and allowing for a greater effect of 3-hydroxychalcone on fibroblast cells. The plasma membranes of tumor cells show changes in the composition and organization of lipids, which cause several changes in cellular functions such as permeability and transport of substances across the plasma membrane [29]. These changes may result in less intense interactions between functionalized silica nanoparticles and tumor cells, which would explain the results found in the cytotoxicity test.