The general idea behind the study is preparation of hydrogel particles capable to incorporate drug molecules (doxorubicin) and be detectable with 3D imaging medical diagnostic tools, e.g. positron emission tomography (PET). The preparative procedure consists of three stages. In the first one, the precursor polystyrene particles undergo a sulfonation reaction, which results in their transformation into hydrogel particles. Next, the hydrogel particles are modified with rice-shaped gallium-containing nanoparticles, optionally doped with Ga-68 radioisotope, through the precipitation of Ga3+ cations with OH− ions, followed by thermal transformation into GaOOH. Finally, the modified hydrogel structures are utilized to incorporate doxorubicin drug molecules based on coulombic cation–anion interaction. The scheme of preparative procedure is shown in Fig. 1.Figure 1Scheme of preparative procedure.Preparation of hydrogel particlesThe first preparative stage was sulfonation of polystyrene particles to yield gel-shell beads. Shown in Fig. 2a and b are scanning electron microscopy (SEM) images of polystyrene (PS) and sulfonated polystyrene (PSS) particles, respectively. While the PS particles appear sharp in the image (their diameter is 3.0 μm), the contours of the PSS beads are diffused. This is not surprising, since it may be assumed the hydrogel shell become collapsed after dryging the sample. We used further transmitted light optical microscopy to image the sulfonated particles (Fig. 2c). First, one can see that the diameter of the particles (4.8 μm) is higher than that of original PS beads. Moreover, one can see much smaller cores (ca. 1.7 μm) within the hydrogel particles. This can be interpreted in such a way, that the sulfonation reaction does not proceed completely, leaving unreacted polystyrene cores.Figure 2Imaging of polymer particles: (a) SEM of polystyrene particles (PS), (b) SEM of sulfonated polystyrene particles (PSS), (c) optical microscopy in the transmitted white light mode of PSS particles, (d) SEM of PSS particles with incorporated doxorubicin.This conclusion is further supported by FTIR data acquired for PS and PSS microparticles, respectively (Supplementary Fig. S1)20,39. Characteristic bands for the sulfonic group appear in the PSS spectrum at 1179, 1039 and 1008 cm−1. The first value can be assigned to the asymmetric stretching of SO2 of the sulfonic group, the next two vibrations are due to symmetric stretching. The band at 3443 cm−1 corresponds to the –OH stretching vibration, absent in the case of polystyrene. The band at 3058 cm−1 corresponds to the stretching vibration of the C–H bond at aromatic carbon, while the bands at 2920 and 2844 cm−1 indicate the presence of asymmetric and symmetric vibrations of the aliphatic CH2 group. Two bands at wavenumbers 758 and 699 cm−1 are characteristic of compounds with a single-substituted aromatic ring, as in the case of polystyrene. However, they are much less intense than in the case of PS, which indicates that the PSS microparticles have not been completely sulfonated which is in accordance with microscopic data. Sulfonation of polystyrene particles is also confirmed by EDS (Energy-dispersive X-ray spectroscopy) analysis (Supplementary Fig. S2).Incorporation of doxorubicin in gel-shell beadsNext, we examined incorporation of doxorubin, a well known cytostatic drug applicable to treat various cancers. Doxorubicin molecule contains an amino group which, when protonated, carries positive charge. The pKa of doxorubicin is 8.4, thus, the molecule is protonated at neutral pH40,41. Since PSS particles contain anionic sulfonic groups, it can be expected that, as a result of electrostatic interactions, the protonated doxorubicin groups (–NH3+) will be attracted to the negatively charged –SO2O- groups. Consequently, doxorubicin molecules should accumulate within the PSS hydrogel layer.Doxorubicin solution was added to the hydrogel particles which resulted in its rapid decoloration (the doxorubicin solution is intensively red). After separation of the particles through centrifugation they have been examined with SEM (Fig. 2d). One can see that the diameter (4.78 ± 0.09 μm) is considerably higher than that of precursor PS beads (and similar to the diameter of PSS beads). Moreover, the contours of the particles are sharp, unlike PSS. It looks like they have swelled due to the absorption of the doxorubicin molecules.Next we studied the doxorubicin-containing beads with optical (fluorescence) microscopy (the drug reveals fluorescent properties42,43,44). Shown in Fig. 3a is fluorescence microscopy image. One can see bright red light projecting from the particles which confirms incorporation of doxorubicin. The corresponding emission spectrum (Fig. 3b) exhibits an emission band with a maximum at 642 nm, which fairly matches the position of solid doxorubicin hydrochloride7. The incorporation of doxorubicin was also confirmed with infrared spectroscopy. Shown in Fig. 3c are FTIR spectra of the particles with incorporated drug (for reference, the spectrum of doxorubicin and PSS beads was included). Comparison of the spectra shows that the spectrum for PSS microparticles with doxorubicin (PSS/Dox) contains signals assigned to both pure doxorubicin and PSS. They are marked with the symbol (*) in colors corresponding to the colors of spectra of individual components. The bands attributable to doxorubicin are found at the following wavenumber values: 1722, 1614, 1579, 1527, 1406, 1280, 1204, 1121, 1078, 966 and 873 cm−1. They are comparable to those for doxorubicin reported in the literature45. On the other hand the PSS bands are found at 1496, 1167 and 1033 cm−1. The signals at the 1442 and 1004 cm−1 wavenumbers are due to both doxorubicin and PSS. The presence of bands from both species in the spectrum confirms the incorporation of doxorubicin in the PSS beads.Figure 3(a) Fluorescence microscopy image of hydrogel particles with incorporated doxorubicin (PSS/Dox), (b) emission spectrum of an individual PSS/Dox particle (excitation 532 nm), (c) FTIR spectra of doxorubicin and PSS and PSS/Dox particles (signals in the PSS/Dox spectrum originating from pure doxorubicin as well as from PSS are marked with the symbol (*) in colors corresponding to the colors of the spectra of the components).In order to determine the amount of incorporated doxorubicin within the gel-shell particles we have recorded the change in the absorbance of the solution after incubations with the beads. A simple calculation, taking into account the known mass of PSS particles, yields the value of 0.96 mg of doxorubicin per 1 mg of PSS, which means that considerably large quantity of the drug, comparable to the mass of PSS, was incorporated.At the same time, the extent to which the incorporated doxorubicin was released into the solution was examined. For this purpose, PSS/Dox particles were incubated in deionized water, and the absorbance of the contacting solution was measured at specific time intervals. Supplementary Fig. S3 shows the curve, indicating that doxorubicin is released gradually at a consistent rate. After 10 h, less than 1.5% of the doxorubicin contained in the particles was released. This result demonstrates that the release of doxorubicin is relatively slow, likely due to the strong interaction between doxorubicin and the sulfonic groups in PSS. The remarkably high sorption capacity of PSS, together with the retarded release of the drug, is promising from the perspective of application in controlled drug delivery.Incorporation of GaOOH in gel-shell beadsIn a set of parallel experiments we examined the possibility to incorporate gallium-containing compounds within the hydrogel beads. It was expected that Ga3+ cations due to positive charge would be accumulated within the hydrogel shells of the particles (as the sulfonic groups are negatively charged). The cations then would be transformed into non-soluble compounds through reaction with anions, e.g. hydroxyl anions.Gallium(III) oxide hydroxide is known to be formed through dehydration of Ga(OH)3 at elevated temperature46. Therefore, the idea was to precipitate gallium ions with OH− and then transform it to GaOOH. However, we have noticed that the Ga(OH)3 did not precipitate effectively at stoichiometric Ga3+/OH− ratio, which is likely due to the presence of numerous equilibria in the solution47. In order to determine the ratio at which both reagents should be mixed to obtain the largest amount of Ga(OH)3, nephelometric titration of Ga3+ with NaOH or NH3 solutions was performed.Shown in Fig. 4a are titration curves of Ga3+ solution. One can see that the highest intensity of scattered light is observed at 1.77:1 for NaOH/Ga3+ and 1.20:1 for NH3/Ga3+ molar ratios. Increasing the amount of the base in both cases results in decrease of the intenstity which can be interpreted as gradual dissolution of the precipitate. These ratios have been selected as optimal in further experiments. Nevertheless, as Ga(OH)3 is prone to dissolution, it should be transformed to GaOOH. To check this possibility, the gallium hydroxide precipitate was heated, and then, after separating the precipitate, subjected to X-ray diffraction measurement. Shown in Fig. 4b is diffraction pattern of the obtained product. All observed diffraction peaks correspond to gallium hydroxide oxide, confirming full transformation of Ga(OH)348,49 (the structural parameters50,51,52 calculated from the the XRD data are included in Supplementary Information; Tables S1 and S2). The morphology of the fabricated GaOOH was also examined using scanning electron microscopy. When NaOH was used in the synthesis, GaOOH crystals were mostly cuboidal, regular shape, with average size of 320 nm × 144 nm (Supplementary Fig. S4). On the other hand, the particles prepared through precipitation with NH3 were longer and thinner than their NaOH counterparts (ca. 375 nm × 103 nm), resembling rice grains (Fig. 4c).Figure 4(a) Nephelometric titration curves of gallium nitrate solution with ammonia solution (blue line) and sodium hydroxide solution (red line), (b) diffraction pattern of rice grain-like GaOOH nanoparticles (prepared from NH3 solution), (c) SEM image of rice grain-like GaOOH nanoparticles (prepared from NH3 solution), (d) SEM image of PSS/GaOOH particles (prepared from NH3 solution).The next aim was to incorporate GaOOH nanoparticles into hydrogel beads. The PSS structures were conditioned with a Ga3+ ion solution, followed by the addition of a base solution and heating. It was expected that Ga3+ ions would accumulate in the hydrogel shell as a result of coulombic interactions with sulfonic groups. The addition of OH- ions would result the precipitation of Ga(OH)3, followed by transformation into GaOOH due to increased temperature. In the case of the synthesis using NaOH, some deposit was indeed obtained on the surface of the PSS particles (Supplementary Fig. S5). Unfortunately, it was not regular. Some PSS particles were completely covered with gallium nanorods, while some were practically uncoated. On the other hand, the use of NH3 solution results in preparation of beads regularly coated with GaOOH particles (Fig. 4d).It can be observed that the particles of the gallium compound have the shape of rice grains, even though they are slightly longer and thicker compared to those prepared in the absence of PSS (593 nm × 155 nm). It seems that the nanoparticles are attached to the surface of the PSS particles, but they do not penetrate deeply. The diameter of the PSS/GaOOH particles is 3.35 ± 0.17 μm, thus, it is slightly higher than the size of the starting PS particles. Some GaOOH particles non-attached to PSS are also visible (it seems they were detached from the PSS matrix during preparative procedures, or they are the result of insufficient separation). Additional detailed analysis with EDS, further confirming the above conclusions, is included in the Supplementary Information (section: EDS analysis of PSS/GaOOH beads, Figs. S6-S9).In order to determine the efficiency of the incorporation of GaOOH particles in PSS, the Ga-68 isotope was used to label the structures in order to perform radiometric measurements. For this purpose, a solution of stable gallium ions was prepared with the addition of Ga-68 radioisotope with an activity of approximately 10 MBq. The synthesis was then carried out according to the experimental procedure described above, and the supernatants were collected to determine their activity. On this basis, the yield of the preparative process was calculated. Figure 5a shows a histogram illustrating the efficiency of GaOH incorporation at individual preparatory stages. The data show that during subsequent centrifugations, GaOOH rice-shaped nanoparticles unbound to hydrogel beads are removed to some extent, but the final efficiency is relatively high, reaching ca. 77%.Figure 5(a) Histogram showing the 68GaOOH content in a sample of hydrogel particles during subsequent preparative stages, (b) thermogravimetric curves of PSS (i) and PSS/GaOOH (ii) particles; (c) SEM image of PSS/GaOOH/Dox particles, (d) gamma-ray spectrum of PSS/68GaOOH/Dox particles.In order to determine the content of GaOOH in the composite we employed thermogravimetric analysis (TGA). Shown in Fig. 5b is thermogram of PSS/GaOOH particles (for reference, the thermogravimetric curve of PSS beads is also shown). One can see, that the decomposition of the PSS/GaOOH beads starts above 100 °C. The mass gradually decrease in several decomposition steps reaching the final value of 43%. On the other hand, the decomposition of PSS yields a residue of 12%. It can therefore be assumed that heating PSS/GaOOH results in the decomposition of PSS, but with the formation of an indecomposable residue. At the same time, GaOOH decomposes into gallium oxide Ga2O3 with the release of water molecules.53 Based on a simple calculation taking into account the above data, it can be estimated that GaOOH content in the composite is ca. 0.64 mg per 1 mg of PSS.Fabrication of Dox/GaOOH gel-shell beadsWe aimed further to obtain PSS particles modified with both doxorubicin and GaOOH (PSS/Dox/GaOOH). For this purpose, first the PSS particles were modified with a gallium compound and then with doxorubicin. The obtained hydrogel structures were examined using scanning electron microscopy. Figure 5c and Fig. S10 (Supplementary Information) show SEM images at different magnifications of hydrogel particles modified with GaOOH and doxorubicin. The diameter of the particles is 3.81 ± 0.09 μm which is somewhat higher than PSS beads with GaOOH (see Fig. 4d), but lower than PSS beads with incorporated doxorubicin (Fig. 2d). This indirectly indicates the incorporation of doxorubicin (fully unambiguous confirmation of the presence of doxorubicin is provided by fluorescence microscopy; the corresponding microscopic images of PSS/GaOOH/Dox particles, in both white light and fluorescence mode, are shown in Supplementary Information, Fig. S11).On the other hand, radiometric measurements (analogous to those presented in the previous chapter) show that the addition of doxorubicin causes a decrease in the amount of GaOOH by only 4%. This indicates that the incorporation of doxorubicin molecules occurs in a gentle manner and does not substantially result in the removal of GaOOH nanoparticles.In addition to determining incorporation efficiency, the use of the gallium-68 isotope may have additional applications. Ga-68 is used in PET tomography, so the particles modified with the isotope could be monitored using this technique after administration into the body, which is especially important if they carry a drug (here: doxorubicin). Shown in Fig. 5d is gamma-ray spectrum of the PSS/68GaOOH/Dox sample. The spectrum shows a signal at 511 keV attributed to gamma rays resulting from the annihilation of positrons emitted by Ga-68 contained in the hybrid particles. The Ga-68 activity used here is to demonstrate feasibility of the concept. However, it should be noted, that for PET imaging, significantly higher radioisotope activities would be required.Cytotoxicity of PSS/GaOOH/Dox particlesIn vitro cytotoxicity studies of PSS/GaOOH/Dox beads were performed on the MDA-MB-213 breast cancer cell line. These cells represent a subtype of triple-negative breast cancer that lacks progesterone and estrogen receptors, which makes them insensitive to treatment with antiestrogens, commonly used in breast cancer therapy. The reference cell line was MCF-10A—non-tumorigenic epithelial cells derived from the MCF-10A mammary gland.Two tests were used to assess antitumor activity: MTT and CVS. The MTT assay is one of the most popular tests assessing the activity of a potential anticancer drug and is an indirect test. It is based on the ability of living cells to reduce a tetrazolium salt to formazan. The obtained formazan has a purple color, the intensity of which is proportional to the amount of the product formed and indirectly proportional to the number of living cells. The second assay used was the direct CVS test, independent of cellular metabolism. The CVS test assesses the amount of dye absorbed by the cell, which depends on the DNA content in the culture and allows for estimating the number of living cells.Figure 6a, b shows the results of viability studies of MDA-MB-231 breast cancer cells after 72-h incubation in the presence of the hybrid particles.Figure 6MTT and CVS assays evaluating the effect of PSS/GaOOH/Dox particles on cancer and non-tumorigenic epithelial cell line cultures: (a) MDA-MB-231 cancer cells, (b) MCF-10A non-tumorigenic epithelial breast cells (right panel), (c) HT-29 colon cancer cells, (d) CRL-1790 non-tumorigenic epithelial colon cells.The histograms for MDA-MB-21 cancer cells and non-tumorigenic epithelial MCF-10A cells incubated in the presence of PSS/GaOOH/Dox beads show that as the concentration of particles increases, cell viability decreases. When comparing histograms for cancer and non-tumorigenic epithelial cells at high concentrations, a slightly higher viability of non-tumorigenic MCF-10A cells that serves as a model of normal cells compared to cancer cells was noted. For the highest concentration of 15 ppm, the viability of cancer cells drops to approximately 15%, and for the model normal cells to approximately 18%.Analogous in vitro cytotoxicity studies were also performed in the HT-29 colorectal cancer cell line, which in in vitro cultures shows a morphology typical of epithelial cells. The reference cell line was CRL-1790, isolated from normal human colon tissue, representing a model of normal colon cells. Figure 6c, d shows histograms for the HT-29 and CRL-1790 cell lines after 72 h of incubation in the presence of particles. For HT-29 cancer cells, it was observed that with increasing particle concentration, cell viability decreased. The cytotoxic effect of the examined structures on colon cancer HT-29 cells was the strongest in the concentration range of 1.67–15 ppm, where cell viability in this range did not exceed 15% (CVS test) and approximately 23% (MTT test). It should be noted that the highest concentration of 15 ppm turned out to be 3 times less toxic for CRL-1790 cells (MTT test) and 4 times less toxic for CRL-1790 cells (CVS test) than for HT-29 cancer cells. The results clearly indicate reduced toxicity towards non-cancerous cells.To quantitatively investigate the cytotoxic effect of the particles, the IC50 values were determined (the concentration at which the number of cells in the culture decreases to 50% compared to the control). Since the data for the particles are expressed in ppm (which complicates comparison with the data for doxorubicin in solution), they were recalculated to an effective molar concentration, taking into account the previously determined doxorubicin content in the particles (vide supra). The IC50 values are presented in Table 1 (the IC50 values reported in the literature for doxorubicin not encapsulated in the carrier are also included).
Table 1 IC50 values determined using MTT and CVS assays for MDA-MB-231, MCF-10A, HT-29, and CRL-1790 cell lines after 72 h of incubation with PSS/GaOOH/Dox.The IC50 values determined for cells incubated with PSS/GaOOH/Dox particles in the MTT assay for breast cancer cells (MDA-MB-231) was 0.40 ppm (0.66 μM), which was almost 4.5 times higher than that for non-tumorigenic MCF-10A cells. In the CVS test, the IC50 value was 0.13 ppm (0.22 μM) for cancer MDA-MB-231 cells, which was 3 times higher than for MCF-10A cells. These values are rather undesirable because they indicate that the cytotoxicity of doxorubicin is higher for MCF-10A cells than for MDA-MB-231 cells. However, similar behavior of doxorubicin towards MDA-MB-231 and MCF-10A cells is reported in the literature (IC50 values of 0.125 µM and 0.025 µM, respectively)54.For HT29 colon cancer cells, the IC50 value for the MTT assay is 0.65 ppm (1.08 μM), and for CVS, it is 0.23 ppm (0.38 μM), which is approximately 40% of the IC50 value for colon CRL-1790 cells (the IC50 values of doxorubicin in solution for HT-29 cells is 0.750 μM (MTT)55; we have found no literature data for CRL-1790 cells). This is a much better result than in the case of breast cells, where cancer cells turned out to be less sensitive to the action of cytostatics (whether encapsulated in a carrier or free) compared to non-cancerous cells.Qualitative microscopic observation of cell cultures was also performed. Staining cells with propidium iodide (PI) and fluorescein diacetate (FDA) allows to visualize live and dead cells and to assess the processes caused by the action of the examined polymer particles in the cells. Figure 7 shows microscopic images of stained MDA-MB-231 and MCF-10A cells after a 72-h incubation in the presence of the PSS/GaOOH/Dox structures.Figure 7Confocal microscopy images of stained MDA-MB-231 cells (A, B) and MCF-10A cells (C, D) after 72 h of incubation in the presence of PSS/GaOOH/Dox particles at concentrations of 15, 1.67, 0.18, and 0.02 ppm; transmitted light (A, D) and fluorescence modes (B, C).For concentrations of 1.67 ppm and 15 ppm, no live or dead cells were recorded for both MDA-MB-231 and normal MCF10A breast cell lines. The obtained result may indicate very high cytotoxicity of the tested structures towards cells in this concentration range. In microscopic images of non-cancerous MCF10A cells incubated with 0.18 ppm, dead cells (red) were observed, which were not visible for MDA-MB-231 cancer cells, which may indicate a different mechanism (cytotoxic in MCF-10A and cytostasis or growth arrest in MDA-MB-231 cells) of action of the beads on individual cell lines56.Microscopic examination was also performed on HT-29 (A, B) and CRL-1790 cell cultures. (Fig. 8). For high concentrations of the studied particles (15 ppm, 1.67 ppm), no live HT-29 cancer cells were observed (only hydrogel particles in the form of red spots were recorded). A different scenario was observed for colon CRL-1790 cells, where, in addition to PSS/GaOOH/Dox particles, numerous living cells were also seen. The findings align with the results from cytotoxicity assessment using MTT and CVS, demonstrating a more pronounced cytotoxic effect on colon cancer HT-29 cells as compared to CRL-1790 normal cells. This is a very important and promising result from the point of view of the potential therapeutic effectiveness of the particles.Figure 8Confocal microscopy images of stained HT-29 cells (A, B) and CRL-1790 cells (C, D) after 72 h of incubation in the presence of PSS/GaOOH/Dox particles at concentrations of 15, 1.67, 0.18, and 0.02 ppm; transmitted light (A, D) and fluorescence modes (B, C).The presented results indicate that the PSS/GaOOH/Dox particles exhibit selective cytotoxicity. Increasing the selectivity of the cytotoxic effect by encapsulating doxorubicin is now a priority due to its serious side effects (cardiotoxicity), which prevent safe and effective treatment.57 In addition, the results indicate that efficient incorporation of gallium allows for imaging and tracking the formulation using PET. Such theranostic approach is considered as a beneficial option for the therapy of cancer. Pan et al.58 shown that doxorubicin may be incorporated in drug delivery vehicles together with iron oxide nanoparticles that enable MRI imaging. Similar approach presented Hasannia et al.59. It was shown that doxorubicin may be effectively delivered to the tumor when incorporated in peptosomes hybridized with gold nanorod that allow CT imaging.
Physicochemical characterization and potential cancer therapy applications of hydrogel beads loaded with doxorubicin and GaOOH nanoparticles
