Synthesis of nanostructured calcium carbonate/methotrexate@silica and its application in cancer therapy

Abstract

Biomedical applications of nontoxic amorphous calcium carbonate (ACC) have mainly been restricted because of its aqueous instability. Herein, we report the successful synthesis of highly stable ACC–methotrexate (MTX)@SiO₂ nanospheres in vitro for use in cancer therapy. Further, vaterite–MTX@SiO₂ nanospheres were also prepared for comparison. In our synthesis procedure, ACC–MTX and vaterite–MTX were firstly prepared at different pH values, and then SiO₂ layers were subsequently deposited by the well-known Stöber method. The results indicated that the special structure of ACC–MTX@SiO₂ presents better controlled-release and results in efficient death of cancer cells, thus showing its great potential as a desirable chemotherapeutic system for cancer therapy.

---

1. Introduction

Calcium carbonate (CaCO₃), as one of the most widely existing biominerals produced by organisms, has shown promising potential for the development of smart carriers for drugs due to its excellent biocompatibility, biodegradability, and simple chemical composition.¹ To date, crystallized CaCO₃ has been widely used as a successful carrier for the delivery of drugs, genes, and proteins.² However, the main problem of its micrometer size limits its application in nanomedicine. In contrast to the crystallized form, amorphous calcium carbonate (ACC) can be readily obtained on a nanoscale.³ Studies have shown that biomaterials based on nanosized particles exhibit enhanced resorbability and much higher bioactivity than micron-sized ones.⁴ However, nanosized ACC is not a stable phase in thermodynamic terms, thus its high aqueous instability including the ease of dissolution and phase transition, has limited the development on medicine.^5–7 Further, a comprehensive comparison of ACC and crystallized CaCO₃ in the delivery efficient and anticancer effect has not been reported before.

In this paper, nanosized ACC-drug and vaterite (a kind of crystallized CaCO₃)-drug hybrids were firstly synthesized, and then a silica-based core–shell structure was introduced to improve the stability. Typically, this kind of system usually consists of a CaCO₃-drug core and a permeable shell, and the permeable shell should protect the core from unexpected drug leakage. Such similar structure of ACC–DOX@SiO₂ was also prepared by Zhao group, however the aqueous instability still restricted its further utilization.⁸ Herein, methotrexate (MTX), which can effectively deactivate the metabolism of diseased cells through programmed cell death or apoptosis, was chosen as the anticancer drug. More importantly, the formation mechanism and anticancer effect was explored emphatically.

2. Experimental section

2.1 Synthesis of vaterite–MTX and ACC–MTX hybrids

A typical synthesis of vaterite–MTX and ACC–MTX was described as follows: CaCl₂·2H₂O (200 mg) and MTX (5 mg) was dissolved in 100 mL of absolute ethanol, NH₃·H₂O (25%) was then used to regulate the pH value to 4.5 or 8.5. After that, the mixture was transferred into a glass bottle and covered by parafilm with several pores. Then, the bottle was left in a desiccator along with two glass bottles of ammonia bicarbonate (NH₄HCO₃) at 35 °C. After vapor diffusion reaction for 2–3 days, the yellow products were centrifuged and redispersed in absolute ethanol. At last, the final samples were named as a and c according to the initial pH value of 4.5 and 8.5.

2.2 Synthesis of vaterite–MTX@silica and ACC–MTX@silica nanoparticles

Firstly, 2 mL of vaterite–MTX and ACC–MTX solution (2 mg mL⁻¹) were re-dispersed in 20 mL of absolute ethanol under stirring for 0.5 hour, and then 0.2 mL of tetraethyl orthosilicate (TEOS) was added. Secondly, the mixture was continuously stirred for another 2 hours. Then 0.5 mL of NH₃·H₂O (25%) was added, and the mixture was continuously stirred overnight. At last, the mixture was centrifuged and washed with absolute ethanol for three times. The final samples were named as b, d according to the original precursors of a and c.

2.3 In vitro drug release

In vitro drug release was performed as follows: 0.02 g of nanostructures was added into 500 mL of phosphate buffer solution (pH = 7.4) in a closed glass bottle at a constant temperature of 37 °C. At selected time after addition of the hybrids, 4 mL of solution was withdrawn and centrifuged, part of the supernatant was used for the measurement. Dissolution medium was maintained at constant volume by replacing the samples with a fresh dissolution medium. Then, the concentration of MTX was measured by UV-vis spectroscopy at λ_max = 306 nm, at last the release profiles were plotted as the relative release percentages of MTX against time. The data were collected in triplicate and presented in Fig. 2A.

2.4 In vitro bioassay

Human lung adenocarcinoma cells (A549) and mouse adrenal pheochromocytoma cells (PC-12) were used here. As a typical of benign tumour, PC-12 was usually regarded as normal cells. Cells were routinely cultured as ref. 9. For cell proliferation and viability study, cells were seeded onto 96-well plates, and then the cells were incubated overnight at 37 °C under a 5% CO₂ atmosphere. After that the medium in the wells was replaced with fresh medium containing examined samples, and further incubated for 24 h. The effect of the examined samples on cell proliferation was determined using MTT (a yellow tetrazole) assay.⁹ Absorbance was recorded at 570 nm on a microplate reader (Thermo MK3, USA). The MTT assays were also performed with the cells being cultured with different incubation time. As for the phase contrast images, after the cell attachment of 24 hours, the medium in the wells was replaced by the examined sample. Then the cells were observed using a light microscope with a 40× objective lens at different incubation time.

3. Results and discussion

As shown in Fig. 1, vaterite–MTX (Fig. 1a) with a diameter of about 80 nm were generated in the acidic environment, and ACC–MTX (Fig. 1c), which obtained under pH = 8.4, exhibited an average particle size of about 350 nm. Later on, silica shell was then deposited to form vaterite–MTX@SiO₂ (Fig. 1b) and ACC–MTX@SiO₂ (Fig. 1d). After coating, the diameter increased accordingly, with the size of about 130 and 450 nm. Noticeably, the particles exhibit excellent monodispersity after coated with silica shells. Further, the distribution of the particle size was also drawn from the corresponding size distributions from SEM statistical results, locating in the top right-hand corner of Fig. 1. At last, the polydispersity index (PDI) was also been determined by dynamic light scattering (DLS) technology and the values were 0.33, 0.25, 0.07 and 0.06 for samples a–d, respectively.

Fig. 1 TEM images of nanostructures (a) sample a, (b) sample b, (c) sample c, (d) sample d. (e) Schematic procedure for the formation of ACC–MTX@SiO₂ nanoparticles, (f) XRD patterns of pristine SiO₂ and samples a–d. (g) FTIR spectra of pristine SiO₂ and samples a–d.

Here, a possible mechanism was proposed for the formation and growth of CaCO₃–MTX@SiO₂ structure, as illustrated in Fig. 1e. Firstly, MTX molecules were bound or chelated with Ca²⁺ to form Ca²⁺–MTX structure.¹⁰ Secondly, the obtained Ca²⁺–MTX would react with CO₃²⁻ anions to form CaCO₃–MTX NPs. Finally, the obtained CaCO₃–MTX NPs were subsequently coated with SiO₂ by the well known Stöber method, and then CaCO₃–MTX@SiO₂ was formed.

Further, these four samples were also characterized by XRD and IR. For XRD analysis, ACC–MTX exhibited no distinct diffraction peaks, and two typical broad lumps at 2θ angle of approximately 30° and 45° represented the formation of ACC.¹¹ On the contrary, vaterite–MTX fitted well with the International Centre for Diffraction Data (ICDD) database of PDF#72-0506, confirming the existence of vaterite. What's more, the weak reflection plane at about 23° for samples b and d (Fig. 1f) belonged to the diffraction pattern of the silica layer. The corresponding Fourier transform infrared (FTIR) spectra were also exhibited in Fig. 1g. For all samples, the bands at 1460, 1090 and 877 cm⁻¹ were due to the ν₃ asymmetric vibration, ν₁ symmetric and ν₂ asymmetric vibration of CO₃²⁻, respectively.^12,13 As for vaterite–MTX hybrid, the formation of vaterite was confirmed by the presence of ν₄ symmetric vibration at 746 cm⁻¹, which is characteristic of crystalline calcium carbonate (vaterite) phase only.¹⁴ In addition, the peak centered at 1095 cm⁻¹ belonged to the deformation mode of Si–O–Si, which appeared in both ACC–MTX@SiO₂ and vaterite–MTX@SiO₂.

In vitro release profiles of four samples had been illustrated in Fig. 2A, which obtained in phosphate buffer saline (PBS) of pH = 7.4. A rapid release of MTX occurred at the initial stage, which followed by a relatively slower one. Such two-step and prolonged release behavior could benefit the therapeutic treatment, as the initial release rapidly provided a therapeutic dose, and the subsequent sustained release allowed maintenance of this dose over a long period of time.¹⁵ In details, the t_0.5 (the time for release fraction of 50%) is 10 min, 4 h, 30 min, 7 h for samples a, b, c and d, respectively. The burst release fractions (defined as the release fraction in the initial 5 h) are 69%, 54%, 59% and 50%, respectively. It is obvious that silica-coated particles present much better controlled-release property than those of non-coating ones. We think that core–shell structure might experience two-step of the drug-release process, and the first one is that MTX dissociates from the CaCO₃–MTX complex by an acid-induced decomposition of the core. The second one is that MTX passes through the silica shell and enters into the external environment. Consequently, the silica-coating structure exhibited much better controlled-release than that of CaCO₃–MTX NPs.

Fig. 2 (A) Release profiles of samples a–d. Error bars represent standard error (n = 3). (B) Plots of different kinetic models of Higuchi equation, Bhaskar equation, R–P equation and parabolic diffusion equation for the release of MTX from samples a–d.

There are about two kinds of mechanism on the drug release, i.e., ion-exchange process and dissolution of outer layers in buffer solution, and the release process could be controlled by any of the steps. To gain more insights into the mechanism of drug release, four types of kinetics models were exploited:¹⁶ the Higuchi model (eqn (1)), the Bhaskar equation (eqn (2)) and the Ritger–Peppas equation (eqn (3)) and parabolic diffusion equation (eqn (4)):

M_t/M_∞ = k_Ht^0.5 (1)

lg(1 − M_t/M_∞) = −k_Bt^0.65 (2)

M_t/M_∞ = ktⁿ (3)

(M_t/M_∞)/t = k_Pt^−0.5 + b (4)

In the above equations, M_t/M_∞, t, k are the fractional drug release, release time and the corresponding release rate constant, respectively, and a and b are constants whose chemical significance is not clearly resolved.

On the basis of the four different kinetic models, the fitting results of drug release profiles are given in Fig. 2B, and the corresponding linear correlation coefficients (R) and n values obtained from the fittings are summarized in Table 1. It can be seen from Fig. 2B and Table 1, the Higuchi and the Bhaskar equation are not suitable to explain the release process reflected by the fact that modeling data points do not form a straight line and small linear correlation coefficients (R < 0.96). As for samples a and c, parabolic diffusion model fits the release data much better (R > 0.98). While for samples b and d, the fitting coefficients of Ritger–Peppas model are the highest (R > 0.98). Usually, the R–P equation (eqn (3)) used to explain drug diffusion and dissolution of layers. The value of n < 0.45 corresponds to the drug diffusion control, which based on ion exchange process; n > 0.89 is attributed to the dissolution of layers; 0.45 < n < 0.89 is due to the cooperation of drug diffusion and layer dissolution. The parabolic diffusion equation (eqn (4)) elucidates that the release process is controlled by an external surface/edge diffusion process.^17,18 For samples a and c, they can be described by external surface diffusion as the rate-controlling step, and MTX anions are diffused into the medium solution from the surface of CaCO₃ particles, owning a rapid drug release process. Whereas, the release process of samples b and d, with release value of n between 0.1861 and 0.2265 (n < 0.45), belongs to the drug diffusion via ion exchange, based on the MTX anions diffusion from the interlayer CaCO₃ core to the surface of particles, which takes a rather longer release time.

Table 1 Fitting parameters of different kinetic models for the release of MTX from different samples

Samples	Higuchi	Bhaskar	Ritger–Peppas		Parabolic diffusion
	R	R	R	n	R
a	0.823	0.831	0.972	0.177	0.980
b	0.842	0.864	0.981	0.070	0.961
c	0.918	0.912	0.971	0.061	0.981
d	0.842	0.856	0.982	0.054	0.958

---

The cell cytotoxicity and anticancer efficiency of samples a–d together with MTX are evaluated by bioassay test. Firstly, the cytotoxicity evaluation of samples a–d and MTX was carried out on both normal (PC-12) and cancer cell (A549) lines, and cell viability was evaluated at a concentration of 100 μg mL⁻¹ after 24 h of incubation (in Fig. 3K and L). As displayed in Fig. 3K, ACC–MTX and ACC–MTX@silica had little toxic effect on normal cells, while vaterite–MTX and vaterite–MTX@silica were fairly toxic and block viability of the cells obviously.^19,20 As expected, MTX was fairly toxic on normal cells. However, as for cancer cells of A549 (Fig. 3L), MTX showed little effect on them. From Fig. 3L, silica-coating nanoparticles exhibited enhanced anticancer effect compared with those of non-coating ones, especially after long-time incubation. In details, the cell viability of ACC–MTX, vaterite–MTX and MTX did not change much with the increase of time. However, the cell viability of vaterite–MTX@silica and ACC–MTX@silica showed a rapid decrease as the extension of time, probably due to the enhanced controlled-release effect after silica coating. Considered from the above results, the formation of ACC–MTX@silica structure not only can decrease the side effects on normal cells but also enhance the anticancer effect on the meanwhile, showing a kind of potential treatment candidate for cancer therapy. Furthermore, a contrast study on cancer and non-cancerous cells from Fig. 3K and L demonstrated that samples a–d have obvious suppression effect on cancer cells than normal cells.

Fig. 3 Morphology changes of A549 cells treated with various samples at the concentration of 100 μg mL⁻¹ (A) DMEM (control), (B) free MTX, (C) sample a, (D) sample c, (E) sample b and (F) sample d. The intracellular location by phase contrast images are measured at different incubation time for sample d at the concentration of 100 μg mL⁻¹ (G) 0 h control, (H) 0.5 h, (I) 1 h, (J) 2 h. (K) Comparisons of cell viabilities (PC-12) for MTX and samples a–d at various incubation time at the concentration of 100 μg mL⁻¹. (L) Comparisons of cell viabilities (A549) for MTX and samples a–d at various incubation time at the concentration of 100 μg mL⁻¹.

As a supplement to MTT assay, morphology changes of A549 cells treated with DMEM, pure MTX (100 μg mL⁻¹), samples a–d (100 μg mL⁻¹) after 72 h of incubation were displayed in Fig. 3A–F. The cells treated with DMEM showed normal cellular morphology of A549, indicating that living cells look like spindles (Fig. 3A). The cell form was slightly changed after treated with MTX, and the co-existence of spindle and circular cells mean that part of A549 cells were withered or close to death after the uptake of MTX (Fig. 3B). As expected, addition of samples a and c had noticeable effect on A549 cells, revealed by lots of circular cells, displaying that most of A549 cells died (Fig. 3C). Notably, it was worthy to reveal that almost all cells treated with samples b and d exhibited circle in morphology, presenting that almost all cells underwent apoptosis (Fig. 3D). In order to further explore the uptake process of ACC–MTX@silica nanostructure into A549 cells, the intracellular location observed by phase contrast images were also carried out at different incubation time for sample d (see Fig. 3G–J). The results demonstrated that cells morphology changed dramatically with the incubation time. After 0.5 h of incubation, the cells became slightly shrank and a small amount of NPs accumulated at the nearby of cell membrane (Fig. 3H). When the duration increased to 1 h, most of particles were internalized into the cytoplasm, such rapid amassing also suggested a distinct mechanism, i.e., the endocytosis effect may be involved in the uptake of ACC–MTX@silica.²¹ At the same time, cell morphology became irregular (Fig. 3I). After 2 h of incubation, ACC–MTX@silica nanoparticles still laid free in the cytoplasm, but the cells shrank into an ellipse (Fig. 3J). For comparisons, the intracellular location of A549 cells without adding any drug was also shown in Fig. 3G. The distinct morphology changes presented that the core–shell structure of ACC–MTX@silica is a kind of promising treatment platform for cancer therapy.

Acknowledgements

The authors are grateful for the financial support of this project by the National Natural Science Foundation of China (21073093), the Foundation of Key Laboratory for Palygorskite Science and Applied Technology of Jiangsu Province (HPK201405), Fund of Priority Academic Program Development of Jiangsu Higher Education Institutions and Jiangsu Collaborative Innovation Center of Biomedical Functional Materials.

References

1. L. Addadi, S. Raz and S. Weiner, Taking Advantage of Disorder: Amorphous Calcium Carbonate and Its Roles in Biomineralization, Paediatric and Perinatal Epidemiology, 2003, 15, 959–970.
2. D. V. Volodkin, V. K. Regine and M. Helmuth, Pure Protein Microspheres by Calcium Carbonate Templating, Angew. Chem., 2010, 49, 9258–9261.
3. G. B. Cai, G. X. Zhao, X. K. Wang and S. H. Yu, Synthesis of Polyacrylic Acid Stabilized Amorphous Calcium Carbonate Nanoparticles and Their Application for Removal of Toxic Heavy Metal Ions in Water, J. Phys. Chem. C, 2010, 114, 12948–12954.
4. J. Wang, J. S. Chen and J. Y. Zong, et al., Calcium Carbonate/Carboxymethyl Chitosan Hybrid Microspheres and Nanospheres for Drug Delivery, J. Phys. Chem. C, 2010, 114, 18940–18945.
5. J. J. J. M. Donners, B. R. Heywood and E. W. Meijer, et al., Amorphous calcium carbonate stabilised by poly(propylene imine) dendrimers, Chem. Commun., 2000, 1937–1938.
6. A. W. Xu, Q. Yu, W. F. Dong and M. Antonietti, et al., Stable Amorphous CaCO₃ Microparticles with Hollow Spherical Superstructures Stabilized by Phytic Acid, Adv. Mater., 2005, 17, 2217–2221.
7. M. Faatz, F. Gröhn and G. Wegner, Amorphous Calcium Carbonate: Synthesis and Potential Intermediate in Biomineralization, Adv. Mater., 2004, 16, 996–1000.
8. Y. Zhao, L. N. Lin and Y. Lu, et al., Templating synthesis of preloaded Doxorubicin in hollow mesoporous silica nanospheres for biomedical applications, Adv. Mater., 2010, 22, 5255–5259.
9. D. Y. Tian, W. Y. Wang and S. P. Li, et al., A novel platform designed by Au core/inorganic shell structure conjugated onto MTX/LDH for chemo-photothermal therapy, Int. J. Pharm., 2016, 505, 96–106.
10. C. F. Dai, S. P. Li and X. D. Li, Synthesis of nanostructured methotrexate/hydroxyapatite: morphology control, growth mechanism, and bioassay explore, Colloids Surf., B, 2015, 136, 262–271.
11. J. D. Rodriguez-Blanco, S. Shaw and L. G. Benning, How to make ‘stable’ ACC: protocol and preliminary structural characterization, Mineral. Mag., 2008, 72, 283–286.
12. J. D. Rodriguez-Blanco, S. Shaw and L. G. Benning, The kinetics and mechanisms of amorphous calcium carbonate (ACC) crystallization to calcite, via vaterite, Nanoscale, 2011, 3, 265–271.
13. X. Wang, K. Rui and X. Pan, et al., Role of ovalbumin in the stabilization of metastable vaterite in calcium carbonate biomineralization, J. Phys. Chem. B, 2009, 113, 8975–8982.
14. K. Naka, Effect of anionic dendrimers on the crystallization of calcium carbonate in aqueous solution, C. R. Chim., 2003, 6, 1193–1200.
15. C. F. Dai, D. Y. Tian and S. P. Li, et al., Methotrexate intercalated layered double hydroxides with the mediation of surfactants: mechanism exploration and bioassay study, Mater. Sci. Eng., C, 2015, 57, 272–278.
16. X. Q. Zhang, M. G. Zeng and S. P. Li, et al., Methotrexate intercalated layered double hydroxides with different particle sizes: structural study and controlled release properties, Colloids Surf., B, 2014, 117, 98–106.
17. R. Bhaskar, R. S. R. Murthy, B. D. Miglani and K. Viswanathan, Novel method to evaluate diffusion controlled release of drug from resinate, Int. J. Pharm., 1986, 28, 59–66.
18. L. Serra, J. Domenech and N. A. Peppas, Drug transport mechanisms and release kinetics from molecularly designed poly(acrylic acid-g-ethylene glycol) hydrogels, Biomaterials, 2006, 27, 5440.
19. H. L. Karlsson, G. Johanna and C. Pontus, et al., Size-dependent toxicity of metal oxide particles-a comparison between nano- and micrometer size, Toxicol. Lett., 2009, 188, 112–118.
20. L. Nastassja, C. Vicki and D. Rebekah, Cytotoxicity of nanoparticles, Small, 2008, 4, 26–49.
21. Y. Svenskaya, B. Parakhonskiy and A. Haase, et al., Anticancer drug delivery system based on calcium carbonate particles loaded with a photosensitizer, Biophys. Chem., 2013, 182, 11–15.

---

Footnote

† These authors contributed equally to the work.

---