Glutathione-triggered release of model drug molecules from mesoporous silica nanoparticles via a non-redox process

Abstract

Model drug-loaded mesoporous silica nanoparticles (MSNs) that are responsive to the pH rather than the redox changes normally related to glutathione (GSH) are prepared using surfactant-free MSNs as precursor. The nanoparticles are successfully shown to serve as GSH-triggered release vehicles for the guest molecules. Unlike the generally known GSH-based redox-triggered drug release action, a non-redox, ionic process associated with GSH and GSH-induced pH change enables the MSNs to release their guest molecules.

---

Owing to their high surface area, tunable nanoscale pores, chemically functionalizable surfaces, and ability to enter cells,¹ mesoporous silica nanoparticles (MSNs) can host a myriad of bioactive and drug molecules with different sizes and hydrophilicity/hydrophobicity, and carry them into cells. Hence they have attracted a great deal of attention for their potential applications as drug, protein, and gene delivery vehicles and in nanomedicines. In recent years, a number of them have also been designed to serve as on-demand drug release systems to unleash payloads of drugs under different stimuli such as changes in pH,² temperature,³ redox conditions,⁴ wavelengths of light,⁵ concentrations of glucose⁶ or magnetic activity,⁷ or in the presence/absence of enzymes.⁸ Materials such as these are quite appealing for biomedical applications because they can significantly minimize the side effects of numerous drugs or improve drugs' therapeutic effects.^1a,1c

One of the most physiologically pertinent stimuli-responsive drug release processes relies on the redox-triggering bioactive agents present within cells such as glutathione (GSH) or dithiothreitol (DTT). So, unsurprisingly some of the most commonly explored and synthetically viable redox-triggered drug delivery materials involve disulfide bonds, which can be placed around the materials and then easily undergo cleavage through reduction upon contact with substances such as GSH and DTT.⁴ Since the concentration of GSH in the interior of a cell is much higher (10 mM) than that in the exterior of the cell (2 μM),^4b,9a,9b GSH is particularly good in triggering disulfide-functionalized materials to release their drug cargoes only when the materials reach intracellular spaces, and not before. This kind of redox-triggered drug release property stimulated by GSH has been extensively investigated. However, GSH, which is an important tripeptide composed of glutamate, cysteine and glycine, possesses not only redox properties but also acidity. To be specific, while the thiol groups inherited from cysteine can render GSH redox activity or the ability to break disulfide bonds, the two carboxylic groups present on its glutamate and glycine species make GSH acidic or help it to lower the pH of the systems in which it is present.

In this communication we report the release profiles of rhodamine 6G (R6G), a model drug molecule, from R6G-loaded MSNs (RMSNs) under different conditions: namely, with or without GSH. Specifically, we show that RMSNs exhibit non-redox drug release properties, unlike the most common redox-triggered drug release processes reported for GSH.⁴ The plausible mechanism by which the RMSNs show such kind of release properties toward the guest molecules upon contact with GSH is also discussed.

The RMSNs are prepared as briefly depicted in Fig. 1 (details are provided in ESI†). First, MSNs, whose surfactant templates are removed by solvent extraction, are synthesized as reported before.^6c The MSNs are then loaded with R6G molecules. Finally, the adsorption properties of MSNs toward R6G and the release properties of the RMSNs for R6G molecules are studied.

Fig. 1 Schematic representation of rhodamine 6G (R6G)-loaded MSNs (RMSNs) and how the RMSNs release their R6G guest molecules when the particles are exposed to GSH.

The presence or immobilization of R6G in the RMSNs is qualitatively evident from their pink color (Fig. S1 in ESI†) (please note that R6G and RMSNs are dark pink in color but the parent MSNs are white in color). The loading of R6G in the MSNs is further corroborated by UV-Vis diffuse reflectance spectroscopy, as the spectrum of RMSNs (Fig. S2 in ESI†) displays two peaks at 347 nm and 527 nm that correspond to R6G.

The structure of the MSNs is examined using transmission electron microscopy (TEM), scanning electron microscopy (SEM), X-ray diffraction (XRD) and N₂ adsorption/desorption. Fig. 2 shows the TEM and SEM images of MSNs. The representative TEM images of MSNs (Fig. 2a, c and d) show oval/spherical shaped particles with an average diameter of ca. 90 nm possessing hexagonally ordered mesoporous structures. The SEM image for MSNs (Fig. 2b) also reveals oval/spherical shaped particles with an average diameter of ca. 90 nm. Additionally, powder X-ray diffraction (PXRD) is used to characterize the structure of MSNs. The PXRD pattern of MSNs (Fig. 3a) displays three Bragg diffraction peaks in low 2θ region corresponding to the typical (100), (110) and (200) reflections of well-ordered mesostructured MCM-41 type MSNs.^3b This result is also in line with the ones obtained from TEM, which are discussed above.

Fig. 2 TEM (a) and SEM (b) images of MSNs, and high magnification TEM images (c and d) of MSNs.

Fig. 3 (a) PXRD patterns of MSNs. (b) N₂ adsorption/desorption isotherms and (c) pore size distributions of MSNs and as-synthesized mesostructured silica.

The MSNs are further characterized with N₂ adsorption/desorption porosimetry, and the result shows type IV isotherm with H1 type hysteresis loop, which is characteristic of mesoporous materials (Fig. 3b). At low relative pressure (0.1 < P/P₀ < 0.4), the isotherm for MSNs displays a large capillary condensation step (compared with that for its parent material, as-synthesized mesostructured silica). This suggests that the MSNs have larger pore volume than the as-synthesized mesostructured silica nanoparticles, or the pores in the former are largely free of surfactant molecules. The pore-size distribution, calculated using the adsorption branch of the isotherm according to the Barrett–Joyner–Halenda (BJH) method, shows a narrow pore size distribution ranging from 2.1 to 3.6 nm for MSNs; however, not surprisingly, the one for the as-synthesized mesostructured silica shows no pores at all (Fig. 3c). The pore volumes of the as-synthesized mesostructured silica and MSNs are found to be 0.42 and 1.18 cm³ g⁻¹, respectively, and their BET surface areas are found to be 81 and 928 m² g⁻¹, respectively. The higher pore volume and surface area of the MSNs compared with those of the as-synthesized mesostructured silica is clearly due to the absence of surfactant molecules in the former, and needless to say, their presence in the latter.

High surface area materials such as MSNs are known to adsorb large cargoes of guest molecules and release them under certain conditions (or they can serve as good drug delivery vehicles). Specifically here, the ability of MSNs to release molecular cargoes under non-redox related GSH stimulus is investigated. For this test, R6G is chosen as the guest molecule because it is relatively easy to detect and widely used as model drug before.¹⁰ Before the tests though, a graph of absorbance at 527 nm vs. concentration of R6G is plotted. Using the graph in accordance with the Beer–Lambert law (Fig. S3, ESI†), the equation “absorbance = 0.17 × concentration (in μg mL⁻¹) − 0.016” is derived, which is then employed to determine the concentrations of R6G released by RMSNs over time during the release experiments (see below). Additional UV-Vis spectra are run to check the stability of R6G in the solutions at room temperature, and this is typically performed by measuring the absorbance of 5 μg mL⁻¹ of R6G at 527 nm for 24 h (Fig. S4, ESI†). In this case, the absorbance is found to change only slightly (from 0.84 to 0.81) over 24 h, literally indicating R6G's stability.

Experiments involving R6G release in the presence of different concentrations of GSH (0, 0.1, or 1 mM) in deionized water are then conducted. As shown in Fig. 4a, the amount of R6G released by RMSNs in 1 min varies (increases) as the concentrations of GSH changes (increases). For example, the absorbance in 0 mM GSH is only 0.097, whereas the absorbance values in 0.1 mM and 1 mM GSH solutions are 0.303 and 0.899, respectively. It should be noted that because there is no disulfide gatekeeper in RMSNs, unlike in several other previously reported cases,^4,9a the observed release of R6G from RMSNs upon the latter's contact with GSH can not be attributed to the redox activity of GSH's thiol groups. Upon determination of the pH of these three solutions (i.e., 0, 0.1 and 1 mM GSH), it became apparent that the solutions possess significantly different pH values: 6.10, 4.51, and 3.65, respectively. The pH value of 6.10 for 0 mM GSH solution is most likely caused by some dissolved carbon dioxide in the deionized water at room temperature. Further decrease in pH upon addition of GSH into the solution is, however, caused by the carboxylic acid groups present on GSH. So, the release of R6G from RMSNs must be triggered with the help of GSH, or the pH change that GSH induces, rather than due to the most commonly reported, redox activity of GSH.⁴

Fig. 4 Results of release experiments of R6G from RMSNs under different conditions. (a) UV-Vis spectra of the supernatants obtained after placing RMSNs in 0, 0.1, or 1 mM GSH solution for 1 min. (b) UV-Vis spectra of the supernatants obtained after placing RMSNs in 1 mM GSH (pH 6.00) or 1 mM NaNO₃ solutions for 1 min. (c) Comparison of the calculated amount of R6G released in μg per mg from RMSNs within 1 min or 60 min in the presence of 0 mM GSH (A), 0.1 mM GSH (B), 1 mM GSH (C), 1 mM GSH with a pH of 6.00 (D), and 1 mM NaNO₃ (E) solutions.

In order to confirm this, two other release experiments were performed, one in 1 mM GSH, whose pH was adjusted to 6.00 using 2 M NaOH and be very close to that of the 0 mM GSH solution above, and another one in 1 mM NaNO₃ with a pH of 6.20. As shown in Fig. 4b, the absorbance is found to be very low in both cases: 0.167 and 0.136 for 1 mM GSH (pH 6.00) and 1 mM NaNO₃, respectively. This indicates that ions such as Na⁺, NO₃⁻ or the anion of GSH, do not stimulate the release of R6G from RMSNs. Thus, the major route by which GSH triggers the release of R6G from RMSNs is GSH's ability to decrease the pH of the solution.

The amount of R6G released by RMSNs (as μg R6G release per mg of RMSNs) in 1 min and 60 min is calculated based on the linear standard curve of R6G, and the result is displayed in Fig. 4c. As can be seen in the graph, there is no obvious difference in the amount of R6G released from RMSNs within 1 min versus 60 min in the different solutions where the experiments were carried out. This indicates that the release of R6G is very quick and reaches equilibrium fast. The result is also consistent with the fast proton exchange processes taking place as GSH is added. However, the values of R6G released in the solutions vary according to the concentrations of GSH. The values include the release of 5.36 μg R6G mg⁻¹ RMSNs in 1 mM GSH solution (whose pH is 3.65), which is much higher than that in water (0.67 μg R6G mg⁻¹ RMSNs). On the other hand, the amount is of R6G released in 1 mM GSH (whose pH has been re-adjusted to 6.00) is 1.07 μg R6G/mg RMSNs, which is only slightly higher than the one released in deionized water.

A possible mechanism reflecting RMSNs' release properties toward the cationic R6G guest molecules is proposed. Since the isoelectric point of silica is ca. 2.0,¹¹ it is quite obvious that MSNs' surfaces are negatively charged in the GSH solutions we employed in the studies. So, the positively charged R6G (ammonium species) can be adsorbed on the MSNs' surfaces predominantly via electrostatic interactions. Besides the ionic interaction, hydrogen bonding between R6G and surface silanols of MSNs may also assist with the adsorption of R6G on the MSNs though. Thus, as the pH of the solution is decreased (as a result of GSH), more Si–O⁻ groups will be protonated, resulting in Si–OH groups and making the MSNs' surfaces less negatively charged. As a result, the R6G molecules come off of the MSN's surfaces and are released into the solution.

In summary, we have shown that the changes in pH caused by GSH can trigger the burst release of cationic guest or model drug molecules loaded in MSNs. The study has also demonstrated the added advantage of GSH, an ubiquitous substance in biological systems, in being able to trigger the release of guest molecules off of MSNs via ionic interaction or pH changes, besides its well-known ability to induce drug release through its redox properties. Our measurements of 1 mM and 10 mM GSH in PBS give pH values of 6.91 and 3.75, respectively, both of which are lower than that of the original PBS (pH = 7.29); this suggests that higher GSH concentration in a buffered solution could reduce the pH to required levels to make these nanoparticles release their guest molecules in pH dependent manner. Thus, besides the redox property, the acidity of GSH can also be taken advantage of and utilized for controlled drug release and biomedical applications.

Acknowledgements

T. A. gratefully acknowledges the financial assistance to his research group by the US National Science Foundation (NSF) under NSF DMR-0968937, NSF NanoEHS-1134289, NSF-ACIF, and NSF Special Creativity grant.

Notes and references

1. (a) F. Tang, L. Li and D. Chen, Adv. Mater., 2012, 24, 1504; (b) S. H. Wu, Y. Hung and C. Y. Mou, Chem. Commun., 2011, 47, 9972; (c) I. I. Slowing, J. L. Vivero-Escoto, C. W. Wu and V. S. Y. Lin, Adv. Drug Delivery Rev., 2008, 60, 1278; (d) N. R. E. N. Impens, P. van der Voort and E. F. Vansant, Microporous Mesoporous Mater., 1999, 28, 217.
2. (a) Q. Zheng, Y. Hao, P. Ye, L. Guo, H. Wu, Q. Guo, J. Jiang, F. Fu and G. Chen, J. Mater. Chem. B, 2013, 1, 1644; (b) D. He, X. He, K. Wang, M. Chen, Y. Zhao and Z. Zou, J. Mater. Chem. B, 2013, 1, 1552; (c) Y. Tian, A. Glogowska, W. Zhong, T. Klonisch and M. Xing, J. Mater. Chem. B, 2013, 1, 5264; (d) F. Muhammad, M. Guo, W. Qi, F. Sun, A. Wang, Y. Guo and G. Zhu, J. Am. Chem. Soc., 2011, 133, 8778.
3. (a) Z. Yu, N. Li, P. Zheng, W. Pan and B. Tang, Chem. Commun., 2014, 50, 3494; (b) C. D. L. Torre, A. Agostini, L. Mondragón, M. Orzáez, F. Sancenón, R. Martínez-Máñez, M. D. Marcos, P. Amorós and E. Pérez-Payá, Chem. Commun., 2014, 50, 3184.
4. (a) H. Kim, S. Kim, C. Park, H. Lee, H. J. Park and C. Kim, Adv. Mater., 2010, 22, 4280; (b) Y. Cui, H. Dong, X. Cai, D. Wang and Y. Li, ACS Appl. Mater. Interfaces, 2012, 4, 3177; (c) Z. Luo, K. Cai, Y. Hu, L. Zhao, P. Liu, L. Duan and W. Yang, Angew. Chem., Int. Ed., 2011, 50, 640; (d) Q. Zhang, F. Liu, K. T. Nguyen, X. Ma, X. Wang, B. Xing and Y. Zhao, Adv. Funct. Mater., 2012, 22, 5144; (e) R. Liu, X. Zhao, T. Wu and P. Feng, J. Am. Chem. Soc., 2008, 130, 14418.
5. (a) Z. Zhang, D. Balogh, F. Wang, R. Tel-Vered, N. Levy, S. Y. Sung, R. Nechushtai and I. Willner, J. Mater. Chem. B, 2013, 1, 3159; (b) J. Lai, X. Mu, Y. Xu, X. Wu, C. Wu, C. Li, J. Chen and Y. Zhao, Chem. Commun., 2010, 46, 7370; (c) X. Mei, S. Yang, D. Chen, N. Li, H. Li, Q. Xu, J. Ge and J. Lu, Chem. Commun., 2012, 48, 10010; (d) S. Yang, N. Li, D. Chen, X. Qi, Y. Xu, Y. Xu, Q. Xu, H. Li and J. Lu, J. Mater. Chem. B, 2013, 1, 4628.
6. (a) M. Chen, C. Huang, C. He, W. Zhu, Y. Xu and Y. Lu, Chem. Commun., 2012, 48, 9522; (b) E. Aznar, R. Villalonga, C. Giménez, F. Sancenón, M. D. Marcos, R. Martínez-Máñez, P. Díez, J. M. Pingarrón and P. Amorós, Chem. Commun., 2013, 49, 6391; (c) Y. Zhao, B. G. Trewyn, I. I. Slowing and V. S.-Y. Lin, J. Am. Chem. Soc., 2009, 131, 8398.
7. E. Bringas, Ö. Köysüren, D. V. Quach, M. Mahmoudi, E. Aznar, J. D. Roehling, M. D. Marcos, R. Martínez-Máñez and P. Stroeve, Chem. Commun., 2012, 48, 5647.
8. Y. L. Sun, Y. Zhou, Q. L. Li and Y. W. Yang, Chem. Commun., 2013, 49, 9033.
9. (a) A. N. Koo, H. J. Lee, S. E. Kim, J. H. Chang, C. Park, C. Kim, J. H. Park and S. C. Lee, Chem. Commun., 2008, 48, 6570; (b) D. P. Jones, J. L. Carlson, P. S. Samiec, P. Sternberg Jr, V. C. Mody Jr, R. L. Reed and L. A. S. Brown, Clin. Chim. Acta, 1998, 275, 175.
10. (a) J. K. Oh, D. J. Siegwart, H. Lee, G. Sherwood, L. Peteanu, J. O. Hollinger, K. Kataoka and K. Matyjaszewski, J. Am. Chem. Soc., 2007, 129, 5939; (b) P. DeMuth, M. Hurley, C. Wu, S. Galanie, M. R. Zachariah and P. DeShong, Microporous Mesoporous Mater., 2011, 141, 128.
11. (a) I. Larson, C. J. Drummond, D. Y. C. Chan and F. Grieser, Langmuir, 1997, 13, 2109; (b) X. Huang, Z. Tao, J. C. Praskavich Jr, A. Goswami, J. F. Al-Sharab, T. Minko, V. Polshettiwar and T. Asefa, Langmuir, 2014, 30, 10886.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details and supporting results. See DOI: 10.1039/c4ra08570a

---