Protective effect of mesoporous silica particles on encapsulated folates

Document downloaded from: This paper must be cited as: The final publication is available at Copyright Additional Information http://dx.doi.org/ 10.1016/j.ejpb.2016.05.016 http://hdl.handle.net/10251/78363 Elsevier Ruiz Rico, M.; Daubenschüz, H.; Pérez Esteve, E.; Marcos Martínez, MD.; Amorós, P.; Martínez Mañez, R.; Barat Baviera, JM. (2016). Protective effect of mesoporous silica particles on encapsulated folates. European Journal of Pharmaceutics and Biopharmaceutics. 105:9-17. doi:10.1016/j.ejpb.2016.05.016. 1 Protective effect of mesoporous silica particles on encapsulated folates 1 2 María Ruiz-Rico1,*, Hanna Daubenschüz1, Édgar Pérez-Esteve1, María D. Marcos2,3, Pedro 3 Amorós4, Ramón Martínez-Máñez2,3, José M. Barat1 4 5 1 Grupo de Investigación e Innovación Alimentaria. Departamento de Tecnología de Alimentos, 6 Universitat Politècnica de València. Camino de Vera s/n, 46022, Valencia, Spain 7 2 Centro de Reconocimiento Molecular y Desarrollo Tecnológico (IDM), Unidad Mixta 8 Universitat Politècnica de València y Universitat de València. Departamento de Química, 9 Universitat Politècnica de València, Camino de Vera s/n, 46022, Valencia, Spain 10 3 CIBER de Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN) 11 4 Institut de Ciència dels Materials (ICMUV), Universitat de València, P.O. Box 22085, 46071 12 Valencia, Spain 13 *(M.R.R.) Phone: +34963877000 ext.83612. E-mail: [email protected]. 14 2 Abstract 15 Mesoporous silica particles (MSPs) are considered suitable supports to design gated materials 16 for the encapsulation of bioactive molecules. Folates are essential micronutrients which are 17 sensitive to external agents that provoke nutritional deficiencies. Folates encapsulation in MSPs 18 to prevent degradation and to allow their controlled delivery is a promising strategy. 19 Nevertheless, no information exists about the protective effect of MSPs encapsulation to prevent 20 their degradation. In this work, 5-formyltetrahydrofolate (FO) and folic acid (FA) were 21 entrapped in MSPs functionalized with polyamines, which acted as pH-dependent molecular 22 gates. The stability of free and entrapped vitamins after acidic pH, high temperature and light 23 exposure was studied. The results showed the degradation of FO after high temperature and 24 acidic pH, whereas entrapped FO displayed enhanced stability. Free FA was degraded by light, 25 but MSPs stabilized the vitamin. The obtained results point towards the potential use of MSPs 26 as candidates to enhance stability and to improve the bioavailability of functional biomolecules. 27 28 Keywords: 5-formyltetrahydrofolate; controlled release; encapsulation; folic acid; mesoporous 29 silica particles; stability 30 31 32 33 Abbreviations used 34 5-formyltetrahydrofolate (FO), ascorbic acid (AA), encapsulated 5-formyltetrahydrofolate (E35 FO), encapsulated folic acid (E-FA), European Food Safety Authority (EFSA), folic acid (FA), 36 free 5-formyltetrahydrofolate (F-FO), free folic acid (F-FA), least significant difference (LSD), 37 mesoporous silica particles (MSPs), N-(3-trimethoxysilylpropyl)diethylenetriamine (N3), N38 cetyltrimethylammonium bromide (CTABr), phosphate-buffered saline (PBS), powder X-ray 39 diffraction (PXRD), tetrabutylammonium hydrogen sulphate (TBAHS), tetraethylorthosilicate 40 (TEOS), transmission electron microscopy (TEM), triethanolamine (TEAH3), ultraviolet (UV). 41 3 Graphical abstract 42 43 0 20 40 60 80 100 810 12 14 16 Vitamin recovery (%) Time of exposure (h) Encapsulated vitamin Free vitamin 4 Introduction 44 In the last years hybrid organic−inorganic materials have attracted considerable interest due to 45 the combination of the beneficial characteristic of organic chemistry and material science in 46 order to develop smart nanodevices. Among different hybrid solids, mesoporous silica particles 47 (MSPs) offer several unique features that allow the design of gated materials for controlled 48 release and sensing/recognition protocols [1]. The first family of MSPs called MCM-X was 49 described in the early 1990s by the Mobil Corporation Laboratories. This family of silica 50 supports include different porous silica exhibiting hexagonal (MCM-41), cubic (MCM-48) and 51 lamellar (MCM-50) pore shapes. After these developments, new ordered materials (e. i. MSU, 52 KIT, FDU, AMS, SBA…) with a wide range of textural properties have been described by 53 different authors [2]. MSPs possess broadly advantageous properties such as biocompatibility, 54 thermal and chemical stability, huge loading capacity, high surface areas, tunable morphologies 55 and pore sizes, as well as facile functionalization of surfaces and pores [3-7]. The surface 56 functionalization of MSPs for the development of gated materials allows that the delivery of the 57 cargo stored in the inorganic support can be triggered by applying selected external stimulus [1]. 58 Furthermore, it is considered that the inorganic framework can effectively protect the payload 59 molecules from enzymatic degradation or denaturation caused by environmental changes [7]. 60 However, there are few studies in the literature about the protective effect of MSPs on the 61 stability of biomolecules in biological solutions. 62 The functionalized MSPs have been used to encapsulate drugs mainly for the biomedical field, 63 but also to encapsulate bioactive molecules for other sectors such as food technology. Different 64 food ingredients and nutraceuticals including vitamins [8-12], antioxidants [13,14], 65 antimicrobials [15-17], aromas [18] or enzymes [19] have been entrapped in gated mesoporous 66 materials. Most studies have been focused on the development and optimization of the 67 encapsulation systems for controlled delivery, but it is expected that the MPSs may be able to 68 enhance the stability of the entrapped bioactive compound. 69 5 Water-soluble vitamins, like folates, are labile compounds in the presence of environmental 70 agents, such as extreme pH values or high temperatures [20]. As folates are essential for the 71 human body and cannot be synthesized de novo by the organism, this indispensable vitamin 72 needs to be obtained from food or dietary supplements [21]. Thus the stability of this vitamin 73 after storage and processing in food products or supplements should be taken into account. Loss 74 of the biochemical activity of natural folates can occur during harvest, storage and food 75 processing [22,23]. In general, pH, temperature, pressure, light and antioxidants, among others, 76 can affect the stability of the natural folates and the synthetic folic acid (FA) [24-30]. FA, with a 77 fully oxidized pteridine ring system, exhibits greater stability than folates. Among folates, large 78 differences in stability exist in susceptibility to oxidative degradation, and 579 formyltetrahydrofolate (FO) is the most stable [31]. Moreover, the stability of folates is 80 influenced by pH and oxygen, which provokes their oxidation [30,32]. The inclusion of 81 antioxidant compounds, such as ascorbic acid (AA) or mercaptoethanol, is required to prevent 82 the destruction of labile folates from thermal exposure and photodegradation during food 83 processing [20,23,33]. 84 Bearing in mind these factors, it is of interest to create folates encapsulation systems which can 85 ensure the required dose and fully guarantee the stability and bioavailability of this vitamin. 86 Therefore, the objective of this study was the encapsulation of FO and FA in a mesoporous 87 silica support (MCM-41) functionalized with amines to create a system to be used in orally 88 delivered applications, and to study the stability of entrapped vitamins to test the efficacy of the 89 MCM-41 support as a protector against external agents, such as acidic pH, high temperature and 90 light. 91 92 6 Materials and methods 93 Chemicals 94 Tetraethylorthosilicate (TEOS), N-cetyltrimethylammonium bromide (CTABr), sodium 95 hydroxide (NaOH), triethanolamine (TEAH3), N-(3-trimethoxysilylpropyl)diethylenetriamine 96 (N3), sodium phosphate monobasic (NaH2PO4), sodium phosphate dibasic (Na2HPO4) and 97 tetrabutylammonium hydrogen sulphate (TBAHS) were provided by Sigma-Aldrich (Madrid, 98 Spain). 5-formyltetrahydrofolate (FO) and folic acid (FA) were purchased from Schircks 99 Laboratories (Jona, Switzerland). Acetonitrile HPLC grade was provided by Scharlab 100 (Barcelona, Spain). 101 102 Mesoporous silica particles synthesis 103 Synthesis of microparticulated MCM-41 particles was carried out following the so-called 104 “atrane route”, where CTABr was used as the structure-directing agent. A molar ratio, fixed at 7 105 TEAH3: 2 TEOS:0.52 CTABr:0.5 NaOH:180 H2O. CTABr, was added to a TEAH3 and NaOH 106 solution, which contained TEOS at 118 ºC. After dissolving CTABr in the solution, water was 107 slowly added along with vigorous stirring at 70 ºC to form a white suspension. This mixture was 108 aged at 100 ºC for 24 h. Following synthesis, the solid was recovered, washed with deionized 109 water and dried at 70 ºC. The as-synthesized microparticles were calcined at 550 ºC in an 110 oxidant atmosphere for 5 h to remove the template phase [16]. 111 112 Synthesis of encapsulated folates 113 The design of the encapsulation system was based on a previous work, in which FA was 114 entrapped in a MSP functionalized with amines to deliver FA during a simulated digestion 115 process [12]. Dissolutions of FO and FA (10 mg/mL) were prepared in distilled water and 116 phosphate-buffered saline (PBS), respectively. Solutions were added to 300 mg of MCM-41 in 117 3 addition cycles (1.5 mL per cycle). After each addition cycle, solids were dried at 37 ºC to 118 7 remove water content. After loading and drying, solids were collected and functionalized with 119 1.29 mL of N3 using different media; i.e. acetonitrile (E-FO) or acetate buffer at pH 2 (E-FA). 120 The final mixtures were stirred for 5.5 h at room temperature, isolated by vacuum filtration, 121 washed with 300 mL of water adjusted to pH 2, and dried at room temperature for 24 h. 122 123 Characterization of solids 124 Powder X-ray diffraction (PXRD), transmission electron microscopy (TEM), N2 adsorption125 desorption isotherms and zeta potential were used to characterize the synthesized materials. 126 PXRD was performed in a BrukerD8 Advance diffractometer using CuKα radiation (Bruker, 127 Coventry, UK). For the TEM analysis, particles were dispersed in dichloromethane and 128 sonicated for 2 min to preclude aggregates. The suspension was then deposited onto copper 129 grids coated with a carbon film (Aname SL, Madrid, Spain). The MSPs samples were imaged 130 by JEOL JEM-1010 (JEOL Europe SAS, Croissy-sur-Seine, France) at an acceleration voltage 131 of 80 kV. The single-particle size was estimated by averaging the measured size values of 50 132 particles. The N2 adsorption-desorption isotherms were recorded with a Micrometrics 133 ASAP2010 automated sorption analyzer (Micromeritics Instrument Corporation, Norcross, 134 USA). Samples were degassed at 90 ºC in vacuum overnight. Specific surface areas were 135 calculated from the adsorption data within the low pressure range by the BET model. Pore size 136 was determined following the BJH method. To determine the zeta potential of the materials, a 137 Zetasizer Nano ZS (Malvern Instruments, UK) was employed. Samples were dispersed in water 138 at a concentration of 1 mg/mL. Before taking each measurement, samples were sonicated for 2 139 min to preclude aggregation. The zeta potential was calculated from the particle mobility values 140 by applying the Smoluchowski model. The average of five recordings was reported as the zeta 141 potential. Measurements were taken at 25 ºC in triplicate. 142 143 8 Release studies 144 Delivery studies were conducted to test the release capacity of the encapsulation system and to 145 confirm the functionality of the gates to modulate the release of vitamins according to the pH of 146 the medium (closed gates at pH 2, opened gates at pH 7.5). To determine the release of FO and 147 FA from the amine-gated mesoporous support (E-FO and E-FA), 10 mg of the solids were 148 placed in 25 mL of PBS at pH 2 and pH 7.5. At certain time points (0, 2, 5, 15, 30, 60, 120, 180 149 min), aliquots were separated, the suspension was filtered and the solution was analyzed by 150 HPLC. 151 152 Stability assays 153 The influence of diverse external agents, such as acidic pH, high temperature and light, on the 154 stability of the free and entrapped vitamin was studied. Free FO and FA were treated, whenever 155 possible, as the encapsulated vitamin in order to ensure reproducibility. In order to simulate not 156 only the 3 day-loading (72 h drying at 37°C) of the particles with the vitamin, but also a further 157 24-hour drying period after functionalization, compounds in their free form were incubated for 158 96 h at 37°C. The stability assays with the free vitamin were all conducted with these incubated 159 samples (F-FO and F-FA). 160 For the stability assays, 4 mg of the entrapped vitamins (E-FO and E-FA) and the 161 correspondent amounts of the free forms (ca. 0.02 mg for FO and ca. 0.3 mg for FA) were 162 dissolved in 10 mL of PBS (pH 2 or pH 7.5). All the stability experiments were performed in 163 triplicate. The vitamin recoveries were presented by assuming the percentage recovered under 164 optimal conditions to be 100% (pH 7.5, no treatment). 165 166 pH stability 167 These experiments were carried out to study the stability and solubility of vitamins at different 168 pH values; e.g., the acidic pH at which FO and FA exhibited very low solubility [30,32]. These 169 15 285 Functionalization efficiency was verified by the zeta potential determinations of bare MCM-41, 286 MCM-41 loaded with FO/FA, and MCM-41 loaded and functionalized with amines. Bare 287 particles revealed an average negative zeta potential of -31 mV. After loading particles with 288 FO/FA, the zeta potential changed slightly to values of ca. -30 mV. Yet after functionalization 289 with N3, the zeta potential changed positively to values of ca. 50 mV for E-FO and E-FA, 290 which confirmed the attachment of amines to the particle surface. 291 292 Release studies 293 The release studies confirmed the mechanism of the amine-gated MSPs to modulate vitamin 294 release according to the pH of the medium. The pH-dependent releases of the encapsulated FO 295 and FA are shown in Figure 4. Gates were largely closed at pH 2 and the vitamin was barely 296 detected, which confirmed that vitamin delivery was hindered by the combination of the low 297 solubility of the vitamins under acidic conditions, the effect of the amines anchored to the 298 surface of MSPs and the polyammonium groups-anionic species interaction. At an acidic pH, 299 polyamines were transformed into polyammonium groups, which adopted a rigid-like 300 conformation due to Coulombic repulsions and coordinate anions (phosphates present in 301 solution), which blocked pores and avoided vitamin release [8,34]. 302 In contrast, FO and FA showed a progressive release among time at pH 7.5. After 2 h, 303 maximum vitamin releases were obtained at pH 7.5, with 41.9±7.2 mg FO/g solid for E-FO and 304 84.3±7.8 mg FA/g solid for E-FA. The maximum released amounts were used to calculate the 305 equivalent amount of solids needed in the stability assays to make a comparison between the 306 free and encapsulated FO and FA. A sustained release was produced because polyamines were 307 less protonated at a neutral pH, and the Coulombic repulsion between them and the affinity for 308 anions significantly reduced. These effects, along with increased vitamin solubility, allowed the 309 delivery of FO and FA from pores. This pH-responsive delivery effect has been suggested to be 310 16 suitable for releasing vitamins in the gastrointestinal tract (closed gates in the stomach, opened 311 gates in the intestine) [12]. Encapsulation was also expected to protect vitamins from 312 degradation after exposure to environmental agents (vide infra). 313 314 315 Figure 4. Release profiles of vitamin from the pores of E-FO (A) and E-FA (B) in PBS at pH 316 2.0 (dotted lines) and pH 7.5 (solid lines). Values are Means ± SD, n = 3. 317 318 Stability assays 319 The influence of diverse external agents related to food processing or storage, such as pH, 320 temperature and light, on the stability of free 5-formyltetrahydrofolate and folic acid (F-FO and 321 F-FA) and the corresponding entrapped vitamins (E-FO and E-FA) was studied. 322 323 pH 324 The study of the effect of pH on the stability of FO and FA at different pH values was 325 conducted in two steps. In the first step, water solutions of free FA and FO were adjusted to 326 different pHs and stirred for 1 h before being analyzed by HPLC. Figure 5 shows the detected 327 concentrations of FA and FO (in terms of recovery) in the aqueous solutions under all the study 328 conditions (i.e. pH 1-10). As observed, recoveries reached values of ca. 100% from pH 5 to 10, 329 which confirms the stability of both molecules at these pH values. Below this pH range, the 330 0 10 20 30 40 50 60 70 80 90 100 060 120 FA release (%) Time (min) 0 10 20 30 40 50 60 70 80 90 100 060 120 FO relesase (%) Time (min) A B 17 concentrations of both molecules lowered. The FO concentration in water gradually lowered 331 from ca. pH 4 to pH 1, whereas this effect was observed for FA below ca. pH 3. 332 333 334 Figure 5. F-FO and F-FA recoveries at different pH values. Different letters in the bars indicate 335 statistically significant differences (p<0.05) from levels of pH. Values are Means ± SD, n = 3. 336 337 The drop in the recovery of FO and FA at an acidic pH can be explained by three phenomena: 338 (a) loss of solubility; (b) interconversion into other derivatives; (c) oxidative degradation. 339 Folates are slightly soluble at an acidic pH, and are highly soluble under neutral/basic 340 conditions due to the protonation and deprotonation of molecules in aqueous environments [32]. 341 In addition to oxidative degradation, FO can nonenzymatically interconvert with 5,10342 methenyltetrahydrofolate through changes in pH, temperature and oxygen [35]. 5,10343 methenyltetrahydrofolate is formed by the acidification of 5-formyltetrahydrofolate because one 344 molecule of water is lost (dehydration), which leads to the cyclization of the molecule in a 345 reversible manner. The equilibrium gradually shifts toward 5,10-methenyltetrahydrofolate, and 346 its formation becomes faster the lower pH becomes [36]. 347 Bearing all these factors in mind, which could explain loss of recovery at an acidic pH, in a 348 second step, experiments were run to determine the amount of vitamins lost at an acidic pH. In 349 0 10 20 30 40 50 60 70 80 90 100 1 2 3 4 5 6 7 8 9 10 Recovery (%) pH F-FO F-FA a a a a b ccc c c a a b c d dd d d d 18 them aqueous solutions of free FA and FO were adjusted to pH 2, stirred for 1 h and then pH 350 was adjusted to 7.5 before the HPLC analysis. The percentage of vitamins determined at pH 2 351 and after neutralization to pH 7.5 is shown in Fig. 6A, where almost no recovery of vitamins is 352 detected after stirring them for 1 h at pH 2 (which agrees with Fig. 5). However, the vitamins 353 reappeared with a percentage of ca. 40% for F-FO and of ca. 72% for F-FA after neutralizing 354 the pH. F-FA gave higher values after adjusting to the neutral value than free FO, but none of 355 them achieved complete recovery. Some studies have revealed that FA might not be soluble at a 356 low pH [12], but other authors have reported its degradation at an acidic pH [29]. The natural 357 folate showed less stability at an acidic pH than FA, probably due to degradation [36]. 358 Similar studies to those shown above have been conducted in the presence of ascorbic acid 359 (AA), and their results are shown in Fig. 6B. This antioxidant was included because previous 360 studies have demonstrated that its incorporation increases folate stability as oxidation reactions 361 are prevented [35]. As for the vitamins supplemented with AA, the F-FO concentration only 362 slightly increased (from 40% in the absence of AA to 50%). However, AA remarkably 363 influenced the stability of F-FA, and revealed a recovery of almost 90%. 364 Lastly, the effect of pH changes on the encapsulated vitamins was studied in order to prove the 365 protective function of the support. As seen in Fig. 6C, minor vitamin recoveries took place at 366 pH 2. When pH was adjusted to 7.5, E-FO and E-FA were almost fully detectable with a 367 recovery of 94% and 99%, respectively. Both encapsulated vitamins were highly preserved in 368 the acidic environment by the pH-responsive gated material. As a result, the highly protective 369 function of the MSPs functionalized with amines at a low pH was evidenced by both vitamins. 370 This approach better improved the stability of vitamins than the strategy reported to enhance the 371 stability of natural folates (addition of antioxidants). Neither the free form nor the vitamins 372 supplemented with AA were as stable at an acidic pH as they were inside the pores of MSPs. 373 374 19 375 Figure 6. FO and FA recoveries after pH changes for free vitamins (A), free vitamins in 376 presence of ascorbic acid (B) and encapsulated vitamins (C). Different letters in the bars 377 indicate statistically significant differences (p<0.05) from levels of pH. Values are Means ± SD, 378 n = 3. 379 380 Temperature 381 Previous experiments conducted with vitamins dissolved in PBS at temperatures below 100 °C 382 had no impact on their stability (data not shown). In order to investigate the impact of higher 383 temperatures on the vitamins, a study was carried out by simulating sterilization conditions (121 384 °C, 1 bar) at different times. The temperature assays performed in the autoclave are presented in 385 Figure 7. The results showed that encapsulated folate did not significantly reduce vitamin 386 content at various exposure times. In contrast, F-FO revealed a significant loss of ca. 27% after 387 15 min, probably due to the formation of interconversion products [30]. With FA, no significant 388 differences were obtained for both the E-FA and F-FA results. These results are in accordance 389 with previous studies that have reported good FA stability after thermal exposure in the solid 390 state and with dissolution [24,26]. Synthetic vitamin has been suggested to be the most stable 391 type in the folate group because of its oxidized p-teridin ring [37]. The thermostability of FA 392 and FO has been previously reported as being similar at a neutral pH [23]. The results obtained 393 with the encapsulated vitamins revealed that entrapped FO could bear up under thermal pressure 394 exposure and greater stability after proving FO encapsulation. However, encapsulated FA could 395 also resist the burden of thermal pressure as well as its free form. 396 0 10 20 30 40 50 60 70 80 90 100 2 7.5 Recovery (%) pH E-FO E-FA 0 10 20 30 40 50 60 70 80 90 100 2 7.5 Recovery (%) pH F-FO+AA F-FA+AA aa 0 10 20 30 40 50 60 70 80 90 100 2 7.5 Recovery (%) pH F-FO F-FA BA aa b b C aa b b b b 20 397 Figure 7. Influence of temperature exposure on the stability of encapsulated (E-) and free (F-) 398 FO (A) and FA (B) vitamins. Different letters in the bars indicate statistically significant 399 differences (p<0.05) from levels of time exposure (small letters) and differences between the 400 encapsulated FO/FA or in their free form (capital letters). Values are Means ± SD, n = 3. 401 402 Light 403 Previous articles, which have reported the influence of various light sources on the degradation 404 of synthetic FA, have investigated the impact of visible and UV light on both vitamins [25]. 405 Preliminary experiments revealed no or very little degradation after 6 h (data not shown). 406 Therefore, assays were carried out from 8 h to 16 h. 407 Visible light assays were conducted with a lamp, which generated visible light with an intensity 408 ca. 8 mW/cm². The results obtained from visible light experiments are presented in Figure 8. 409 The good stability of FO (Fig. 8A) after light exposure was evidenced. Neither F-FO nor E-FO 410 showed degradation during visible light exposure. 411 With FA (Fig. 8B), F-FA showed considerable loss after 8 h of visible light exposure, with a 412 remaining averaged amount of 40%. Gradual reduction of F-FA was detected up to 12 h 413 irradiation, with a low value of 12%, and total degradation occurred after 16 h with an average 414 remaining amount of 3%. Conversely, E-FA was well-protected by the functionalized support 415 and a non-significant decrease was detected. 416 0 10 20 30 40 50 60 70 80 90 100 510 15 Recovery (%) Time (min) E-FA F-FA 0 10 20 30 40 50 60 70 80 90 100 510 15 Recovery (%) Time (min) E-FO F-FO aA aA aA aA aA bA aA aA aA aA aA aA BA 21 Given F-FA’s tendency to be degraded by light in solution, the effect of antioxidant AA was 417 also evaluated as a strategy to improve stability and to confirm the mechanism of degradation 418 (i.e. oxidation). Free FA dissolution supplemented with AA showed marginal fluctuation after 419 16 h of visible light exposure, but appeared to stabilize FA substantially in the same way as the 420 encapsulation system. 421 422 423 Figure 8. Influence of visible light exposure on the stability of encapsulated (E-) and free (F-) 424 FO (A) and FA (B) in presence or not of ascorbic acid. Different letters in the bars indicate 425 statistically significant differences (p<0.05) from levels of visible light exposure (small letters) 426 and differences between the encapsulated FO/FA or in their free form (capital letters). Values 427 are Means ± SD, n = 3. 428 429 Figure 9 shows the free and encapsulated FO and FA recoveries after UV light exposure with an 430 estimated intensity of 4 mW/cm². The results revealed that E-FO and F-FO exhibited good 431 stability under UV light (Fig. 9A). Hence natural folate was highly stable in both the free and 432 encapsulated forms. FA stability was affected by UV light (Fig. 9B) and F-FA showed losses 433 after 8 h of UV exposure. As in the visible light assays, the impact of AA enhanced the good 434 stability of the synthetic vitamin and demonstrated more effective protection than the 435 mesoporous system after 16 h of light exposure. 436 437 0 10 20 30 40 50 60 70 80 90 100 810 12 14 16 Recovery (%) Time (h) E-FO F-FO aAaA aA aA aA aAaA aA aA aA A 0 10 20 30 40 50 60 70 80 90 100 810 12 14 16 Recovery (%) Time (h) E-FA F-FA F-FA+AA aB aB aB aB aB abB aA bA cA cA cA aB abcB cB bcB B 22 438 Figure 9. Influence of UV light exposure on the stability of encapsulated (E-) and free (F-) FO 439 (A) and FA (B) in presence or not of ascorbic acid. Different letters in the bars indicate 440 statistically significant differences (p<0.05) from levels of visible light exposure (small letters) 441 and differences between the encapsulated FO/FA or in their free form (capital letters). Values 442 are Means ± SD, n = 3. 443 444 Summing up the two light experiments, it was noted that FO did not degrade under either visible 445 light exposure or UV light stress, but FA was susceptible to visible and UV light exposure. 446 However, the mesoporous system was able to efficiently protect the vitamin and stability was 447 greatly enhanced. 448 This study simulated indirect light-induced stress and, therefore, intensities until degradation 449 occurred. We were unable to compare these results with previous studies as they all used direct 450 exposures [25,27,28]. Even though it has been suggested to be considerably more stable at pH 451 7.5 than in acidic media [29], we detected complete oxidative degradation after 16 h of visible 452 light and UV exposure. Akhtar et al. [25] suggested the mechanism of this oxidative 453 degradation as they hinted at degradation being produced in the C9-N10 position. Aqueous 454 solution can form various radiolytic products, which initiate oxidative dehydrogenation and lead 455 to an enamine compound. This intermediate form is highly susceptible in acidic media and can 456 undergo fast degradation. In alkaline media, this decomposition process has been proposed to 457 take place more slowly, but it also led to irreversible cleavage between C9-N10 bonding after 16 458 0 10 20 30 40 50 60 70 80 90 100 810 12 14 16 Recovery (%) Time (h) E-FO F-FO 0 10 20 30 40 50 60 70 80 90 100 810 12 14 16 Recovery (%) Time (h) E-FA F-FA F-FA+AA aA aAaAaA aAaA aA aA aA aA aA aB aB bB bB aB aA aA bA bA aB aB aB aB aC A B 23 h of light exposure. This caused final decay in the separation of the p-teridin moiety from p459 aminobenzoylglutamate [25,27]. 460 Encapsulated FA, which was solved in PBS (pH 2) in the stability assays, showed good 461 stability. Not even the degradation-favorable acidic surrounding could hardly affect E-FA 462 compared to the free form. Apart from successful FA improvement through encapsulation, 463 antioxidant AA enhanced stability in the same way. Compared to the well-known protective 464 mechanism of antioxidants [38], the protective function of the mesoporous support has not been 465 reported to date, and the stabilizing mechanism remains unclear. Although it is fully accepted 466 that MSPs show very little absorption within the visible and ultraviolet range [39,40], enhanced 467 encapsulated vitamin recovery was confirmed herein. The possible role of MSPs as a stability 468 enhancer could hinder access to weak points (C9-N10 bonding) by entrapping the vitamin in 469 mesopores in such a way that conformational transformations of the molecule are avoided. 470 471 Conclusions 472 The successful entrapment of natural FO and synthetic FA in pH-responsive MSPs and the 473 controlled release of the compounds that mimic the gastrointestinal tract were accomplished 474 herein. The ability of MSPs to protect vitamins after environmental degradation was clearly 475 evidenced. The stability assays revealed that encapsulated FO and FA were effectively protected 476 against degradation at an acidic pH compared to their free from. The sterilization studies 477 showed that encapsulation allowed vitamins to withstand thermal exposure and enhanced their 478 stability. The results obtained after exposure to visible and UV light displayed good stability for 479 free FO, which was not influenced by encapsulation, but improved FA stability after entrapment 480 in MSPs. When considering the protective effect of MSPs against external agents and acidic 481 stomach conditions, and progressive delivery with time under the intestinal conditions, the FO482 and FA-loaded supports proposed herein can be considered promising potential systems as 483 supplements for food systems. 484 485 24 Acknowledgements 486 Authors gratefully acknowledge the financial support from the Ministerio de Economía y 487 Competitividad (Projects AGL2012-39597-C02-01, AGL2012-39597-C02-02 and MAT2012488 38429-C04-01), FEDER founding and the Generalitat Valenciana (Project 489 PROMETEOII/2014/047). M.R.R. and E.P.E. are grateful to the Ministerio de Ciencia e 490 Innovación for their grants (AP2010-4369 and AP2008-0620). 491 492 References 493 [1] Aznar, E., Oroval, M., Pascual, L., - , Gated 494 materials for on-command release of guest molecules, Chem. Rev. 116 (2016) 561−718. 495 [2] Choudhari, Y., Hoefer, H., Libanati, C., Monsuur, F., McCarthy, W., Mesoporous silica 496 drug delivery systems, in: Shah, N., Sandhu, H., Choi, D.S., Chokshi, H., Malick, A.W. (Eds.), 497 Amorphous solid dispersions, Springer, New York, USA, 2014, pp. 665-693. 498 [3] Slowing, I.I., Vivero-Escoto, J.L., Wu, C.W., Lin, V.S.Y., Mesoporous silica nanoparticles 499 as controlled release drug delivery and gene transfection carriers, Adv. Drug Deliv. Rev. 60 500 (2008) 1278-1288. 501 [4] Popat, A., Hartono, S.B., Stahr, F., Liu, J., Qiao, S.Z., Lu, G.Q., Mesoporous silica 502 nanoparticles for bioadsorption, enzyme immobilisation, and delivery carriers, Nanoscale 503 3(2011), 2801-2181. 504 [5] Li, Z., Barnes, J.C., Bosoy, A., Stoddart, J.F., Zink, J.I., Mesoporous silica nanoparticles in 505 biomedical applications, Chem. Soc. Rev. 41(2012) 2590-2605. 506 [6] Pérez-Esteve, E., Oliver, L., García, L., Nieuwland, M., de Jongh, H.H., Martínez-Máñez, 507 R., Barat, J.M., Incorporation of mesoporous silica particles in gelatine gels: effect of particle 508 type and surface modification on physical properties, Langmuir 30 (2014) 6970-6979. 509 [7] Song, N., Yang, Y.W., Molecular and supramolecular switches on mesoporous silica 510 nanoparticles, Chem. Soc. Rev. 44 (2015) 3474-3504. 511