Use of Zeolite ZSM-5 for Loading and Release of 5-Fluorouracil

Samples of zeolite ZSM-5 have been synthesized in both the sodium form (ZSM-5) and the acid activated form (H-ZSM-5). In addition, each of these two forms was prepared in the two molar SiO 2 /Al 2 O 3 ratios of 169 and 15. All samples of these ZSM-5 derivatives were characterized by X-ray diffraction (XRD), nitrogen adsorption-desorption isotherms, thermal gravimetric analysis (TGA), X-ray fluorescence (XRF), and scanning electron microscopy (SEM). The samples were successfully loaded with the anticancer drug 5-fluorouracil (5-FU) with loading capacities varying from 22% (for the sodium form having the lower molar SiO 2 /Al 2 O 3 ratio of 15, ZSM-5-(15)) to 43% (for the corresponding acid form, H-ZSM-5-(15)). Percent release of the drug-loaded ZSM-5 samples into simulated body fluid (SBF) was measured at pH 7.4 and 37C. The results showed a slight variation in the % release within the range 84–93%, while the first-order rate constant (k) varied from 2.2 h for ZSM-5-(15) to 3.9 h for H-ZSM5-(15). It was interesting to note that at the higher molar SiO 2 /Al 2 O 3 ratios of 169, both the sodium form, ZSM-5-(169), and the acid form, H-ZSM-5-(169), exhibit an intermediate efficiency in either % loading (38%) or first-order kinetic release constant (k = 2.9 h).


Introduction
Chemotherapy is the most commonly used route for cancer treatment employing organic-and inorganic-based drugs.However, most of these drugs are known for their poor physicochemical properties such as low solubility, low stability, short circulating half-life, and cytotoxicity [1].One of the problematic drugs that have been widely used in anticancer chemotherapy is 5-fluorouracil (Figure 1).5-Fluorouracil (5-FU) is a water-soluble pyrimidine analogue which is widely used in the treatment of various kinds of cancer, especially colon, head, neck, and ovary cancers [2].5-FU can be used in oral, topical, and aerosolized formulations [3].As a derivative of uracil, it rapidly enters the cell where it is converted into active metabolites that disrupt RNA synthesis and acts as inhibitor of thymidylate synthase, a nucleotide synthesis enzyme [4].However, the oral use of 5-FU exhibits a short plasma half-life (30 min) as a result of its rapid enzymatic metabolism [5], which demands continuous high doses thus leading to high toxicity, incomplete and nonuniform oral distribution, drug resistance by tumor cells, and nonselective action against normal cells [6,7].
Different delivery systems have been used to ensure prolonged and sustained release of 5-FU.Examples include hydrogels [8], biodegradable polymers [9], layered inorganic nanocomposites [6], mesoporous organosilica [10], magnetic nanocarriers [7], the clay mineral montmorillonite [11], organic-modified montmorillonite [12], hybrid TiO 2 /ZnO nanotubes [13], porous silica calcium phosphate nanocomposites [14], mesoporous silica systems [15], and thiol-functionalized mesoporous silica nanoparticles [16].Recently, zeolites have been used in biomedical applications due to their biocompatibility and low toxicity [17,18].The small pore size of zeolites (0.3-0.8 nm) that matches the size of small drug molecules, in addition to the tunable properties that can be obtained by varying the molar SiO 2 /Al 2 O 3 ratio, adds more advantages for utilizing zeolites in the medicinal field.Zeolites have been used as delivery systems for different types of drugs, including the adsorption of sulfonamide antibiotics onto zeolite Y [19], the use of NaY zeolite for encapsulation of the anticancer drug -cyano-4-hydroxycinnamic acid [20], the study of sustained release of the anticancer drug doxorubicin from zeolite magnetite nanocomposites [21], and the loading onto and release of aspirin from zeolite HY [22].Recently, both zeolite HY [23] and zeolite NaX-FAU [24] have been explored as delivery systems for 5-FU.
In this study, zeolites ZSM-5 and H-ZSM-5, each prepared in two different molar SiO 2 /Al 2 O 3 ratios (169 and 15), have been investigated as probable drug delivery systems for 5-FU.This choice of ZSM-5 zeolite derivatives for investigation as a possible drug carrier for 5-FU was dictated by their having a promising 3D-channel structure with a suitable pore size, in addition to its reasonable good acid and thermal stability, in comparison with low silica zeolites.

Instrumentation (Physical Measurements
). X-ray powder diffraction spectra were measured using a Philips 2KW model X-ray diffractometer (Cu-K radiation source,  = 1.5418Å) at a scan rate of 2 ∘ C/min.Specific surface areas were estimated through BET modeling using a nitrogen adsorption/desorption instrument (Nova 2200e).UV-Visible spectra were recorded using a Cary 100 Bio UV-Visible spectrophotometer (Varian).FT-IR spectra (4000 to 400 cm −1 , 4 cm −1 spectral resolution, KBr pellets) were measured using a Thermo Nicolet NEXUS 670 FT-IR spectrometer.Thermal gravimetric analysis (TGA) was conducted using a Netzsch Sta 409 PC instrument (Netzsch-Ger) and a Mettler-Toledo DSC 823 E instrument.The morphologies of the solid samples were monitored by scanning electron microscopy (SEM) using an FEI-FEG INSPEC F50 instrument.The chemical composition of samples was determined using an X-ray fluorescence spectrometer (SHIMADZU model XRF-1800).
Solutions 2 and 3 were mixed, and then solution 1 was added dropwise.After vigorous stirring (∼30.0 min), the pH was adjusted to the range 10.3-10.6 by adding concentrated sulfuric acid (98%).The hydrogel formed was transferred into an autoclave and heated for two days at 170 ∘ C. The solid product was filtered and washed with deionized water until the filtrate is sulfate-free (tested by BaCl 2 ).The zeolite was dried at 110 ∘ C for one hour and then calcined at 550 ∘ C for 5 h to allow for the decomposition of the template.

Acid Activation of Zeolite ZSM-5.
The sodium zeolite ZSM-5 was acid activated into H-ZSM-5 using the following typical procedure: a sample of ZSM-5 (1.00 g) was added to a 2.0 M solution of ammonium nitrite (50.0 mL) and the mixture was placed in a thermostated water bath/shaker at 80 ∘ C for 24 h, to obtain the ammonium exchanged zeolite, NH 4 -ZSM-5.The ammonium form of the zeolite was transformed to the acid form (H-ZSM-5) by calcination at 500 ∘ C for 4 h [27].

Loading of 5-FU into ZSM-5. 5-FU was loaded into
samples of ZSM-5 with two different molar SiO 2 /Al 2 O 3 ratios via the following general procedure: 5-FU (≈0.400 g) was dissolved in water (25.0 mL) and then added to a flask containing ZSM-5 (∼0.200 g).The mixture was stirred in the dark, at room temperature, for 24 h.The drug-loaded carrier was filtered and dried under vacuum (30 mmHg) for two hours at 60 ∘ C.
The loading capacities of drug-loaded samples were determined by TGA measurements within the temperature range 25-1100 ∘ C, under nitrogen.The drug loading capacity was also determined by UV-Visible absorption spectrophotometer, in which a predetermined amount of drug-loaded carrier (around 0.0100 g) was stirred with water (100.00mL) for 24 hours at room temperature, the solution was then filtered, and the concentration of 5-FU was determined by measuring the absorbance at 266 nm.The total amount of drug contained in the sample was calculated with reference to a calibration curve.

5-FU Release Profiles.
The in vitro release of each 5-FU-loaded carrier was measured using the dialysis method (dialysis bag diffusion technique).A 5-FU-loaded sample (∼0.0500 g) was introduced into a dialysis bag (MWCO = 3500 Da) containing SBF solution (2.00 mL; pH = 7.4).The sealed dialysis bag was immersed in a flask containing 198 mL SBF (pH = 7.4).The flask was covered and placed in a thermostatic water bath/shaker maintained at 37 ∘ C and a constant shaking rate of 120 rpm.
The concentration of 5-FU released was determined by measuring the absorbance at 266 nm relative to a calibration curve of 5-FU in SBF.The calibration curve covered the concentration range of 0.50 to 20.00 ppm.

Synthesis of ZSM-5 and H-ZSM-5. ZSM-5 (sodium form)
was prepared with two molar SiO 2 /Al 2 O 3 ratios (169 and 15) by using a constant amount of sodium silicate and changing the aluminum content in the hydrogel (Table 1).Zeolite ZSM-5 was then acid activated into H-ZSM-5 via a two-step procedure; the first step involves exchange of the sodium ion by ammonium ion to form NH 4 -ZSM-5, which is then transformed into the acid form (H-ZSM-5) by calcination.

Characterization of ZSM-5 and H-ZSM-5
3.2.1.X-Ray Diffraction.The XRD spectra of ZSM-5 samples having the molar SiO 2 /Al 2 O 3 ratios of 15 and 169 appear almost exactly the same.Similarly, the XRD spectra for the acid activated samples were in agreement with the typical pattern for zeolite ZSM-5, thus confirming that the zeolite keeps its structure and crystallinity upon acid activation as shown in the spectra of ZSM-5-( 15) and H-ZSM-5- (15) depicted in Figure 2.

Nitrogen Adsorption-Desorption Isotherms.
In an ideal case, microporous materials should exhibit type I isotherm according to IUPAC classification [27].A representative nitrogen adsorption-desorption isotherm of ZSM-5 samples is shown in Figure 3.In the relative pressure / 0 range from 0.0 to 1.0, the adsorption increased gradually with relative / 0 due to monolayer adsorption of N 2 onto the microporous channels of ZSM-5.The desorption curve is very close to the adsorption curve in the whole range of / 0 .It was reported that highly siliceous ZSM-5 exhibits two hysteresis loops: one at high relative pressure and the other at low relative pressure, which eventually disappears as the aluminum content increases [28,29].
The specific surface areas of ZSM-5 derivatives were estimated from the adsorption branch using the Brunauer-Emmett-Teller (BET) theory [30].The results (Table 2) indicate a higher surface area for the acid activated sample than that of the corresponding sodium form.It also shows an increase in the surface area with an increase in the molar SiO 2 /Al 2 O 3 ratio, for both forms [31].
As to the effect of variations in acid activation and/or molar SiO 2 /Al 2 O 3 ratio on the specific surface area of ZSM-5 zeolite, the following three observations are evident: (i) For the acid activated form H-ZSM-

Scanning Electron Microscopy (SEM) Micrographs of ZSM-5.
Estimates of the sizes of ZSM-5 and H-ZSM-5 particles, and their morphologies, were obtained through SEM micrographs.All micrographs depict homogeneous distribution of particles, with no significant effect of molar  SiO 2 /Al 2 O 3 ratio or acid activation on their size or shape.Some of the particles have elongated cubic shapes while others have hexagonal prismatic units with particle size distributions in the range of 0.4-0.8m.Examples of these micrographs are depicted for ZSM-5-( 15) and H-ZSM-5-( 15) in Figure 4.

5-FU Loading into Zeolite ZSM-5.
With the dimensions shown in Figure 1, 5-FU is expected to enter the micropores of ZSM-5 readily.5-FU was loaded into samples of zeolites ZSM-5 and H-ZSM-5 with the two different molar SiO 2 /Al 2 O 3 ratios 169 and 15 by the impregnation method.
The optimum conditions to obtain the highest % loading were attained using an aqueous 25.00 mL of 16.00 mg/mL solution of 5-FU and 0.20 g of the zeolite carrier (2 : 1 mass ratio of drug to carrier) and a stirring period of 24 hours.Loading of the drug into the ZSM-5 was confirmed by FTIR-spectrometry and XRD spectra.Inspection of the FTIR spectra of ZSM-5, 5-FU, and loaded ZSM-5, in the region 2000-1500 cm −1 , confirms the presence of the drug in the loaded zeolite.The strong band centered at about 1770 cm −1 in the spectrum of free 5-FU (which is absent from the spectrum of unloaded ZSM-5) appears as a very weak band in the spectra of loaded ZSM-5 and loaded H-ZSM-5.Such drastic decrease in intensity is characteristic of compounds enclosed by porous materials [23].Drug loading was also confirmed by the appearance of the 5-FU characteristic peak at 2 = 29 ∘ in the XRD-spectrum of the drug-loaded ZSM-5 (Figure 5).This is an indication of 5-FU crystallization in the carrier pores.The zeolite loading capacity for 5-FU was determined from the TGA thermograms of pure ZSM-5 and ZSM-5 loaded with 5-FU. Figure 6(a) shows decomposition of 5-FU within the range 280-350 ∘ C (m.p. of 5-FU = 282-286 ∘ C).However, one of the drug-loaded samples, 5-FU-ZSM-5-( 15), lost about 40% of its mass below 400 ∘ C but eventually ended with 50% residual mass at 1100 ∘ C, thus indicating the thermal stability of the silicate carrier.
The zeolite loading capacity was also estimated by UV-Visible spectrophotometry, according to the following equation: % Drug loading = Mass of drug loaded into carrier (estimated by UV absorption) Mass of loaded carrier × 100.In view of data presented in Table 3, a lower molar SiO 2 /Al 2 O 3 ratio plus acid activation of the sodium form of ZSM-5 yields the highest loading capacity for 5-FU.For example, at the lower molar SiO 2 /Al 2 O 3 ratio of 15, acid activation of ZSM-5 doubles the % loading of 5-FU (22% up to 43%).

5-FU Release from Zeolite ZSM-5.
The in vitro cumulative release of 5-FU loaded into three different samples of zeolite ZSM-5 was studied.The effects of variation in the molar SiO 2 /Al 2 O 3 ratio on the release profiles of 5-FU that was loaded into ZSM-5 having a molar SiO 2 /Al 2 O 3 ratio of 15 and also into ZSM-5 having a molar ratio of 169 were measured.In addition, the effect of acid activation of zeolite on the release profile from ZSM-5 and H-ZSM-5 having the same molar SiO 2 /Al 2 O 3 ratio (15) was also measured.The % release was evaluated utilizing the dialysis bag diffusion technique, in which dialysis bags with MWCO of 3500 Da were used.The dialysis bags retain the carrier particles within the bag but allow the released free drug molecules to diffuse into the simulated body fluid (SBF) medium at pH 7.4 and 37 ∘ C.
At predetermined time intervals (as specified in the experimental section), aliquots were withdrawn from the release medium and were readily replaced by fresh samples of SBF at the same conditions, in order to maintain a constant SBF volume.The release experiments were conducted in  triplicate, and corresponding data averages of the three experiments were used for nonlinear regression analysis.The withdrawn samples were analyzed for 5-FU concentration by measuring the absorbance at  max = 266 nm, following appropriate dilution and using a blank sample for base The measured concentration of 5-FU (in ppm) released into the buffer solution was used to estimate the cumulative drug release,   , at time .The cumulative release represents the total amount of drug released (in mg) from the ZSM-5 carrier into the release medium, which was kept at a constant volume of 200 mL.For each sampling time , the amount of drug in all previously withdrawn samples is added to the amount of the drug in the release medium measured at time , thus giving rise to the cumulative drug release,   .The first-order kinetic model was used to simulate the kinetics of the drug release process.Nonlinear regression analysis was used to fit the experimental cumulative drug release (  ) against time , to the first-order drug release model given by the following equation [32]: where   is the cumulative drug release (in mg),   is the maximum amount of drug (in mg) that is released into the medium,  is the release rate constant in h −1 , and  is the sampling time in hours.
Both   and  were used as floating parameters in nonlinear regression of   against   (1 −  − ), and the best parameter estimates, thus obtained, were used to obtain a corresponding estimate of the best data fit.The % release was calculated by the normalization of the cumulative drug release of drug estimated for each time interval (  ) into the total quantity of drug in the loaded ZSM-5 zeolite sample.Plots of % release against time profiles for the three systems are depicted in Figure 7, where solid lines represent the best data fits.
The kinetic parameters for the release of 5-FU from the three carriers ZSM-5-( 15), H-ZSM-5-( 15), and ZSM-5-(169) are listed in Table 4. Analysis of those data indicates that upon acid activation the first-order release rate constant () increases from 2.2 h −1 for ZSM-5-(15) to 3.9 h −1 for H-ZSM-5- (15).The effect of molar SiO 2 /Al 2 O 3 ratio can be seen by comparing the  values for ZSM-5-(15) and ZSM-5-(169), which are 2.2 h −1 and 2.9 h −1 , respectively.This might be attributed to the corresponding decrease of the aluminum content, which is responsible for the stronger attractive interaction forces between 5-FU and the zeolite ZSM-5 pore surface, thus impeding drug release.These results are in agreement with those reported for the release of 5-FU from zeolite HY [23].
The Higuchi square root time model [33], which is based on Fickian diffusion, was also applied to the initial first burst of drug release.This model assumes that the drug % release varies with the square root of time ( 1/2 ), which typically occurs relatively fast compared with subsequent release, and is given by the relation where   is the mass of drug released at time ,   is the maximum mass released,   (h −1/2 ) is the Higuchi constant, and  is a constant characteristic of the drug and host being formulated.Plots of % release of 5-FU, from zeolite ZSM-5, against  1/2 are depicted in Figure 8.
In addition, estimates of the individual contributions of drug diffusion (), and drug erosion (), to 5-FU first burst were obtained through nonlinear regression analysis of data using Kopcha's model [34] according to The / ratios thus obtained (Table 4) indicate that diffusion is the principal mechanism for release of 5-FU from zeolite ZSM-5 with different molar SiO 2 /Al 2 O 3 ratios, both in the sodium and acid activated forms.Table 5 lists the % loading and % release results of 5-FU obtained in this work, in addition to related data on micro-and mesoporous silicates that have been published in the literature.Inspection of the data reveals the following observations: (i) Successful use of zeolite ZSM-5 as a carrier for 5-FU with a relatively high loading capacity (≥40%) can be partially attributed to the close matching between  for Na-ZSM-5-(15) to 2.9 h −1 for Na-ZSM-5-(169) and increases upon acid activation (from 2.2 h −1 for Na-ZSM-5-(15) to 3.9 h −1 for H-ZSM-5-( 15)).(v) Diffusion is the principal process governing 5-FU release from zeolite ZSM-5.

Conclusion
In this research work, zeolite ZSM-5 was synthesized with two different molar SiO 2 /Al 2 O 3 ratios, which were both acid activated.The acid activated ZSM-5 with a molar SiO 2 /Al 2 O 3 ratio of 169 shows the highest % loading and % release.As shown in Table 5, ZSM-5 exhibits about fourfold loading capacity and about twofold % release of 5-FU as compared with zeolite Y.With its relative good thermal and acid stability (compared with low silica zeolites) in addition to its high % loading (∼40%) and % release (∼85%), ZSM-5 may serve as a potential carrier for 5-FU.
5, increasing the molar SiO 2 /Al 2 O 3 ratio from 15 to 169 raises the specific surface area by ≈50% (247 up to 369 m 2 /g).(ii) For the sodium form ZSM-5, increasing the molar SiO 2 /Al 2 O 3 ratio from 15 to 169 raises the specific surface area by ≈70% (194 up to 326 m 2 /g).(iii) For the sodium form ZSM-5, a combination of acid activation and increase in the molar SiO 2 /Al 2 O 3 ratio, from 15 to 169, raises the specific surface area by ≈90% (194 up to 369 m 2 /g).Therefore, a combination of increasing molar SiO 2 /Al 2 O 3 ratio and acid activation of ZSM-5 yields the highest specific surface area.

Table 3 lists
% loading of 5-FU into the sodium form of zeolite (ZSM-5) and the corresponding acid activated form (H-ZSM-5) for each of the two molar SiO 2 /Al 2 O 3 ratios, 169 and 15.Inspection of the table indicates that the estimates of % loading, obtained for each of the four zeolite derivatives from TGA thermograms and UV absorption, are equivalent within experimental error.

Table 3 :
The measured % loading of 5-FU into the four zeolite samples.

Table 5 :
A comparison of % loading, % release, and kinetic model parameter data of 5-FU obtained in this work, as well as related data on siliceous carriers published literature, for comparison.5-FUkinetic release models' data Siliceous material  /  = 1 −  −   /  =    1/2 +  Enhancements on the loading capacity and the firstorder rate constant () were observed upon acid activation of ZSM-5 and also upon increasing the molar SiO 2 /Al 2 O 3 ratio from 15 to 169.(iii) A relatively high % release of 5-FU was achieved with the highest % release exhibited by ZSM-5 having a molar SiO 2 /Al 2 O 3 ratio of 169 (≥85%).(iv) The first-order rate constant () increases with increasing the molar SiO 2 /Al 2 O 3 ratio from 2.2 h −1