1. Introduction
Caused by the pathogenic agent
Mycobacterium tuberculosis, tuberculosis (TB) is an infectious and highly transmissible disease that can affect not only the lungs (pulmonary TB) but also other organs of the body [
1,
2]. Despite being curable, until the arrival of the coronavirus (COVID-19) pandemic, TB was the leading cause of death from a single infectious agent worldwide. According to data from the World Health Organization (WHO), in 2021, about 10.6 million people were diagnosed with TB, an increase of 4.5% compared to 2020, and about 1.4 million people died as a result of the disease in the same year (WHO) [
3].
The alarming data cited above is due to problems associated with the currently recommended therapeutic regimen, which consists of chemotherapy based on the administration of four antibiotics: isoniazid, rifampicin, pyrazinamide, and ethambutol [
2,
4,
5]. These drugs are administered in high doses and for a prolonged period of time, which, depending on the patient’s response to treatment, can last up to 9–12 months, causing the appearance of serious side effects. Another worrying factor is the development of bacterial resistance to these antibiotics, which can lead to the emergence of multidrug-resistant TB. This problem has been declared a global health emergency by the WHO and may jeopardize disease control worldwide [
1,
2,
5].
Rifampicin (RIF) is the main first-line drug used in the treatment of TB and is also used in multidrug therapy and the prophylaxis of leprosy. Despite having high bacteriostatic activity against
M. tuberculosis, RIF has low bioavailability, mainly due to its low solubility and occurrence of polymorphism. Furthermore, RIF undergoes degradation in the acidic environment of the stomach, resulting in the formation of the degradation product 3-formyl rifamycin SV (3-FRSV). It can also undergo autoxidation, turning into rifampicin quinone. Therefore, it is necessary to administer high doses (approximately 10 mg/kg of the patient’s weight) for the drug to reach its desired therapeutic efficacy, causing the appearance of potentially toxic side effects, such as hepatitis, anorexia, skin eruptions, and renal failure [
6,
7,
8,
9].
Among the strategies reported in the literature to overcome these difficulties and improve the therapeutic efficacy of anti-TB drugs is to devise modified drug delivery systems (MDDS). Several works in the literature have reported obtaining MDDS for RIF, using the most diverse types of carriers, such as polymeric nanoparticles [
10,
11,
12], microemulsions [
13], dendrimers [
14,
15], hydrogels [
6,
16,
17], liposomes [
18], and lipid nanoparticles [
19], among others. Despite being effective, such systems are based on expensive technologies that require the use of improved materials and/or devices. For TB, a disease that mainly affects the poorest of populations in underdeveloped countries, the search for MDDS that are effective, safe, and obtained from abundant and cheap materials is still necessary, since until recently, tuberculosis was considered a neglected disease [
20].
Nanotechnology plays an important role in the development of new drug delivery systems. Systems based on nanoparticles allow delivery at lower doses, increase the solubility of the drug, improve its bioavailability, and thus minimize the side effects associated with its administration [
21,
22]. Inorganic nanoparticles (clay minerals, magnetic nanoparticles, silica, and others) have gained prominence in nanomedicine, mainly in disease diagnosis and drug delivery, due to their unique properties such as high stability, optical and magnetic properties, biocompatibility, biodegradability, and low cost [
23,
24].
Clay minerals are widely used in the pharmaceutical industry, as an active ingredient or excipient, in formulations for topical and oral use [
25]. Recently, this type of material has attracted great interest from the scientific community for its potential application as a carrier of active substances. This interest is intrinsically linked to its physicochemical properties, such as adequate surface area, considerable adsorption capacity, and chemical inertness. In addition, clay minerals are biocompatible, abundant, and low-cost [
26,
27,
28]. According to García-Villén et al. [
29], clay minerals allow multiple pathways for interactions with microorganisms, decreasing water activity and cell adhesion, proliferation, and the formation of neotissues. All these characteristics make these nanoparticles viable to be explored in obtaining effective anti-TB drug delivery systems, making them promising candidates for improving therapeutic efficacy in disease treatment, including the treatment of multidrug-resistant infections [
30].
Systems based on clay–drug nanohybrids have been widely reported in the literature, presenting interesting characteristics that have increased the bioavailability of several biomolecules, such as increased solubility of hydrophobic drugs, protection against degradation along the gastrointestinal tract, controlled release, and being pH responsive [
20,
31,
32,
33]. These systems are obtained through the establishment of interactions between the active sites of the nanocarrier and the functional groups present in the drug molecule. They can be prepared using several different methods, including: encapsulation, immobilization, ion exchange, and electrostatic interactions [
33]. We have previously described how to obtain this type of system for the drugs isoniazid and rifampicin using the clay mineral palygorskite [
4,
20]. Other systems for anti-TB drugs, based on montmorillonite [
34], halloysite [
30,
35,
36], and palygorskite [
1,
37], have also been reported.
Montmorillonite (Mt) is a clay mineral of the 2:1 type and belongs to the smectite group. It has an organized structure in the form of layers, with a net negative charge between −0.2 and −0.6, resulting from random isomorphic substitution processes in the sheets of tetrahedra and octahedra. To balance this negative charge, Mt presents in its interlamellar region (between layers), exchangeable cations such as Na
+, K
+, Ca
2+, and Mg
2+, which are usually hydrated. This type of clay mineral swells easily, causing the formation of gels with well-defined rheological properties that exhibit a pseudo-plastic behavior. In addition, it has a cation exchange capacity (80–150 mEq/100 g) and a high specific area. All these characteristics make it possible for Mt to be applied in the development of new hybrid devices for modified drug delivery [
38,
39,
40]. The ions present in the interlamellar region of montmorillonite are easily exchangeable for organic molecules, such as drugs, through cation exchange processes. Therefore, it is possible to obtain clay–drug hybrids through the intercalation of these bioactive molecules between the layers that make up the structure of the clay mineral [
41,
42]. In addition, Mt has surface silanol (Si-OH) groups that undergo different types of interactions with organic molecules, such as hydrogen bonds, ion–dipole interactions, coordination bonds, acid–base reactions, electrostatic attraction, and van der Waals interactions [
43,
44].
The main objective of this work is to obtain an Mt–RIF nanohybrid for the modified release of the anti-TB drug, as a way of overcoming the problems associated with its oral administration. As far as we know, this is the first work that reports the acquirement of this type of system for rifampicin that uses only montmorillonite as a carrier. The intercalation of the drug was evaluated through an experimental design, in order to investigate the influence of some variables in this process. The obtained hybrid was characterized by different techniques in order to further elucidate the clay–drug interaction. In vitro release studies were carried out to evaluate the performance of the obtained pH-responsive system.
2. Materials and Methods
2.1. Materials
Rifampicin was provided by NUPLAM-UFRN (Núcleo de Pesquisa em Alimentos e Medicamentos, Natal, Brazil) and supplied by the pharmaceutical company Sanofi, presenting a purity of 99.24%. The montmorillonite (Cloisite® Na+, BYK Additives & Instruments, Wesel, Germany) used in the present work is of the sodium type—containing Na+ ions in the interlamellar region—and was provided by the company Colormix Especialidades (Barueri, Brazil). The other reagents used are analytical grade: ethyl alcohol (64-17-5, Dinâmica Química Contemporânea LTDA, Indaiatuba, Brazil), hydrochloric acid (37% p/p, 7647-01-0, Vetec, Duque de Caxias, Brazil), sodium hydroxide (1310-73-2, Vetec, Brazil), L-ascorbic acid (50-81-7, Labsynth, Diadema, Brazil), sodium chloride (7647-14-5, Vetec, Brazil), and monobasic sodium phosphate (10049-21-5, Dinâmica Química Contemporânea LTDA, Indaiatuba, Brazil).
2.2. Experimental Design
To evaluate the efficiency of drug incorporation into the clay mineral, a factorial design 2
4 of the central composite design (CCD) type was used, consisting of sixteen experiments and three repetitions at the central point (mean level). The variables studied that could influence the RIF intercalation process in the Mt structure were as follows: Mt mass, RIF solution concentration, pH of RIF solution, and contact time. The volume of solution was fixed at 100 mL. The planning, estimation of the effects, and validation of the built model were obtained using Protimiza Experimental Design software (Campinas, Brazil).
Table 1 presents the values used in the CCD for the four variables described above.
The incorporation occurred through an adsorption process and followed the methodology described by Oliveira, Alcântara, and Pergher [
45], with some adaptations. Initially, the corresponding mass of Mt was weighed in a 250 mL Erlenmeyer flask. Then, 20 mL of deionized water was added. The mixture remained under constant agitation (180 rpm) at 20 ± 0.5 °C in an incubator for two hours to allow for the expansion of the clay mineral. 80 mL of a 40% (
v/
v) hydroethanolic solution of the drug was added, whose pH was adjusted by adding 1.0 mol/L HCl or 1.0 mol/L NaOH solutions. The solutions contained ascorbic acid at a concentration of 1 mg/mL to avoid RIF degradation during the experiments [
4,
46]. The flasks remained under agitation in an incubator, under the same conditions of rotation speed and temperature, for the pre-established period of time. At the end of each experiment, a 15 mL aliquot of the suspension was transferred to a Falcon tube and centrifuged for 15 min at 4000 rpm. The absorbance of each aliquot was determined using a UV/Vis spectrophotometer (UV-1800, Shimadzu, Kyoto, Japan) at 475 nm. Obtaining and validating the calibration curve followed the parameters in the ICH Harmonized Tripartite Guideline-Validation of Analytical Procedures [
47]. The detection limit and quantification limit values obtained were 0.00023 and 0.00069 mg/mL respectively, indicating that the drug can accurately be detected and quantified by the technique, even at low concentrations [
20]. The incorporated drug dose was determined by the equation below (Equation (1)) [
48]:
where
C0 is the initial concentration of RIF (mg/mL),
C is the concentration of RIF in solution (mg/mL) after the incorporation process,
VRIF is the volume of RIF solution (mL), and
MMt is the mass of Mt (mg).
To validate the obtained model, the process was repeated under the optimized conditions in triplicate. The Mt–RIF nanohybrid was prepared under the same conditions. After adsorption, the powder was separated by vacuum filtration and dried in an oven at 60 °C for six hours. Then, the dried material was pulverized, sieved through a 100-mesh sieve, and stored in a desiccator for later use.
2.3. Characterization of Materials
The materials (RIF, Mt, and Mt–RIF) were characterized using X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), thermogravimetry (TGA), and differential scanning calorimetry (DSC) techniques.
For structural characterization using the XRD technique, a Bruker D2 Phaser device (Karlsruhe, Germany) was used with the following specifications: CuKα radiation (λ = 1.54 Å), Ni filter, 0.02° step, 10 mA current, voltage 30 kV and Lynxeye detector.
FTIR spectra were obtained on a Shimadzu FTIR-8400S Iraffinity-1 spectrophotometer (Kyoto, Japan), with the following specifications: 32 scans, analysis range of 400–4000 cm−1, and resolution of 4 cm−1. Samples were prepared on KBr pellets.
TGA analyses were carried out on a Shimadzu TGA-50 thermobalance (Kyoto, Japan) using the following analysis conditions: platinum crucible, nitrogen as purge gas, gas flow rate of 50 mL/min, heating rate of 10 °C/min, and final analysis temperature of 900 °C.
DSC analyses for the samples were performed on a TA Instruments model Q20 instrument (New Castle, DE, USA). The analysis conditions were: gas flow rate of 50 mL/min, inert atmosphere (nitrogen gas), heating rate of 10 °C/min, and aluminum crucible. For all analyses, a mass of approximately 2 mg of sample was used.
Scanning electron microscopy (SEM) images were obtained using a TESCAN MIRA 4 scanning electron microscope (Brno–Kohoutovice, Czech Republic), using an in-beam SE secondary detector with 10 KeV energy. The samples were covered with gold film using a Denton Vacuum model Desk V vaporizer for 60 s and dispersed on carbon tape for high vacuum analysis.
2.4. In Vitro Release Studies and Release Kinetics
The in vitro release studies were carried out sequentially in four different dissolution media, in order to simulate the different pH values in the transit of a pharmaceutical form for oral use along the gastrointestinal tract. To simulate the conditions in the stomach region, the tests were performed for two hours in two different fluids: pH 1.2 (empty stomach) and pH 4.0 (‘fed’ stomach). Tests for simulated release in the small intestine were performed in a pH 6.8 fluid for four hours. Finally, to simulate the physiological conditions of the intestinal colon, a pH 7.4 fluid was used for ten hours [
45,
49,
50].
All tests were carried out in a dissolving apparatus (model 800-2TS, Ethik Technology, São Paulo, Brazil), basket-type apparatus, at a constant speed of 100 rpm and temperature of 37 ± 0.5 °C. Five hundred micrograms of the Mt–RIF hybrid (approximately 50 mg of RIF) was encapsulated in size #00 gelatin capsules. Six hundred microliters of dissolution medium was used, maintaining the sink condition. After each time interval, 5 mL aliquots were removed, and subsequently, the withdrawn volume was replaced with fresh dissolution medium. The absorbance of each aliquot was determined using a UV/Vis spectrophotometer at 475 nm. Preparation of dissolution media followed the methodology described by Almeida et al. [
50]. To prepare the dissolution medium at pH 1.2, 1 g of sodium chloride (NaCl) and 70 mL of concentrated hydrochloric acid (HCl) were dissolved in 1000 mL of deionized water. For the preparation of the pH 4 dissolution medium, 2.99 g of tribasic sodium acetate (NaC
2H
3O
2•3H
2O) and 1.66 mL of concentrated acetic acid (CH
3COOH) were dissolved in 1000 mL of deionized water. The pH was adjusted using a 1.0 mol/L HCl solution. For the preparation of pH 6.8 dissolution medium, 0.3 g of sodium hydroxide (NaOH), 4 g of monobasic sodium phosphate (NaH
2PO
4•H
2O), and 6.2 g of NaCl were dissolved in 1000 mL of deionized water. The pH 7.4 dissolution medium was prepared in the same way, with its pH adjusted by adding 1 mol/L NaOH. Ascorbic acid was added to each dissolution medium at a concentration of 1 mg/m to avoid oxidative degradation of RIF during release studies [
46,
51].
To evaluate the release mechanism, the experimental data were fitted to zero-order, first-order, Higuchi, and Korsmeyer–Peppas kinetic models. The general formulas of each model are shown in
Table 2.
4. Discussion
Through the experimental planning, it was possible to evaluate the influence of important parameters that affect the intercalation process of the drug in the structure of the clay mineral. In addition to minimizing the number of experiments, and consequently, reducing operating costs and time, this methodology allowed the construction of an empirical, statistically validated model that is capable of predicting the dose incorporated from the prior establishment of levels for the variables considered significant [
54]. This model can also help in the design process of pediatric formulations, which generally require an adaptation of the incorporated drug dose.
As previously reported, isomorphic substitution processes occur in the tetrahedral and octahedral sheets of Mt, resulting in a net negative charge. To compensate for this charge, the structure of this clay mineral presents hydrated cations in the interlamellar region, which are easily exchangeable in a cation exchange process [
27]. This is the physicochemical characteristic that supports the process of incorporation of drugs and other organic molecules into the Mt structure, which can be adsorbed in the interlamellar region or on the external surface. It can then be stated that the adsorption of RIF on Mt is a highly pH-dependent process, since the ionization state of the drug plays a significant role in the cation exchange process, directly impacting the performance of drug retention in the clay mineral [
53,
66]. Rifampin has two pK
a values. The first, 1.7, is related to the phenolic hydroxyl groups at C1, C4, and C8. The second value, 7.9, is associated with the nitrogen atoms of the piperazine ring [
67]. According to the calculation presented in a previous work, RIF presents 100% of its molecules ionized at pH 2 due to the protonation of the nitrogen atoms in the piperazine ring. Thus, the drug exists predominantly in its cationic form (RIF+). At pH 6, there was no intercalation of the drug in the structure, since only 1.24% of the molecules were in the form of RIF+ [
4]. This is why, as the pH of the drug solution decreases, the incorporated drug dose increases, as the cation exchange process actually takes place.
Similar results were presented by Afarani, Sarvi, and Alavijeh [
68]. The authors studied the encapsulation process of vitamin B6 in montmorillonite. At pH 6.5, the vitamin was in its protonated form. Thus, its molecules began to occupy the interlamellar region and incorporation was governed by a cation exchange process. When the pH was increased to 9.5, the molecules were in their deprotonated form and began to interact with the edges of the clay mineral, which have a more negative charge. Bello et al. [
69] studied the interaction of several drugs, all containing amine groups, with the sodium Mt structure. The authors demonstrated that the molecules in their protonated form replace sodium ions in the interlamellar region; therefore, the process is governed predominantly by cation exchange.
Regarding the mass, it was observed that the lower its value, the higher the drug dose that is incorporated. Mt has a high cation exchange capacity, being able to retain large quantities of organic molecules [
45]. This result is interesting from an economic point of view, as it is possible to obtain higher incorporated drug dose values using a reduced amount of carrier. Similar results were found in works previously published by the group [
4,
20]. Belaroui et al. [
70] also reported similar observations when evaluating the adsorption of an herbicide on palygorskite clay minerals. The authors showed that maximum adsorption capacity was reached with smaller masses of the adsorbent. Furthermore, as adsorbent mass increased, the amount of adsorbed herbicide tended to plateau.
Regarding the concentration, the response of interest is maximized when the concentration of RIF in solution is increased. This result is due to the fact that, when increasing the concentration, there is an increase in the driving force of mass transfer, facilitating the movement of molecules towards the active sites of the clay mineral [
71,
72]. Li et al. [
48] studied the incorporation of dexibuprofen in pure montmorillonite and in acid-modified montmorillonite. In both cases, the drug load gradually increased with increasing concentration, until a saturation point was reached.
The solid-state characterization of the Mt–RIF hybrid is of paramount importance in proving the obtainment of the hybrid and in elucidating the interaction mechanism between the materials. As observed in the structural analysis by XRD, there is a displacement of the reflection (001) from Mt to smaller angles in 2θ, with an increase in the basal spacing d
001, which went from 11.7 Å to 15.6 Å. This result clearly indicates the intercalation of RIF molecules in the interlamellar region of Mt. Similar results were reported in works in the literature that studied the intercalation of other molecules in the montmorillonite structure [
28,
29,
48,
66,
69]. Also, the diffraction peak becomes more intense and defined. According to García-Villén et al. [
29], this result is an indication that the structure of the nanohybrid is highly ordered. The disappearance of the reflection at 28° in 2θ may be associated with the process of Mt dealumination, caused by the addition of acid used in the pH regulation of the RIF solution. This process is characterized by the leaching of octahedral cations and the formation of amorphous silica, leading to partial damage to the clay mineral structure. This favors the formation of pores, and consequently increases the drug loading capacity [
48,
73]. The absence of peaks referring to the crystalline structure of RIF in the Mt-RIF diffraction pattern (
Figure 4b) indicates amorphization of the drug during the cation exchange process, in addition to confirming the absence of RIF crystals on the surface of the nanohybrid [
29,
48]. Such observations are also confirmed by DSC and SEM analysis.
The FTIR analyses indicated the functional groups involved in the interaction process between the materials. By analyzing the spectrum obtained for the Mt–RIF hybrid, it is observed that there was no change in the positions of the bands, suggesting that the adsorption process occurs preferentially in the interlamellar region, with weak interaction on the outer surface [
74]. According to Silverstein et al. [
57], the increase/decrease in band intensity or splitting are indications of interaction between materials. Therefore, the changes in certain bands observed in the spectrum obtained for the nanohybrid suggest the establishment of hydrogen bond-type interactions between the nitrogen atoms of the RIF molecule and the OH of the water present in the interlamellar region [
73]. Such results indicate the coordination of drug molecules to silanol (Si-OH) groups on the surface, confirming the effective inclusion of RIF molecules in the interlamellar region of Mt, with drug solubilization in the carrier structure [
35,
75].
From the TGA curve obtained for the Mt–RIF hybrid (
Figure 6b), it was possible to confirm the presence of RIF in the structure. It is observed that the first mass loss, referring to the first stage of Mt dehydration, was smaller, corresponding to approximately 3%. Soon after, a mass loss corresponding to 15% of the sample begins in the temperature range between 15 °C and 750 °C. This event corresponds to the thermal degradation of the incorporated RIF, followed by the Mt dehydroxylation process. In the DSC curve (
Figure 6d), only the endothermic peak referring to the first dehydration of Mt is observed, without the appearance of the peak referring to the melting of RIF. The absence of this peak indicates a change in the crystalline form of RIF after interaction with the clay mineral, as confirmed by XRD analysis [
28].
From the SEM images obtained for the Mt–RIF hybrid (
Figure 7b), we can see that the drug incorporation process does not change the morphology of the nanocarrier. The absence of drug crystals on the surface corroborates the results obtained by the other characterization techniques, indicating an amorphization of RIF and good homogenization of the drug in the nanocarrier structure [
48].
The in vitro release studies demonstrated the ability of montmorillonite to protect the drug molecules from degradation in the stomach environment since there was no release in conditions that simulated this environment. In this case, the drug molecules remain in the form of RIF+, occupying the interlamellar region of Mt. According to Damasceno-Junior et al. [
4], at acidic pH values, the oxygen atoms of the silanol groups and water molecules act as bond acceptors, while the protonated nitrogens of the piperazine ring act as donor atoms. Therefore, as oxygen is more electronegative, the hydrogen bond established between materials is stronger [
50]. In the intestinal environment, RIF molecules are deprotonated, being gradually replaced by cations present in the dissolution fluid through a process governed by the cation exchange mechanism [
45]. The hydrogen bond becomes weaker since the nitrogen atoms, which are less negative, start to act as acceptors [
4,
20].
The results demonstrate that montmorillonite can, in this specific case, be used alone as a carrier, without the need for modification or combination with other materials (obtaining a composite). The system was able to protect the drug from the degradation suffered by RIF in the stomach, demonstrating that the hybrid increases drug stability. In addition, RIF was released in a controlled manner in the simulated intestinal fluid, the preferential environment for the absorption of orally administered drugs [
75]. Such results promote a significant improvement in the bioavailability of RIF, meaning that the patient would need a lower dose for therapeutic efficacy to be achieved. The prolonged-release also makes it possible to reduce the frequency of dosing, thereby minimizing or eliminating any associated side effects [
76,
77,
78]. The
n exponent values obtained by adjusting the Korsmeyer–Peppas model were 0.275 and 0.268 for pH 6.8 and pH 7.4, respectively, characterizing a release that can be described as a diffusion process based on Fick’s law (
n < 0.5) [
39,
52,
79]. These results indicate that the RIF release process is diffusion-controlled, whereas Higuchi’s model also describes the release as a diffusion process, without matrix erosion or swelling [
79]. The cations present in the release medium diffuse, promoting cation exchange with the drug molecules present in the interlamellar region [
48].