An Aerosol Nanocomposite Microparticle Formulation Using Rifampicin-Cyclodextrin Inclusion Complexes for the Treatment of Pulmonary Diseases

Abstract

Rifampicin (RIF) is commonly used in the treatment of tuberculosis (TB), a bacterium that currently infects one fourth of the world’s population. Despite the effectiveness of RIF in treating TB, current RIF treatment regimens require frequent and prolonged dosing, leading to decreased patient compliance and, ultimately, increased mortality rates. This project aims to provide an alternative to oral RIF by means of an inhalable spray-dried formulation. TB uses alveolar macrophages to hide and replicate until the cells rupture, further spreading the bacteria. Therefore, delivering RIF directly to the lungs can increase the drug concentration at the site of infection while reducing off-site side effects. Cyclodextrin (CD) was used to create a RIF-CD inclusion complex to increase RIF solubility and biodegradable RIF-loaded NP (RIF NP) were developed to provide sustained release of RIF. RIF NP and RIF-CD inclusion complex were spray dried to form a dry powder nanocomposite microparticles (nCmP) formulation (RIF-CD nCmP). RIF-CD nCmP displayed appropriate aerosol dispersion characteristics for effective deposition in the alveolar region of the lungs (4.0 μm) with a fine particle fraction of 89%. The nCmP provided both a burst release of RIF due to the RIF-CD complex and pH-sensitive release of RIF due to the RIF NP incorporated into the formulation. RIF-CD nCmP did not adversely affect lung epithelial cell viability and RIF NP were able to effectively redisperse from the nCmP after spray drying. These results suggest that RIF-CD nCmP can successfully deliver RIF to the site of TB infection while providing both immediate and sustained release of RIF. Overall, the RIF-CD nCmP formulation has the potential to improve the efficacy for the treatment of TB.

Graphical Abstract

1. INTRODUCTION

Tuberculosis (TB) is a highly contagious infectious disease caused predominantly by the pathogen, Mycobacterium tuberculosis. Once inhaled, this pathogen deposits in the alveolar region of the lungs where it is phagocytized into and replicated within alveolar macrophages. Alveolar macrophages typically easily degrade phagocytized bacteria in their acidic lysosomes; however, TB is capable of evading lysosomal degradation and infecting additional macrophages (O’Hara and Hickey, 2000; Upadhyay et al., 2018). Once a macrophage is infected with TB, the pathogen hijacks normal signaling pathways to prevent the formation of lysosomes, thereby preventing apoptosis, which initiates macrophage necrosis. Following necrosis, TB infects additional macrophages and continually repeats this infection/necrosis cycle, potentially infecting other organs (Upadhyay et al., 2018). The effectiveness of TB to evade the innate and adaptive immune responses of macrophages has resulted in TB being a leading cause of death worldwide with millions of new cases diagnosed each year (WHO Global TB Report, 2021).

The current treatment for TB entails a six-to nine-month daily administration of an assortment of drugs, often including the first-line antibiotic, rifampicin (RIF). The efficacy of RIF in treating TB has been shown to be concentration dependent as it inhibits bacterial DNA-dependent RNA polymerase by binding to the β subunit of RNA polymerase, physically blocking the elongation of RNA (He et al., 2013; Pang et al., 2013). The required length and frequency of dosing in the treatment of TB has led to poor patient compliance, resulting in increased patient mortality rates and a rise in multidrug-resistant TB. Furthermore, current treatment options are typically delivered orally and can result in off-site side effects associated with liver or kidney toxicity (Maguire et al., 2009).

The limitations of current treatment regimens and dosage forms for TB-related infections has led to the need for the development of alternative therapeutics, including aerosol particle formulations delivered directly to the site of infection. The lungs are unique in comparison to other routes of administration, as they possess an extremely large surface area (approximately 100 m²), with only a single layer of cells in the alveolar region and very low enzymatic activity. Importantly, drugs absorbed through the lungs are not subject to the first-pass effect that occurs when delivered orally. The large surface area and facile accessibility of aerosol therapeutics to alveoli results in a rapid onset of action and high bioavailability (Bhat et al., 2006; Labiris and Dolovich, 2003; Shoyele and Cawthorne, 2006). Furthermore, direct delivery of drug to the lungs can reduce off-site side effects, decrease dosing requirements, and improve treatment efficacy with improved pharmacokinetics (Mangal et al., 2017).

The bioavailability of drugs in formulations can be further increased through the use cyclodextrins (CD), which are cyclic oligosaccharides with a hydrophilic exterior and a hydrophobic interior (see Figure A1). CD are non-toxic and have the ability to form inclusion complexes with other molecules, which has been shown to increase the solubility of poorly water-soluble drugs, such as RIF (He et al., 2013; Jeon et al., 2018; Tewes et al., 2008; Williams et al., 2013). CD have also been shown to enhance the permeation of drugs through biological barriers (Patil and Suresh, 2009). Drug release from CD inclusion complexes is dilution controlled and as a result, the complexed drug can be released rapidly (Stella et al., 1999), which is beneficial for quickly reaching effective therapeutic levels but detrimental for achieving sustained drug release which can dosing frequency.

Controlled and sustained drug release can be achieved through the use of drug-loaded nanoparticle (NP) formulations. Acetalated dextran (Ac-Dex), a biodegradable and biocompatible polymer, has been used extensively in particulate delivery systems (Bachelder et al., 2017; Shah et al., 2019a; Wang et al., 2017). Ac-Dex has a simple synthesis process that allows for tunable degradation rates that correlate to its synthesis, where shorter reaction times result in a higher ratio of acyclic acetal groups and longer reaction times result in a higher ratio of cyclic acetal groups. By adjusting the ratio of cyclic (slower degrading) to acyclic (faster degrading) acetal groups, the degradation can be adjusted from hours to days. Furthermore, Ac-Dex degradation is pH-sensitive, with faster degradation occurring in lower pH conditions (Bachelder et al., 2017, 2008; Broaders et al., 2009). It has been shown that Ac-Dex NP are capable of providing sustained drug release at neutral pH and rapid drug release at lower pH (Gupta et al., 2017; Jakaria et al., 2021; Torrico Guzmán et al., 2019), such as that experienced in a phagolysosome of an alveolar macrophage (O’Hara and Hickey, 2000; Shah et al., 2019a).

To achieve effective delivery to the lungs, formulations must have appropriate aerosol dispersion properties. Particles that are too large (> 10 μm) will deposit in the oropharyngeal region prior to reaching the lungs, whereas particles that are too small (< 500 nm) will mostly be exhaled (Labiris and Dolovich, 2003). For deposition in the alveolar region, particles with aerodynamic diameters between 1–5 μm are ideal (Labiris and Dolovich, 2003; O’Hara and Hickey, 2000; Smaldone et al., 1988). Spray drying is a one-step, scalable process that has been used to create inhalable dry powder microparticles capable of deposition in the alveolar region (Mansour et al., 2013; Meenach et al., 2014) and spray drying parameters can be easily tuned to modify particle characteristics (e.g., particle size) to meet specific formulation needs.

Delivery of NP to the alveolar region of the lungs is vital in allowing NP to avoid mucociliary clearance in the upper airways (Hickey, 2020; Labiris and Dolovich, 2003; Mangal et al., 2017). Phagocytosis by alveolar microphages is the predominant clearance mechanism in the alveolar region, which has a very thin mucus layer, allowing for deposition of particles on the epithelial lining (Labiris and Dolovich, 2003). NP on their own are not able to effectively deposit in the alveolar region upon aerosolization to the lungs. Spray-dried nanocomposite microparticle (nCmP) formulations incorporate NP and excipients into a microparticle system to the achieve desired particle size (1–5 μm) for effective pulmonary delivery (Elsayed and Aboughaly, 2016; Wang et al., 2017, 2016), allowing for deposition of NP to the lungs following nCmP dissolution.

In the present study, inhalable dry powder RIF-loaded nCmP formulations were developed as a pulmonary drug delivery platform for the potential treatment of TB. RIF-loaded NP were spray dried with RIF-CD inclusion complex to formulate RIF-CD nCmP, with the inclusion complex replacing typical spray drying excipients such as mannitol. The RIF-CD nCmP formulation was able to deliver a high payload of RIF with a unique dual release profile, with rapid RIF release from the RIF-CD complex and controlled RIF release from the RIF NP, in addition to the potential to be effectively aerosolized to the site of TB infection within the alveolar region of the lungs. To our knowledge, this is the first time RIF has been released from a nCmP formulation that allows for a dual release profile for the potential treatment of TB-related infection. The developed RIF-CD nCmP formulations were evaluated for their physicochemical properties, aerosol dispersion, and cytotoxicity to demonstrate their use as a drug delivery platform for the treatment of pulmonary diseases such as TB.

2. MATERIALS AND METHODS

2.1. Materials

Dextran from Leuconoctoc mesenteroides (9–11 kDa), phosphate buffered saline (PBS) powder, pyridinium p-toluenesulfonate (PPTS, 98%), D-mannitol (≥ 98%), 2-methoxypropene (2-MOP, 97%), triethylamine (TEA, ≥ 99%), anhydrous dimethyl sulfoxide (DMSO, ≥ 99.9%), resazurin powder, and methanol (Chromasolv^™, ≥ 99%) were purchased from Sigma Aldrich (St. Louis, MO). 2-hydroxypropyl-β-cyclodextrin (β-CD, 1,400 Da) and 2-hydroxypropyl-γ-cyclodextrin (γ-CD, 1,540 Da) were generously provided by Wacker Chemical Corporation (Adrian, MI). Rifampicin (RIF) was purchased from TCI Chemicals (Portland. OR). Polyvinyl alcohol (PVA, 88% hydrolyzed, average MW 22 kDa) was obtained from Acros Organics (Geel, Belgium). Dulbecco’s Modified Eagle Medium (DMEM), penicillin-streptomycin (Pen-Strep), and Fungizone^® were purchased from Life Technologies (Norwalk, CT). Trypsin-EDTA, sodium pyruvate, and Dulbecco’s PBS were purchased from Fisher Scientific (Waltham, MA). Fetal bovine serum (FBS) was purchased from Atlanta Biologics (Flowery Branch, GA). A549 lung adenocarcinoma cells were obtained from American Type Culture Collection (ATCC, Manassas, VA).

2.2. Synthesis and Characterization of Acetalated Dextran

Acetalated dextran (Ac-Dex) was synthesized as previously described, as seen in Figure A1 (Z. Wang and Meenach, 2016). Lyophilized dextran (1 g) and (30 mg) were dissolved in anhydrous DMSO (10 mL) in a vial purged with nitrogen. After the addition of 2-MOP (5 mL), the reaction was carried out for 3 hours before being quenched with TEA. The polymer was precipitated in a spinning solution of basic water (pH 9.0), vacuum filtered, lyophilized, and stored at −20 °C in a desiccator.

2.3. Phase Solubility Analysis

A RIF phase solubility study was carried out in aqueous solution containing either β-CD and γ-CD. Excess RIF was added to each CD solution at concentrations ranging from 0 to 70 mM at 37 °C for 48 hours to allow for saturation. The concentration of RIF in the solutions was determined via absorbance at 474 nm using a BioTek Cytation 3 microplate reader (BioTek Instruments Inc., Winooski, VT). CD versus RIF concentrations were plotted to create a phase solubility diagram, and a linear regression line was fit to each RIF/CD combination. Following the methodology described by Higuchi and Connors, stability constants (K_1:1) were calculated for the complexes using the following equation (Higuchi, 1967):

$$K_{1:1}=\frac{\mathit{\text{slope}}}{S_0(1-\mathit{\text{slope}})}$$

where S₀ is the solubility of RIF in pure water and the slope was determined from the linear regression fit on the phase solubility diagram.

2.4. Synthesis of Rifampicin-Loaded Nanoparticles

RIF-loaded nanoparticles (RIF NP) were synthesized via an organic-in-water (O/W) emulsion solvent evaporation method (see Figure A1). RIF (10 mg) and Ac-Dex (40 mg) were dissolved in DCM (1 mL) to create the organic phase and the aqueous phase consisted of 6 mL of 3 % (w/v) PVA solution in PBS. The aqueous phase was added to the organic phase and probe sonication was performed for 30 seconds (1 second on/off pulse) at 70 % amplitude to form the O/W emulsion. Following sonication, the emulsion was transferred to a spinning solution of 0.3 % (w/v) PVA and stirred for 4 hours to allow for solvent evaporation and particles hardening. The RIF NP were purified via centrifugation at 19,802 × g for 20 minutes and washed twice with water. Following the second centrifugation, the NP were redispersed in 0.05 % (w/v) PVA, frozen, lyophilized for 24 hours, and stored in a desiccator at −20 °C until later use.

2.5. Preparation of Nanocomposite Microparticles and Microparticle Formulations

The microparticle formulations were fabricated via spray drying, including nCmP and solid-phase microparticles (MP). As seen in Figure 1, the formulations were comprised of RIF NP, RIF-CD inclusion complex, mannitol, or raw RIF. RIF-CD inclusion complex was formed in water under mixing at a 1:1 molecular ratio of RIF and β-CD. To formulate the RIF-CD nCmP, RIF NP were added to the RIF-CD inclusion complex solution (82:18 by weight complex-to-NP, 1 % w/v total solids in water) and allowed to mix thoroughly prior to spray drying. Mann nCmP were formed similarly to RIF-CD nCmP by replacing the RIF-CD inclusion complex with mannitol and Pure nCmP were comprised of 100% NP in water. RIF-CD MP were comprised of 100% RIF-CD inclusion complex in water. Raw RIF was dissolved in methanol prior to spray drying for the SD MP formulation. The final solutions were fed into a Büchi B-290 spray dryer coupled with a B-295 inert loop (Büchi Labortechnik, AG, Switzerland) at the following conditions: 0.7 mm nozzle diameter, inlet temperature of 70 °C, feed rate of 1.6 mL/min, gas flow rate of 414 L/h, and aspiration rate of 28 m³/h. The spray-dried formulations were separated by a high-performance cyclone, collected, and stored in a desiccator at −20°C. Yield was determined by measuring the mass of particles collected after spray drying versus the initial mass of solids in solution prior to spray drying.

Figure 1.

Schematic representing the spray-dried powder aerosol microparticle systems. Rifampicin-loaded nanoparticles (RIF NP) were spray dried into three nanocomposite microparticle (nCmP) formulations, which were comprised of RIF NP with RIF/β-cyclodextrin inclusion complex (RIF-CD Comp), mannitol, or pure NP to formulate RIF-CD nCmP, Mann nCmP, and Pure nCmP, respectively. RIF-CD inclusion complex and raw RIF were spray dried to formulate RIF-CD MP and SD RIF, respectively, to serve as control particles.

2.6. Solubility Analysis

The solubility of RIF in water was evaluated for raw RIF, RIF-CD MP, and SD MP. Each sample was added to water until precipitation was evident, creating a saturated solution. The samples were placed on a shaker plate and allowed to shake for 24 hours at room temperature and the concentration of RIF in each sample was determined via UV-Vis spectroscopy as described previously.

2.7. Fourier Transform Infrared Spectroscopy

The infrared activity of the formulations and raw materials was investigated via Fourier transform infrared (FTIR) spectroscopy using a Shimadzu IRAffinity-1S spectrometer (Shimadzu, Japan). Samples were analyzed for a wavenumber range of 4000 to 700 cm⁻¹ and each spectrum was the average of 24 scans.

2.8. Scanning Electron Microscopy

The morphology and homogeneity of the NP and nCmP formulations were analyzed using a Zeiss Sigma VP Field Emission-Scanning Electron Microscope (SEM, Germany). The samples were sputter coated with a thin film of gold/palladium alloy prior to imaging using a BIO-RAD (Hercules, CA) system at 20 μA for 10 seconds under argon. All images were captured at 3 kV. The geometric diameter (d_g) of the formulations was determined using ImageJ software by averaging the diameter of at least 100 particles for each formulation (Abramoff et al., 2004).

2.9. Evaluation of Diameter, Polydispersity Index, and Surface Charge

The diameter, polydispersity index (PDI), and surface charge (ζ potential) of the NP before and after spray drying was evaluated using a Malvern Nano Zetasizer (Malvern Instruments, Worcestershire, UK). The samples were diluted to 0.1 mg/mL in water and measurements were taken in triplicate at room temperature using a scattering angle of 173 °C.

2.10. Drug Loading Analysis

Particles were dissolved in DMSO (1 mg/mL) and the amount of RIF in the formulations was evaluated via UV-vis spectroscopy as previously described. The encapsulation efficiency (EE) and drug loading were calculated using the following equations:

$$\mathit{\text{Encapsulation}}\mathit{\text{Efficiency}}(\mathit{\text{EE}})=\frac{\mathit{\text{actual}}\mathit{\text{mass}}\mathit{\text{of}}\mathit{\text{RIF}}\mathit{\text{in}}\mathit{\text{particles}}}{\mathit{\text{theorectical}}\mathit{\text{mass}}\mathit{\text{of}}\mathit{\text{RIF}}\mathit{\text{in}}\mathit{\text{particles}}}\times 100\%$$

$$\mathit{\text{Drug}}\mathit{\text{Loading}}=\frac{\mathit{\text{mass}}\mathit{\text{of}}\mathit{\text{RIF}}\mathit{\text{in}}\mathit{\text{particles}}}{\mathit{\text{mass}}\mathit{\text{of}}\mathit{\text{particles}}}$$

2.11. Differential Scanning Calorimetry (DSC)

The thermal phase transitions of the formulations and raw materials were analyzed via differential scanning calorimetry (DSC) using a TA Q10 DSC system (TA Instruments, New Castle, DE) connected to an RSC-90 cooling accessory. Each sample (3–5 mg) was hermetically sealed in an aluminum pan with an empty pan of similar mass used as a reference. Samples were heated from 0 to 300 °C at 10 °C/min and the heat flow for the samples was normalized by the sample mass.

2.12. Powder X-Ray Diffraction (XRD)

The crystallinity of the formulations and raw materials were evaluated using a Rigaku Multiflex X-ray diffractometer (The Midlands, TX) with a Cu Kα source set to 40 kV and 44 mA. The samples were placed in a 3 mm horizontal quartz glass holder for analysis and scans were performed from 5–60° with a step of 0.1° and a scan rate of 2 °/min.

2.13. In Vitro Drug Release

The release of RIF from the nCmP systems was evaluated using a previously described method (Wang et al., 2017). Formulations were dispersed in 1X PBS (pH 7.4) or sodium acetate buffer (pH 5.2) supplemented with 0.2% Tween^® 80 and incubated at 37 °C and 100 rpm. Ascorbic acid (200 μg/mL) was added to each solution to prevent the oxidative degradation of RIF. At predetermined time points, samples were centrifuged at 23,000 × g for 15 minutes, after which the supernatant was removed and frozen at −20°C for later analysis. The same volume of fresh media was added back to each sample to maintain sink conditions. Prior to analysis, samples were diluted with an equal volume of DMSO. Calibration curves were made with the same ratio of DMSO to buffer and samples were analyzed via UV-Vis spectroscopy.

The RIF release data was fit to several semi-empirical release kinetic models (see Table A1), including the zero order, first order, Korsmeyer-Peppas, Higuchi, and the Hixson-Crowell models to determine the release mechanisms exhibited by the formulations (Costa and Sousa Lobo, 2001; Higuchi, 1963; Koizumi et al., 1975; Korsmeyer et al., 1983; Ritger and Peppas, 1987). The adjusted coefficients of determination ($R_{\mathit{\text{adjusted}}}^2$) was calculated for the fit of each model to the data to provide a more meaningful analysis of the drug release kinetics, as seen below:

$$R_{\mathit{\text{adjusted}}}^2=1-\frac{(n-1)}{(n-p)}\left(1-R^2\right)$$

Where n is the number of release data points, p is the number of parameters in the model to be determined, and R² is the original coefficient of determination calculated by fitting the data to each semi-empirical model.

2.14. In Vitro Aerosol Dispersion Performance

A Next Generation Impactor (NGI, Copley Scientific, Nottingham, UK) equipped with a stainless-steel induction port and stainless steel gravimetric insert cups attached to a Copley HCPS vacuum pump was used to investigate the aerosol dispersion characteristics of the MP formulations. To model the air flow rate of a healthy adult, the pump flow rate was set to 60 L/min (Meenach et al., 2013). Particle bounce and re-entrapment was minimized by placing glass fiber filters (55 mm, Type A/E, Pall Life Sciences, PA) in the gravimetric cups (Edwards et al., 1998). The MP were added to hydroxypropyl methylcellulose capsules (HPMC, size 3, Quali-V, Qualicaps Inc., Whitsett, NC). A human dry powder inhaler device (HandiHaler, Boehringer Ingelhelm Pharmaceuticals, CT) attached to a rubber mouthpiece was used to deliver the MP from the capsules to the NGI. Each measurement consisted of three HPMC capsules and all experiments were run in triplicate. The NGI was run with a 10 s delay time and a 10 s running time. Filter mass was recorded before and after each run to determine particle mass deposited on each stage. For a flow rate of 60 L/min, each stage has the following effective cutoff diameter: stage 1 (8.06 μm), stage 2 (4.46 μm), stage 3 (2.82 μm), stage 4 (1.66 μm), stage 5 (0.94 μm), stage 6 (0.55 μm), and stage 7 (0.34 μm). Fine particle dose (FPD), fine particle fraction (FPF), and emitted dose (ED) were calculated using the following equations:

$$\mathit{\text{Fine}}\mathit{\text{Particle}}\mathit{\text{Dose}}(\mathit{\text{FPD}})=\mathit{\text{Mass}}\mathit{\text{Particles}}<4.4\mu m$$

$$\mathit{\text{Fine}}\mathit{\text{Particle}}\mathit{\text{Fraction}}(\mathit{\text{FPF}})=\frac{\mathit{\text{FPD}}}{\mathit{\text{Total}}\mathit{\text{Particle}}\mathit{\text{Mass}}\mathit{\text{on}}\mathit{\text{Filters}}}\times 100\%$$

$$\mathit{\text{Emitted}}\mathit{\text{Dose}}(ED)=\frac{\mathit{\text{Initial}}\mathit{\text{Mass}}\mathit{\text{in}}\mathit{\text{Capsules}}-\mathit{\text{Final}}\mathit{\text{Mass}}\mathit{\text{in}}\mathit{\text{Capsules}}}{\mathit{\text{Initial}}\mathit{\text{Mass}}\mathit{\text{in}}\mathit{\text{Capsules}}}\times 100\%$$

The experimental mass median aerodynamic diameter (MMAD) and geometric standard deviation of each formulation (GSD) were calculated using a Mathematic^® program written by Dr. Warren Finlay (Finlay, 2008, 2001).

2.15. In Vitro Cytotoxicity Analysis

The cytotoxic effect of the raw RIF and RIF-CD nCmP on A549 cells was evaluated using a cell viability resazurin assay. Cells were seeded into a 96-well plate (5,000 cells/well) and incubated overnight at 37 °C. Varying concentrations of raw RIF or RIF-CD nCmP were added to wells and incubated for 48 hours. Untreated cells were used as a negative control. After 48 hours, resazurin solution (60 μM) was added to each well and incubated for 4 hours. The resulting resorufin fluorescence intensity was analyzed at an excitation at 544 nm and emission of 590 nm using a BioTek Cytation 3 microplate reader. Relative cell viability was calculated using the following equation:

$$\mathit{\text{Relative}}\mathit{\text{Cell}}\mathit{\text{Viability}}=\frac{\mathit{\text{Sample}}\mathit{\text{Fluorescence}}\mathit{\text{Intensity}}}{\mathit{\text{Control}}\mathit{\text{Fluorescence}}\mathit{\text{Intensity}}}\times 100$$

2.16. Statistical Analysis

All measurements were performed in at least triplicate. Statistical differences for in vitro cell studies were determined using one-way or two-way ANOVA with Dunnettt’s and Sidak’s multiple comparison test (GraphPad Prism Version 9). Fit factor analysis was used to compare the drug release profiles in a pairwise fashion (Diaz et al., 2016; Moore and Flanner, 1996; Sathe et al., 1996), as described in Table A2. Student’s t-test was used to determine statistical significance for other studies. A p-value < 0.05 was considered statistically significant.

3. RESULTS AND DISCUSSION

3.1. Rifampicin-Loaded nCmP Formulations

Three distinct rifampicin-loaded nanocomposite microparticle (nCmP) systems (i.e., RIF-CD nCmP, Mann nCmP, and Pure nCmP) were developed and characterized. Each nCmP system contained RIF-loaded NP made of biodegradable Ac-Dex and the systems varied by the excipient used to create the spray-dried formulations (Figure 1). The unique RIF-CD nCmP formulation utilized the RIF-CD inclusion complex to provide a fast initial RIF release phase to aid in the therapeutic efficacy as well as serving as a bulk/binding material to create microparticles with desired particle size during the spray drying process for effective pulmonary delivery, as is the case with mannitol. Two other nCmP systems were formulated as controls, including Mann nCmP, which contained RIF NP and mannitol and Pure nCmP, which contained only spray-dried RIF NP.

Cyclodextrins (CD) have the ability to form complexes with hydrophobic molecules, which shields the molecules in their hydrophobic interior cavity (de Miranda et al., 2011), as seen in Figure A2. CD complexation can result in an increase in solubility and bioavailability of poorly water-soluble drugs (Tewes et al., 2008; Thatiparti and Von Recum, 2010; Williams et al., 2013). Since the anti-tuberculosis activity of RIF is concentration dependent, is it important to maximize drug solubility and bioavailability to enhance treatment efficacy. Ac-Dex was used as the polymer carrier for the RIF NP owing to its pH-sensitive biodegradation properties and its extensive use in drug delivery applications. Ac-Dex NP are capable of providing sustained drug release over a period of hours to days and the Ac-Dex degradation rate can be easily tuned to meet treatment requirements (Broaders et al., 2009; Jakaria et al., 2022; Shah et al., 2019b).

3.2. Phase Solubility of Rifampicin in Cyclodextrin Solutions

The interactions between RIF and two CD molecules, 2-hydroxyproypl-β-cyclodextrin (β-CD) and 2-hydroxypropyl-γ-cyclodextrin (γ-CD), were investigated to determine the optimal CD to use in the nCmP formulation. Phase solubility analysis was carried out to determine the effect each CD had on RIF solubility and to determine the type of interaction between the CD and RIF. For both types of CD, the solubility of RIF increased with increasing CD concentration (see Figure 2A), which is indicative of molecular interactions between the CD and RIF via the formation of a host-guest inclusion complex (Cid-Samamed et al., 2022; Iacovino et al., 2013; Saokham et al., 2018a). The relationship between the RIF and CD concentrations was linear, indicating an A-type phase solubility profile, according to the Higuchi and Connors classification of phase solubility diagrams (Cid-Samamed et al., 2022; Saokham et al., 2018a). A-type phase solubility is characterized by an increase in substrate solubility with increasing ligand concentration. In particular, RIF and CD exhibit an A_L-type solubility profile, where the RIF/CD inclusion complex is first order with respect to the ligand (e.g., CD) and first order or higher with respect to the substrate, RIF (Thatiparti and Von Recum, 2010; Williams et al., 2013). Most often, A_L-type phase solubility profiles result in a 1:1 ratio inclusion complex. The potential of a 1:1 molecular ratio between RIF to CD is further supported as both phase solubility diagrams have a slope less than one (Patil and Suresh, 2009; Tewes et al., 2008). The stability constant (K_1:1) for each RIF/CD complex was calculated to determine the strength of the interaction between RIF and each CD. The K_1:1 values for RIF/β-CD and RIF/γ-D were similar (45.24 and 43.08 M⁻¹, respectively), indicating the same type of interactions between RIF and each CD, which is limited to the internal cavity without involvement of the methyl or hydroxypropyl pendant groups present on the outside of the CD molecules (Tewes et al., 2008). Since the CD displayed similar complexation stability with RIF, β-CD was used for further studies as it had a slightly higher stability constant, which can be attributed to a better fit of rifampicin inside the internal cavity (Anjani et al., 2022).

Figure 2.

Analysis of rifampicin-cyclodextrin (RIF-CD) inclusion complexes comprised of hydroxypropyl-β-cyclodextrin (β-CD) and hydroxypropyl-γ-cyclodextrin (γ-CD), including (A) phase solubility of RIF in CD solutions with the linear relationship between RIF and CD concentrations and the resulting coefficient of determination (R²); (B) solubility of raw RIF, spray-dried RIF (SD RIF), and spray-dried RIF-CD complex (RIF-CD MP) in water; (C) Fourier transform infrared (FTIR) spectroscopy of RIF-CD complexes, raw β-CD, RIF-CD MP, raw RIF, and SD RIF; and (D) Zoomed-in spectra of the FTIR data. Data represents the mean ± standard deviation (n = 6), ***p < 0.0001 and ****p < 0.00001.

3.3. Rifampicin-Cyclodextrin Complexation

The interaction between RIF and β-CD during complexation was further investigated using FTIR spectroscopy. As seen in Figure 2C/D, raw RIF displayed several characteristics peaks, including peaks at 2970 cm⁻¹ and 2940 cm⁻¹ due to the stretching vibration of H-C=O and C-H, a peak at 1720 cm⁻¹ from the stretching vibration of C=O, and a peak 1648 cm⁻¹ from the C=N stretching vibration of the piperazine ring (Anjani et al., 2022; He et al., 2013). In comparison to raw β-CD, RIF-CD MP displayed a broadening and decrease in peak intensity in the range of 3600 – 3000 cm⁻¹, which is associated with O-H stretching and indicates that RIF is interacting with - OH groups in the internal cavity of the CD. Furthermore, after complexation, the sharp peak observed at 1720 cm⁻¹ was no longer present for the RIF-CD MP, indicating that the piperazine ring is inside the CD cavity, “shielding” it from the spectrum (Anjani et al., 2022; He et al., 2013; Saokham et al., 2018a). A small, broad peak was observed around 1648 cm⁻¹ for both RIF-CD and raw β-CD, which is associated with β-CD and not the piperazine ring in RIF. The full spectra for all compounds evaluated can be found in Figure A3.

SD RIF was analyzed to investigate molecular changes due to the spray drying process, and the spectra for raw RIF and SD RIF were identical and the characteristic peak associated with the RIF piperazine ring was still present at 1648 cm⁻¹, indicating that spray drying does not affect RIF at a molecular level. Therefore, the disappearance of the peak in the RIF-CD MP spectrum results from an interaction between the piperazine and CD. These results further confirm that RIF and β-CD successfully formed an inclusion complex and that the complexation involves the piperazine ring of RIF.

3.4. Rifampicin Solubility

The complexation of CD and compounds has been extensively used to improve the solubility and bioavailability of drugs (Saokham et al., 2018b). To investigate this effect, the solubility of RIF from raw RIF, spray-dried RIF (SD RIF), and spray-dried RIF-CD (RIF-CD MP) was determined, as seen in Figure 2B. Spray drying has been shown to increase solubility of drugs due to their transition from the crystalline to amorphous forms, placing them in a high energy state that more readily solubilized (Elsayed and Aboughaly, 2016; Shoyele and Cawthorne, 2006; Vehring, 2008). SD RIF displayed a significant increase in RIF solubility compared to raw RIF (p < 0.0001) and RIF-CD MP resulted in significantly higher RIF solubility in comparison to both raw RIF and SD RIF (p < 0.00001). These results confirm that β-CD effectively increased RIF solubility by complexation and remained complexed after spray drying.

3.5. Evaluation of Drug Loading, Particle Size, Homogeneity, and Surface Charge

As seen in Table A3, the diameter of the parent RIF NP is appropriate for drug delivery applications (< 200 nm) and the low polydispersity index (PDI) indicates that the NP are monodispersed. The NP surface charge was nearly neutral, which is consistent with other PVA-coated NP systems (Shah et al., 2019a; Wang et al., 2017). While a large zeta potential is often desirable to prevent particle aggregation, PVA provides colloidal stability by steric repulsion in addition to minimizing non-specific binding to cells (Gutiérrez et al., 2019; Mngadi et al., 2021), which explains the favorable PDI value. Furthermore, SEM confirmed the parent RIF NP were spherical and monodispersed in nature (Figure 3). RIF was successfully encapsulated into the RIF NP (Table A3) with an encapsulation efficiency of 7.5% due to its slight solubility in water.

Figure 3.

Representative scanning electron microscopy images of (A) parent RIF NP prior to spray drying (scale bar = 2 μm) and (B) spray-dried RIF-CD MP (scale bar = 1 μm); spray-dried dry powder aerosol formulations, including (C) RIF-CD nCmP, (D) Mann nCmP, and (E) Pure nCmP (scale bar = 1 μm); and redispersed RIF NP of the nCmP formulations in water corresponding to (F) RIF-CD nCmP, (G) Mann nCmP, and (H) Pure nCmP (scale bar = 2.5 μm). RIF = rifampicin, CD = cyclodextrin, nCmP = nanocomposite microparticles, Mann = mannitol, RIF-CD = RIF-cyclodextrin complex, MP = microparticles, and NP = nanoparticles.

As seen in Table 1, the diameter, PDI, and surface charge of the RIF NP redispersed from the dissolved nCmP were similar to the parent RIF NP, indicating that spray drying did not have a significant impact on the NP properties. The RIF loading in RIF-CD nCmP was 8- and 10-fold higher than the loading for Mann nCmP and Pure nCmP, respectively, which shows that using RIF-CD complex as a bulk/binding material during spray drying can not only drastically increase RIF loading but also achieve similar particle characteristics compared to similar nCmP systems.

Table 1.

Properties of rifampicin-loaded nanoparticles (RIF NP) after redispersion from RIF-CD nCmP, Mann nCmP, and Pure nCmP, including the diameter, polydispersity index (PDI), zeta (ζ) potential, and RIF loading and the aerosol dispersion properties of the spray-dried (SD) formulations, including geometric diameter (d_g), mass median aerodynamic diameter (MMAD), geometric standard diameter (GSD), and spray drying yield. RIF = rifampicin, CD = cyclodextrin, nCmP = nanocomposite microparticles, Mann = mannitol, RIF-CD = RIF-CD complex, MP = microparticles, NP = nanoparticles. Data represents the mean ± standard deviation (n = 3).

	Parameter	RIF-CD nCmP	Mann nCmP	Pure nCmP	RIF-CD MP
Particle Properties	Redispersed NP Diameter (nm)	164 ± 2	171 ± 1	172 ± 2	---
	Redispersed NP PDI	0.099 ± 0.014	0.134 ± 0.059	0.092 ± 0.012	---
	Redispersed NP ζ Potential (mV)	−6.7 ± 1.0	−9.3 ± 1.6	−11.4 ± 0.83	---
	RIF Loading (μg RIF/mg MP)	55.1 ± 1.2	6.8 ± 0.2	11.5 ± 0.5	122 ± 4.3
Aerosol Properties	(nm)	1.8 ± 0.4	1.7 ± 0.5	2.9 ± 0.9	1.9 ± 0.8
	MMAD (μm)	4.0 ± 0.0	6.7 ± 0.5	5.0 ± 0.3	4.6 ± 0.1
	GSD (μm)	1.6 ± 0.0	1.7 ± 0.1	1.6 ± 0.1	1.8 ± 0.0
	Yield (%)	59.1	46.8	61.2	60.6

3.6. Microparticle Morphology and Size

The morphology and geometric diameters (d_g) of the spray-dried microparticles was investigated via SEM and ImageJ analysis. The d_g of the formulations ranged from 1.7 to 2.9 μm and the particles exhibited raisin-like wrinkled morphology, respectively (see Table 1 and Figure 3). The morphology of the particles can be attributed to the spray drying parameters used in this study, likely resulting in a large Péclet (Pe) number, which is a dimensionless parameter that relates the solvent evaporation rate to the diffusion rate of the solutes (Vehring, 2008; Vehring et al., 2007). When the Pe number is less than one, the solvent evaporation rate and diffusion rate of the solutes are similar and the solute diffuses towards the center of the receding droplet and relatively dense, spherical particles are formed. Conversely, a Pe number greater than one means the solvent evaporates faster than the solute diffusion rate. This parameter is even more prominent when NP are present in a drying droplet, as they are relatively immobile compared to solute molecules in the solution (Arpagaus, 2018; Malamatari et al., 2020).

Evaporation of the droplet during spray drying results in the formation of a shell as the solutes become concentrated at the liquid-air interface and additional drying after shell formation causes an increase in vapor pressure within the particle and, eventually, the shell collapses and wrinkles as the remaining solvent is removed (Vehring, 2008). Mann nCmP exhibited a slightly wrinkled morphology but were the most spherical particles, which can be attributed to using mannitol as the excipient, as mannitol helps to stabilize the shell during droplet drying, resulting in less shell collapse following solvent evaporation (Arpagaus, 2018; Mangal et al., 2017). Studies have shown that wrinkled particles such as the described nCmP formulations can improve aerosol dispersion characteristics and reduce particle-particle interactions (O’Hara and Hickey, 2000; Vehring et al., 2007).

3.7. Thermal Phase Transitions

As seen in Figures 4 and A4, raw mannitol and PVA exhibited endothermic peaks, whereas raw RIF exhibited an exothermic peak, which are indicative of a crystalline melting point for each compound (Drzeżdżon et al., 2019; PerkinElmer and Inc, 2013), and the resulting melting points were 160, 240, and 265 °C for raw mannitol, PVA, and RIF, respectively. Spray drying has been shown to cause crystalline materials to transition to their amorphous state, which can be seen for the spray-dried materials. The characteristic melting peaks associated with raw mannitol and RIF are not present after spray drying, indicating they transitioned to their amorphous states. Importantly, RIF remained in an amorphous state in all nCmP systems, as indicated by a lack of sharp peaks in the thermograms. Amorphous forms of drugs have higher solubility and better dissolution compared to their crystalline forms, which can increase their bioavailability (Hancock and Parks, 2000; Williams et al., 2013).

Figure 4.

Solid-state characteristics of the spray-dried dry powder aerosol formulations RIF-CD nCmP, Mann nCmP, and Pure nCmP and their raw components, including (A) differential scanning calorimetry (DSC) thermographs and (B) X-ray diffraction (XRD) diffractograms. RIF = rifampicin, SD = spray-dried, Mann = mannitol, RIF-CD = RIF-CD complex, NP = nanoparticles, nCmP = nanocomposite microparticles.

3.8. Formulation Crystallinity

X-ray diffraction was used to further elucidate the crystallinity and microstructure of the raw materials and formulations, as seen in Figure 4. Raw RIF displayed sharp diffraction peaks from 5 to 30°, which is consistent with previously reported results (Anjani et al., 2022; Guirguis and Moselhey, 2012), and raw mannitol displayed sharp peaks from 10 to 45°. The sharp diffraction peaks and consistent diffraction angles are a result of the long-range ordered crystalline structure of RIF and mannitol (Chauhan, 2014). Since amorphous materials lack such long-range order, they produce broad diffraction patterns without sharp peaks, as observed for raw Ac-Dex and SD RIF (see Figure A5), indicating they are amorphous in nature. Raw β-CD displayed a halo type diffraction pattern, which is indicative of an amorphous material (Han et al., 2020; Subramani, 2016), whereas SD RIF did not display any peaks, further confirming it remains amorphous after spray drying. Interestingly, the spray-dried nCmP and parent RIF NP displayed four distinct peaks at 28, 32, 46, and 56°, which were not characteristic peaks of raw RIF, mannitol, or Ac-Dex, indicating these materials remained amorphous after spray drying. Instead, these peaks are from residual PBS used during the NP synthesis, as seen by the PBS diffractogram. Overall, DSC and XRD analysis confirm that spray drying resulted in the successful transition of RIF to an amorphous state and that it remained as such in the nCmP systems, and this has important implications by increasing RIF solubility and bioavailability (Alves et al., 2010; Tewes et al., 2008).

3.9. Aerosol Dispersion of Dry Powder Microparticle Formulations

The effectiveness of using inhalable dry powders in pulmonary drug delivery applications is largely dependent on their aerosol deposition in the lungs. As seen in Figure 5, deposition occurred for the nCmP systems and RIF-CD MP on stages 1–6 of the Next Generation Impactor, indicating that they are capable of effective aerosol pulmonary delivery. As seen in Table 1, the mass median aerodynamic diameters (MMAD) for the nCmP formulations ranged from 4.0–6.7 μm. MMAD is an important value that describes how particles behave in air flow and determines where particles will deposit in the lungs. To effectively treat diseases such as tuberculosis via pulmonary delivery, particles must be capable of depositing in the alveolar region of the lungs following aerosolization. Dry power particles with an MMAD around 1–5 μm are able to effectively deposit in the alveolar region, while particles up to 10 μm deposit in the conducting airways, and particles smaller than 1 μm will most likely be exhaled (He et al., 2022; Hickey, 2020; Labiris and Dolovich, 2003; Shoyele and Cawthorne, 2006). The MMAD of the RIF-CD nCmP formulation (4 μm) indicates that these particles are capable of deposition in the alveolar region of the lungs. Mann nCmP displayed the largest MMAD (6.7 μm) of the three nCmP systems with similar particle diameters (Table 1). As seen in Figure 3, Mann nCmP were the most spherical of the systems and this morphology may potentially increase particle-particle interactions, adversely affecting its aerosol dispersion characteristics, which was consistent with a previous report (Wang and Meenach, 2016).

Figure 5.

In vitro aerosol dispersion performance of the spray-dried aerosol formulations, including (A) particle deposition represented as the percentage (%) of particles deposited on each stage of a Next Generation Impactor. The effective cutoff diameters (D₅₀) for each stage are as follows for an air flow rate of 60 L/min: Stage 1 = 8.06 μm, Stage 2 = 4.46 μm, Stage 3 = 2.82 μm, Stage 4 = 1.66 μm, Stage 5 = 0.94 μm, Stage 6 = 0.55 μm, and Stage 7 = 0.34 μm. (B) Fine particle fraction (FPF), respirable fraction (RF), and emitted dose (ED) of each particle system. Data represents the mean ± standard deviation (n = 3).

Fine particle fraction (FPF), respirable fraction (RF), and emitted dose (ED) are also important aerosol dispersion characteristics for inhalable dry powder therapeutics. It can be seen in Figure 5 that the FPF for the RIF-CD nCmP was the highest, indicating that a large amount of the inhaled particles will be able to reach the alveolar region of the lungs. RIF-CD nCmP also displayed high RF and ED values. The likelihood of increased particle-particle interactions for the Mann nCmP formulation is evident by its FPF, as it is the lowest of the particles evaluated in this study. The spry-dried microparticle formulations displayed a GSD of 1.6 to 1.8 μm, which is an indication of the heterogeneity of the particles, as confirmed in the SEM images (Labiris and Dolovich, 2003).

The yield for RIF-CD nCmP, Pure nCmP, and RIF-CD MP was relatively high (~60%) for what is typically seen for a mini spray dryer and is consistent with previously reported results for nCmP (Sollohub and Cal, 2010; Zimeng Wang and Meenach, 2016; Wolska, 2021). Mannitol nCmP resulted in the lowest yield (47%) due to increased accumulation on spray drying glassware, as mannitol has a glass transition temperature of 13 °C, well below the spray drying temperatures, which causes increased “stickiness” (Chan et al., 2004; Costantino et al., 1998; Lechanteur and Evrard, 2020). RIF-CD nCmP displayed appropriate aerosol dispersion characteristics with a high FPF, RF, and ED, indicating its potential to effectively deliver a high concentration of RIF at the site of infection, the alveolar region of the lungs.

3.10. In Vitro Release Characteristics of Rifampicin from Particle Formulations

The in vitro release of RIF from the nCmP formulations was carried out in two physiologically relevant media supplemented with Tween 80 to aid RIF solubility and ascorbic acid to prevent oxidative degradation of RIF. As seen in Figure 6, at pH 7.4, the cumulative release of RIF from RIF-CD nCmP after 6 days was 83.4%, which is significantly higher than the release for Pure nCmP and Mann nCmP at 22.7 and 25.9%, respectively (see Table A4 for statistical analysis). In comparison, at pH 5.2, the release of RIF after 6 days was the same for the nCmP formulations, with RIF-CD nCmP, Mann nCmP, and Pure nCmP releasing 98.9, 95.6, and 88.3% of RIF, respectively. At both pH conditions, RIF-CD nCmP exhibited a large initial burst release of RIF, where after 8 hours, significantly more RIF was released in comparison to the other systems, indicating the impact of using water-soluble RIF-CD as the excipient in the formulation. The impact of pH on RIF release is clear when looking at the amount released after 6 days, as there was a significant difference in RIF release for all nCmP formulations at pH 7.4 versus 5.2 (see Table A4) due to the pH-sensitive nature of Ac-Dex (Bachelder et al., 2017; Broaders et al., 2009; Shah et al., 2019b), resulting in faster degradation (and subsequent RIF release) in acidic conditions. Interestingly, the release of RIF after 8 hours is the same for RIF-CD nCmP at both pH conditions, indicating that this formulation exhibited burst release of RIF regardless of pH.

Figure 6.

In vitro release profiles of rifampicin from the spray-dried aerosol particle formulations at (A) pH 7.4 and (B) pH 5.2. RIF = rifampicin, CD = cyclodextrin, RIF-CD = RIF-CD complex, nCmP = nanocomposite microparticles, and Mann = mannitol. Data represents mean ± standard deviation (n = 3).

The RIF release profiles from the formulations were evaluated to determine their “sameness” using fit factor analysis (Diaz et al., 2016; Moore and Flanner, 1996; Sathe et al., 1996), as seen in Table A5. Given the significant difference in the formulations and pH testing conditions, none of the profiles in their entirety exhibited any similarity with respect to this analysis. As seen in Table A6, the release data was also fitted to several common semi-empirical models (see Table A1) to determine the type of release exhibited by each system. All of the release profile fit either the Korsmeyer-Peppas or Higuchi models the best, both of which model drug release from polymeric matrix formulations (Higuchi, 1963; Korsmeyer et al., 1983), which fit the given formulations well.

The unique combination of RIF-CD complex and Ac-Dex NP was responsible for the dual release profiles with a burst release of RIF from the RIF-CD complex followed by controlled release of RIF from the RIF NP (Anjani et al., 2022; Cid-Samamed et al., 2022; Doile et al., 2008; Tewes et al., 2008). While there are several mechanisms for the release of drug from CD inclusion complexes, the main driving force is dilution (Stella et al., 1999). Therefore, RIF can rapidly disassociate from the CD once introduced to the release media, resulting in the burst release observed. The burst release of RIF from β-CD would result in rapid delivery of the drug in the alveolar region, while the RIF NP can provide controlled release after being redispersed from the nCmP. Achieving a high concentration of RIF is important given the concentration dependency of RIF in anti-TB applications (Pang et al., 2013). The slow degradation rate of Ac-Dex at neutral pH provides sustained release of the RIF from the NP, allowing for longer residence time in the lungs (Sung et al., 2007), which can potentially increase alveolar macrophage clearance, where at phagosomal pH (pH 5.2) RIF will rapidly be released from the RIF NP. Overall, the RIF-CD nCmP formulation is capable of delivering NP to the alveolar region of the lungs, while providing both sustained and pH-sensitive release in addition to an initial burst release of RIF at the target site.

3.11. Cytotoxic Impact of Microparticles on Lung Epithelia

The in vitro cytotoxicity of RIF-CD nCmP and raw RIF was investigated by dosing A549 pulmonary epithelial cells with RIF or RIF-CD nCmP for 48 hours. The concentration range evaluated was based on where the raw RIF and RIF-CD nCmP formulation had an impact on the viability of the A549 cells. It is challenging to equate these concentrations to therapeutic levels of RIF as both the typical RIF dosage (600 mg/day for the treatment of TB) or typical plasma concentrations of RIF are not relevant to a formulation delivered as an aerosol. To date, there is little (if any) information on the dosage needed to treat TB using an aerosol formulation. As seen in Figure 7, the nCmP and raw RIF imparted no adverse effects on the cells with respect to relative viability up through 1 μM RIF dosing, however, the relative viability of the A549 cells significantly decreased upon exposure to RIF-CD nCmP at 10 and 25 μM RIF and for raw RIF at 25 μM. This decrease in cell viability can potentially be attributed to the presence of CD in the RIF-CD nCmP formulation, as they can interact with cellular membrane components to increase membrane fluidity and permeability, enhancing particle absorption into cells compared to free drug (Fenyvesi et al., 2014; Hammoud et al., 2019; Patil and Suresh, 2009; Tewes et al., 2008). Enhanced uptake of RIF-NP released from the RIF-CD nCmP into cells could potentially mean a lower dose is required. Overall, exposure of RIF-CD nCmP on A549 cells resulted in a decrease in cell viability at high RIF concentrations and does not display cytotoxicity at lower RIF concentrations, indicating a concentration-dependent effect on cytotoxicity.

Figure 7.

Cytotoxicity (relative viability) of A549 lung adenocarcinoma cells exposed to RIF-CD nCmP and raw RIF at varying RIF concentrations for 48 hours in comparison to untreated cells (control). RIF-CD nCmP = rifampicin-cyclodextrin nanocomposite microparticles, RIF = rifampicin. Data represents mean ± standard deviation (n = 3), **p < 0.001, ***p < 0.001.

4. CONCLUSION

Tuberculosis is a leading cause of infection worldwide and results in millions of deaths annually. While RIF is effective in the treatment of tuberculosis, its efficacy is hindered greatly by the length of treatment and dosing frequency, leading to decreased patient compliance. This has a two-fold effect – increased mortality rates and an increase in multi-drug resistant tuberculosis. The aim of this study was to develop and characterize a unique RIF nCmP formulation comprised of RIF-CD and Ac-Dex NP. Rifampicin and cyclodextrin were confirmed to form inclusion complexes and effectively enhanced RIF solubility following the spray drying of RIF-CD inclusion complexes. The spray-dried RIF-CD nCmP formulation displayed favorable aerosol dispersion characteristics for pulmonary deposition, with an MMAD well within the 1–5 μm range required for alveolar deposition and a high fine particle fraction. This indicates a high percentage of the respirable dose will reach the alveolar region of the lungs where RIF NP can be effectively redispersed from the nCmP. Notably, the RIF-CD nCmP release profile displayed a burst release from the RIF-CD complex while the RIF NP provided sustained and pH-sensitive release, important for the concentration dependent efficacy of RIF. The formulation was not toxic to lung epithelial cells at low concentrations indicating its safe use. Overall, the developed RIF-CD nCmP has the potential to improve upon current tuberculosis treatment options and can be easily modified to adjust particle characteristics to meet a variety of treatment needs.

Abbreviations:

Ac-Dex

acetalated dextran

CD

cyclodextrin

DMEM

Dulbecco’s Modified Eagle’s medium

DMSO

dimethyl sulfoxide

DSC

differential scanning calorimetry

ED

emitted dose

EE

encapsulation efficiency

FPD

fine particle dose

FPF

fine particle fraction

FTIR

Fourier transform infrared

GSD

geometric standard deviation

Mann

mannitol

2-MOP

2-methoxypropene

MMAD

mass median aerodynamic diameter

MP

microparticles

nCmP

nanocomposite microparticles

NGI

Next Generation Impactor

NP

nanoparticles

PDI

polydispersity index

PBS

phosphate buffered saline

PPTS

pyridium p-toluenesulfonate

RD

respirable dose

RIF

rifampicin

SEM

scanning electron microscopy

TB

tuberculosis

TEA

triethylamine

XRD

x-ray diffraction