Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Aug 3.
Published in final edited form as: Int J Pharm. 2016 Apr 26;514(2):384–391. doi: 10.1016/j.ijpharm.2016.04.063

A dry powder combination of pyrazinoic acid and its n-propyl ester for aerosol administration to animals

PG Durham a, EF Young b, MS Braunstein b, JT Welch c, AJ Hickey a,*
PMCID: PMC9347698  NIHMSID: NIHMS1814821  PMID: 27130363

Abstract

Combining the advantage of higher efficacy due to local pulmonary administration of pyrazinoic acid (POA) and potent effect of pyrazinoic acid ester (PAE) delivered as an aerosol would aid in tuberculosis therapy. A combination spray dried dry powder, composed of POA, PAE (n-propyl POA), maltodextrin and leucine, was prepared for aerosol delivery to animals. Solid-state characteristics of morphology (scanning electron microscopy) crystallinity (X-ray powder diffraction), thermal properties (thermogravimetric analysis and differential scanning calorimetry) and moisture content (Karl Fisher) were evaluated. Particle size distributions, by volume (laser diffraction) for the dispersed powder and by mass (inertial impaction) were determined. Efficient delivery of the powder to a nose only animal exposure chamber employed a novel rotating brush/micro-fan apparatus. Spherical, crystalline particles were prepared. The volume median diameter, ~1.5 μm, was smaller than the mass median aerodynamic diameter, ~3.0 μm, indicating modest aggregation. Drug content variations were observed across the particle size distribution and may be explained by PAE evaporative losses. Delivery to the nose-only exposure chamber indicated that boluses could be administered at approximately 3 min intervals to avoid aerosol accumulation and effect uniform dose delivery with successive doses suitable for future pharmacokinetic and pharmacodynamic studies.

Keywords: Aerosol, Lungs, Tuberculosis, Pyrazinoic acid, n-Propyl pyrazinoic acid, Spray drying, Powder dispersion apparatus, Nose-only exposure chamber

1. Introduction

1.1. Tuberculosis

Tuberculosis continues to be one of the World’s most serious infectious diseases, accounting for 1.5 million deaths in 2014 alone. (World Health Organization, 2015) Worldwide, the incidence of multi-drug and extensively drug resistant (MDR and XDR) disease poses a serious challenge to global health (Acosta et al., 2014; Cohen et al., 2015; Smaoui Fourati et al., 2015).

1.2. First line drugs

Mycobacterium tuberculosis (MTB) originated from a soil microbe, which evolved to withstand exposure to ambient environment, having a thick mycolic acid coat (Draper, 1998). Its ability to survive outside the body for a period sufficient to support the pulmonary route of transmission and slow rate of growth make intervention to prevent or treat emergence of disease in the host challenging.

The arsenal of efficacious compounds at the disposal of doctors and pharmacists has not changed significantly over the decades with only one new drug, bedaquiline, being approved recently. The current standard of care begins with two months of first line agents – isoniazid (INH), rifampicin (RIF), pyrazinamide (PZA), and ethambutol (E). However, this is insufficient to adequately treat multiple (MDR-TB) and extensively (XDR-TB) tuberculosis patients and much longer periods of treatment with many more drugs are required (Karo et al., 2015; Yuen et al., 2015).

1.3. Pyrazinamide

PZA is included in initial treatment with other first-line agents (INH, RIF and E) and allows a reduction in the duration of therapy. The mechanism of therapeutic action of PZA has yet to be fully elucidated, and a current theory outlined by Zhang and Mitchison, briefly proposes PZA arrives proximal to the micro-organism where it traffics into the MTB. It is then converted to pyrazinoic acid (POA) by a mycobacterial enzyme, pyrazinamidase. POA is exported from the pathogen, but intracellular macrophage concentration of POA and decreased intracellular macrophage pH at the site of the pathogen, may lead to bioaccumulation and ultimately to the death of the micro-organism (Zhang and Mitchison, 2003).

Given that the efficacy of administering PZA largely relies on a functional microbial enzyme to activate POA, a major cause of PZA resistance is a mutation in the gene coding pyrazinamidase. It has also been observed that PZA resistant strains show lack of uptake via an ATP-dependent transport system (Raynaud et al., 1999).

PZA is administered orally, in the range of 20–30 mg/kg daily. For a 75 kg adult, a 25 mg/kg dose amounts to almost 2 g (1.875 g). Duration of treatment of PZA is limited by hepatotoxicity. Though POA is known as the active moiety, administering POA directly via the oral route has been shown to be less successful in eradicating MTB. In culture, conflicting results have been observed regarding the efficacy of POA in comparison to PZA. Recent work has shown that wild-type MTB is more susceptible to the action of POA than PZA at neutral and alkaline media conditions (Peterson et al., 2015).

1.4. Pyrazinoic acid esters

The reliance on pyrazinamidase for activation of PZA has been noted (Zhang et al., 2014), and other prodrugs have been explored. Many pyrazinoic acid esters (PAEs) have been synthesized and evaluated in vitro for efficacy (Cynamon et al., 1992). Given the abundance of nonspecific esterases ubiquitous in the body, activation to POA should occur regardless of pyrazinamidase activity, mitigating that mechanism of resistance. In the catalog of ester variants explored, one of the more efficacious (in terms of MIC) was the n-propyl ester of POA, achieving a 2-log reduction in MTB (H37Rv) in vitro at pH 5.8 at a concentration ≤ 3.12 μg/mL (Cynamon et al., 1992). Previously, the MIC of POA alone was shown to be 240–480 μg/mL at pH 5.6. (Heifets et al., 1989). The authors found this dose to be too high to be achievable in humans, they do note POA’s ability to lower the pH sufficiently in vitro, and it is known that PZA is more effective at lower pH. (McDermott and Tompsett, 1954).

1.5. In vivo efficacy

PZA is an important drug in the first line treatment of tuberculosis. PZA is a prodrug of the active moiety, POA. POA, delivered to the gastro-intestinal tract experiences barriers to bioavailability that reduce its efficacy. Consequently, it has not been developed for oral delivery. PAEs have significantly lower minimum inhibitory concentrations for MTB than POA but are liquid at room temperature. Combining the advantage of higher efficacy due to local pulmonary administration of POA and potent effect of PAE delivered as an aerosol would aid in tuberculosis therapy. A strategy for delivery as a dry powder aerosol would render PAE more suitable for use in regions of the world that do not have ready access to electricity and compressed gas.

1.6. The pulmonary route of administration

Administering drugs by inhalation has a long history. The Egyptians recognized the power of inhaled vapors, and cigarette smoke is known to efficiently deliver nicotine by the pulmonary route to the systemic circulation (Patton and Byron, 2007). However, the inhaled route has limited or no benefit over the oral route for most pharmaceutical applications. Most frequently, drugs are administered to the lungs to treat lung or airway diseases and infections by direct aerosol delivery.

Scientific progress in aerosol delivery systems has advanced to a point that targeting drugs to the lungs is readily achievable with increasing reproducibility and specificity (Hoppentocht et al., 2014). Nebulizers have been widely used in clinical settings and are a useful tool for a variety of applications, but the logistics of nebulizer use make it less practical for patient use under ambulatory conditions (Roche and Huchon, 2000). It is worth noting that the major burden of tuberculosis is in the developing world, where resources such as consistent supply of electricity, cold storage and compressed gas may be a challenge (World Health Organization, 2015).

Recent scientific and technical discoveries, evolving understanding of airway anatomy and physiology, and improvements in particle engineering capabilities have paved the way for dry powder formulations to be delivered as aerosols from specialized inhalers. They afford ease of use and high dose delivery required for treatment of local microbial infection (Hickey et al., 2013).

Although tuberculosis is a systemic infection, the lungs present as the major symptomatic region and the site of transmission (Smith, 2003). To achieve high lung concentrations sufficient to exceed the MIC, as required for therapy via the oral route, the systemic drug load must also be very high. This increases off-target effects and the prospects of toxicity, most notably in the liver (Durand et al.,1996). A local application of drugs via the pulmonary route can result in high lung concentrations and reduced side effects, while circumventing the first-pass metabolism experienced by orally administered drugs. The use of aerosols to treat TB is not a novel concept but has received considerable attention in recent years (Hickey et al., 2015).

n-Propyl pyrazinoate has been evaluated for the treatment of MTB (strain H37Rv) in vivo in infected guinea pigs when administered as a nebulized aerosol. Aerosol PAE therapy reduced the bacterial burden significantly not only in the lungs, but also in the spleen and thoracic lymph nodes (Young et al., 2015). Additionally, co-delivery of POA locally via the airways may increase the efficacy of PAE by lowering pH and could have additional local therapeutic effect, as it is accepted as the active moiety of PZA and the hydrolysis product of PAE.

1.7. Rationale for dry powder

n-Propyl pyrazinoate is a hydrophobic liquid at room temperature. Previous work with liquid ester formulations for nebulization were limited in the scope of formulation requirements as it was sufficient to prepare a solution for delivery from a vibrating mesh nebulizer for these initial experiments. However, having demonstrated efficacy with this approach it may be limited in its application worldwide by the absence of sufficient infrastructure as mentioned earlier (i.e. electricity, compressed gas) and convenience. A dry powder inhaler (DPI) product would not require a power source, would be stable at room temperature and convenient for both the patient and the health care provider.

Maltodextrin has been widely utilized to create flowable powders containing material that is otherwise liquid at room temperature, such as ethanol (Mitchell and Seidel, 1974) and hydrophobic molecules such as vegetable oil (Fuchs et al., 2006). Maltodextrins, as dextrose polymers, are generally categorized by the dextrose equivalent (DE) rating. Maltodextrins with a lower DE rating exhibit higher degrees of branching and, with increased surface area, have a greater absorptive potential. For this reason, a low DE maltodextrin was selected for spray drying to increase the encapsulation efficiency of the PAE. Maltodextrins are hygroscopic (U.S. Pharmacopeial Convention, 2011). However, lower DE maltodextrins are less hygroscopic (Goula and Adamopoulos, 2008). l-leucine was selected as an excipient due to its anti-hygroscopic effect, notable ability to improve powder flow and reduction in caking (Chang et al., 2014). Additionally, leucine is incorporated into approved inhaled products to enhance delivery and is considered safe for inhalation (Cipolla et al., 2014).

2. Materials and methods

2.1. Materials

n-Propyl pyrazinoate (PAE) was synthesized for these experiments using a method previously employed by one of the authors (Cynamon et al., 1992) and confirmed by proton NMR. Maltodextrin, POA and reagent grade ethanol were obtained from Sigma-Aldrich (St. Louis, MO, USA). L-leucine was obtained from Alfa Aesar (Ward Hill, MA, USA).

2.2. Methods

2.2.1. Spray drying

Particles were spray dried (Model B-290, Buchi, Flawil, Switzerland). 500 mg each of PAE, POA, maltodextrin and l-leucine were combined and dissolved in 200 mL of 20% ethanol for a total solids concentration of 10 mg/mL. Liquid feed was atomized at a rate of 5 mL/min using a two-fluid nozzle of inner and outer orifice diameters 0.7 and 1.5 mm, respectively. Nitrogen was used as the atomizing gas at a flow rate of 1052 L/h, as estimated by the calibrated rotameter adjusted for pressure drop, supplied with the apparatus by the manufacturer. Room air was used as the drying gas, pulled through the spray dryer at a rate of 35 m3/h. Inlet temperature was 85 °C, resulting in an outlet temperature of 45 °C. Particles were collected from the airstream using the high efficiency cyclone.

2.2.2. Physicochemical characterization

2.2.2.1. Scanning electron microscopy (SEM).

Scanning electron micrographs (Quanta 200, FEI, Hillsboro, OR) were obtained after sampling spray dried powder onto carbon adhesive mounted on an aluminum stub. Samples were sputter coated with gold/palladium (Hummer Sputtering System, Anatech Ltd., Union City, CA) for a period of 120 s. Particles were imaged at various magnifications with an accelerating voltage of 15 kV and a spot size of 3.0.

2.2.2.2. Thermal analysis.

The melting point of PAE was determined by differential scanning calorimeter (DSC) (Q200, TA Instruments, New Castle, Delaware, USA). Sample was loaded into an aluminum pan and weighed before analysis. The chamber was cooled at a rate of 10 °C/min to −90 °C, then heated at 10 °C/min to 30 °C. Spray dried powder and components (POA, maltodextrin and leucine, 1–5 mg per sample) were analyzed by DSC in capped aluminum pans from 30 °C to 300 °C at a ramp rate of 10 °C/min.

Thermogravimetric analysis was performed (TA Instruments, New Castle, Delaware, USA) using approximately 10 mg of powder weighed into a tarred platinum sample pan. Sample was heated at 5 °C/min from 20 °C to 300 °C under nitrogen purge.

2.2.2.3. X-ray powder diffraction (XRPD).

The XRPD (XRD-6100 diffractometer, Shimadzu, Japan) pattern of spray dried powder was determined from 8 to 70 degrees two-theta with a step rate and dwell time of 0.04 ° two-theta for 4 s using a low-background glass sample holder with a copper X-ray source with a wavelength of 1.5418 nm. The diffraction pattern was processed using Jade version 9.6 pattern processing software (Materials Data Inc., Livermore, California, USA). PAE powder, POA, maltodextrin and leucine were examined by XRPD

2.2.2.4. Moisture content.

Moisture content of the PAE powder was determined by Karl Fischer titration using a V30Compact Volumetric Karl Fischer Titrator (Mettler Toledo, Columbus, Ohio USA) using 10–13 mg per replicate (n = 3).

2.2.3. Aerosol particle size determination

2.2.3.1. Inertial impaction.

Aerodynamic size was determined by inertial impaction (Next Generation Impactor, NGI, MSI Corp, USA) operated at 60 L/min for 4 s per actuation. A mass of 10 mg of spray dried powder was loaded into each of three, size #3 hydroxypropylmethylcellulose (HPMC) capsules and delivered to the impactor by a Cyclohaler (Plastiape, Italy) dry powder inhaler. The stages were coated with silicone oil to prevent particle bounce and re-entrainment by applying a solution of 1% w/v silicone oil in hexane, removing excess, and allowing hexane to evaporate. Impaction runs were performed in triplicate.

Deposited powder on each stage was dissolved in deionized water and quantified by liquid chromatography (Acquity UPLC, Waters, Milford, Massachusetts, USA) using a HSST3 1.8 μm, 2.1 × 50 mm column. Mobile phases A and B were water with 0.05% (v/v) TFA and acetonitrile with 0.04% (v/v) TFA, respectively. The gradient was 0–85% B over 8 min at a flow rate of 0.35 mL/min. The proportion of TFA in the mobile phase and gradient selected with respect to the balance between clear separation of the POA and PAE peaks and period of elution. Signal was detected by photodiode array at a wavelength of 270 nm. The same quantification method was utilized to determine the average percentage of each drug in the bulk powder.

The mass median aerodynamic diameter (MMAD) was calculated by determining the median (d50) of the cumulative percentage of mass of each drug delivered to the impactor stages, on a probability scale, plotted against the logarithm of its corresponding aerodynamic diameter.

2.2.3.2. Laser diffraction.

Laser diffraction data was obtained for the PAE powder (Mastersizer 2000, Malvern Instruments, Worcestershire, UK, with Sirocco 2000 (A) dry powder dispersion system). Approximately 100 mg of spray dried powder was deposited into the sample tray. Triplicate measurements were taken for 5 s each with a dispersive air pressure of 4.0 bar and a vibration feed rate between 40% and 60%.

2.2.4. Powder disperser and nose-only exposure chamber

Many mechanisms of dispersing dry powder efficiently have been described in the literature. (Gill et al., 2006) The intent of the present study was to administer the powder to a small-contained volume, to minimize dose required to maximize airborne concentration. Dispersion methods relying on high velocity airflow, such as a Venturi system, to achieve high aerosol concentration would not be suitable for these studies as very large drug quantities would be required.

A rotating brush based system was designed consisting of a spiral-wound conical brush and a complimentary housing, constructed from a centrifuge tube notched at the bottom. Powder was introduced at the top of the housing. The brush was rotated at fixed speed using a DC motor (Maxon Precision Motors, Inc., San Antonio, TX). The spiral brush acts as an auger to move the bulk powder to the bottom of the housing while loading the brush bristles. At the bottom of the housing the bristles enter the notches and straighten, dispersing the particles into the chamber. A small DC brushless fan (SparkFun Electronics, Niwot, CO) below the dispersion unit mixes the air and distributes the particles throughout the chamber (Fig. 1).

Fig. 1.

Fig. 1.

Open in a new tab

Aerosol dispersion head and chamber. (a) Six animal exposure chamber shown without restraint tubes with (b) dry powder dispersion apparatus schematic consisting of a powder reservoir constructed from a modified centrifuge tube and a wound brush, (b) six animal exposure chamber with 2.47 L volume central housing and (c) actual dispersion apparatus depicting the brush, housing and fan.

The chamber was constructed of acrylic tube with 6 animal ports sufficiently large to accommodate guinea pigs over the period of the study. The nose-only exposure ports were tiered in three levels with 2 opposite ports on each level. The levels were offset 120°. The central chamber interior diameter was 4 inches with a total height of 12 inches. Nose ports were 1 inch in diameter. Restraint tubes were 3.8 inches inner diameter. Tapered cones to guide the animal’s nose into the chamber were fabricated using a custom mold and food grade epoxy resin (EPON 862 cured with 10% DMP30).

2.2.4.1. Chamber residence time characterization with MicroPEM.

To assess the duration of aerosol residence in the chamber following dispersion, 3 personal aerosol exposure monitors were employed (RTI MicroPEM, NC, USA (Rodes et al., 2012)). The MicroPEM features a calibrated nephelometer and single stage impactor. Adjusting the onboard pump to a flow rate of 0.2 L/min, approximately twice the respiration rate of a guinea pig (Amdur and Mead, 1958) shifts the impactor cutoff from 2.5 μm to 4 μm. Three MicroPEM units were initialized and placed into the restraint tubes at each tier, with the inlet positioned at the exposure hole. To characterize the powder disperser performance a mass of 10 mg of powder was loaded into the dispersion head and the rotating brush and the fan were turned on simultaneously and run for 30 s. Nephelometer data was collected at 3 s intervals.

2.2.4.2. Chamber dose determination.

To determine an approximate dose delivered to each animal, 10 mg of PAE powder was loaded into the disperser. A personal environmental monitor (PEM) aerosol sampler (MSP Corporation, Shoreview, MN) with a cutoff of 2.5 μm was placed at the bottom of the chamber and operated at 2 L/min. Based on the MicroPEM duration assessment, the impactor was operated for 3 min. The dispersion head was turned on for the first 30 s of the 3 min operation time, after which the pump was turned off. The 37 mm diameter PTFE filter was removed from the impactor, washed and quantified via UPLC.

3. Results

3.1. Methods

3.1.1. Spray drying

The yield of the spray drying process was 62.5 ± 1.3% (n = 3) of total solids.

3.2. Characterization

3.2.1. SEM

Particles appeared spherical and solid by scanning electron microscopy with a smooth surface (Fig. 2).

Fig. 2.

Fig. 2.

Open in a new tab

Scanning electron micrographs of PAE powder at two different magnifications.

3.2.2. Thermal analysis

The melting point of PAE, a liquid at room temperature, as determined by DSC was 4.16 °C. Analysis of the PAE powder shows two endothermic events in the range analyzed at 180° and 214 °C. Maltodextrin exhibits a broad endothermic peak at approximately 120 °C, and an exothermic event that onsets at approximately 255 °C. POA displays a very sharp endothermic peak at 225 °C. Leucine begins an endothermic event at approximately 280 °C, the peak of which was not resolved over the temperature range evaluated (Figs. 3 and 4).

Fig. 3.

Fig. 3.

Open in a new tab

DSC thermogram used to establish melting point of PAE. Tm = 4.16° C.

Fig. 4.

Fig. 4.

Open in a new tab

DSC thermograms for PAE powder, POA and bulk excipients (Maltodextrin and l-leucine).

TGA (Fig. 5) shows curves consistent with the constituents in the final spray dried powder containing PAE powder with onset of thermal degradation consistent with POA, leucine and maltodextrin occurring at 100, 227 and 278 °C, respectively. A loss of mass prior to 100 °C may correspond to PAE.

Fig. 5.

Fig. 5.

Open in a new tab

TGA thermograms for PAE powder, POA and bulk excipients (maltodextrin and l-leucine).

3.2.3. XRPD

X-ray diffraction peaks (Fig. 6) for PAE powder were evident, indicating crystallinity, with defined peaks at 14.315, 19.041, 24.081 and 27.602 2-theta. POA shows one major peak at 27.698° 2-theta, with smaller peaks (<10% area) at 14.361, 19.899, 31.279 and 57.101° 2-theta. Maltodextrin was amorphous without well-defined peaks in the spectrum. Leucine had peaks at 12.117, 24.353, 30.559, and 36.874 2-theta.

Fig. 6.

Fig. 6.

Open in a new tab

XRPD of PAE powder, POA and bulk excipients (maltodextrin and l-leucine). Moisture content of spray dried powder as determined by Karl Fischer analysis was 2.44 ± 0.1%.

Moisture content of spray dried powder as determined by Karl Fischer analysis was 2.44 ± 0.1%.

3.3. Aerosol particle size determination

3.3.1. Inertial impaction

Analysis by inertial impaction revealed slight differences in the mass median aerodynamic diameter (MMAD) relative to the two forms of the drug in the PAE powder, with a MMAD of 3.05 ± 0.04 μm (geometric standard deviation, GSD 1.54 ± 0.03) and 2.68 ± 0.12 μm (GSD 1.85 ± 0.05) for POA and PAE respectively. Distribution of the two drugs within the particle sizes also varied. Higher PAE/POA ratio was observed in smaller particles. Emitted dose taken as the percentage of each drug delivered to the impactor and not retained in either the capsules or the inhaler was 76.0% ± 5.7% and 77.9% ± 5.7% for POA and PAE respectively. Fine particle fraction, expressed as proportion of the mass deposited on stage 3 and below (<4.46 μm) of the total mass was 53.9% ± 3.7% and 60.5% ± 4.5% for POA and PAE respectively (Fig. 7).

Fig. 7.

Fig. 7.

Open in a new tab

Results of inertial impaction of PAE powder using a Next Generation Impactor at 60 L/min showing stage deposition. Error bars represent one standard deviation above and below the mean (n = 3).

3.3.2. Laser diffraction

Volume median particle size (d50) as determined by laser diffraction was 1.52 μm (±0.05) with a span ((d90–d10)/d50) of 1.705 (n = 3) (Fig. 8).

Fig. 8.

Fig. 8.

Open in a new tab

Laser diffraction cumulative percent undersize, representative distribution.

3.4. Powder Disperser and Nose-only Exposure chamber

3.4.1. Chamber residence time characterization by MicroPEM

Nephelometer signal increased rapidly at actuation to saturate the detectors with concentrations at each tier over 10000 μg/m3, and remained saturated for approximately 2.6 min. Particle concentration in the chamber decreased to near baseline levels by approximately 5.3 min. The concentration measured at the top tier persisted longer than the others, returning to baseline after approximately 6.5 min. The concentration in the chamber when the dose was administered was 4.17 g/m3, based on the total mass delivered (10 mg) and the volume of the chamber (2.47 L). Consequently, the threshold at which the MicroPEM registers a decline in particle concentration (10 mg/m3) is approximately 2% of the starting concentration and sufficient to indicate when a subsequent dose could be administered without accumulation. Therefore, to prevent accumulation of drug in the chamber the frequency of dosing should not be more than once every 3 min (Fig. 9).

Fig. 9.

Fig. 9.

Open in a new tab

Aerosol lifetime in the exposure chamber of PAE powder dispersed by the brush-based powder disperser for 30 s period of generation as determined by MicroPEM nephelometer data.

3.4.2. Chamber dose

The impactor collected 13.4 μg POA and 1.7 μg PAE over the three minutes. The PAE/POA ratio for the collected powder was 0.126.

4. Discussion

The yield from the spray drying process was acceptable to support initial studies of aerosol dispersion and delivery. Spray dried particles were spherical in appearance and in a size range suitable for aerosol delivery.

The melting point of PAE, a liquid at room temperature was slightly above the freezing point of water and at common refrigeration temperature. Thermogravimetric analysis under nitrogen flow revealed that the pyrazinoate ester had negligible vapor pressure below approximately 50 °C relative to water, suggesting that significant evaporation occurs at approximately this temperature of the pure PAE. This was used to set a target for the outlet temperature during spray drying to minimize evaporation of the ester before the particle had dried.

Spray drying the feedstock liquid at 85 °C resulted in an outlet temperature of 45° C, below the approximate target and just below the reported boiling point of the ester. The atomizing conditions employed a high mass flow ratio of atomizing gas to feedstock liquid considering the feedstock concentration (10 mg/mL) to produce particles in the respirable range (1–5 μm).

DSC analysis of PAE powder showed two distinct endothermic events that do not directly correspond with any of the bulk excipients (maltodextrin and l-leucine), potentially indicating an interaction with the ester or between excipients. This could be further examined by spray drying the feedstock solution without PAE and subsequent thermal and powder diffraction analysis.

XRD pattern analysis comparing the spray dried powder to unprocessed components show the dominant peak at approximately 27.7° 2-theta may correspond to the major explicit peak observed for crystalline POA, with additional shared peak locations at approximately 14.32, 19.8 and 31° 2-theta, though these peaks are not prominent in the diffraction pattern of POA and the differences approximate the margin of error of the method. This may indicate some presence of crystalline POA, but the DSC data suggests a crystalline structure corresponding to the melting temperature of crystalline POA alone is absent. Leucine also displayed a highly crystalline peak, but correspondent peaks indicating crystalline leucine were not seen in the spray dried powder. Small, dense particles are difficult to deaggregate due to high interparticulate forces from increased surface. Deaggregation to a respirable size is critical for efficiency of dose delivery of aerosol powders. Even at concentrations insufficient to saturate at the surface and alter particle morphology, incorporation of leucine has been shown to increase the fine particle fraction and improve dispersion. (Chew et al., 2005)

PAE was encapsulated into the dry powder, composed of approximately 8.5% of the bulk powder mass, for an encapsulation efficiency of 34%. However, aerodynamic sizing of PAE powder showed slight differences in the distribution of the two drug forms in the product. Aerodynamic sizing of active components revealed both were in the respirable range, and comparison with laser diffraction data suggests some degree of aggregation as particles were dispersed from the dry powder inhaler. PAE was more broadly distributed than POA in the particles, with less proportion of PAE in the larger particle sizes, with the largest particles having only 16% of mass relative to POA (Fig. 10). The particles collecting on stage 6 and 7 (less than 0.94 and 0.55 μm, respectively) contained approximately 40–45%. As POA has a melting point of roughly 225 °C and the ester has a reported boiling point of 46–48 °C (Cynamon et al., 1992), it is probable that ester is evaporating with water as the particle dries, while an insignificant amount of POA is volatilized.

Fig. 10.

Fig. 10.

Open in a new tab

Ratio of PAE to POA across the impactor stages, showing higher PAE incorporation at smaller particle sizes. Error bars represent one standard deviation above and below the mean (n = 3).

Atomized droplets should have been homogeneous and contain the same proportion of components as all components dissolved. Smaller droplets have higher surface area and less total water, and dry faster at the same temperature than larger particles. Once the powder is dry, the ester appears to be encapsulated sufficiently to prevent further loss as the particles are still under high airflow in the cyclone while spray drying proceeds. When determining the effect of spray drying process variables on powder properties encapsulating fish oil an increase in the drying temperature was observed that led to an increase in encapsulation efficiency. The higher efficiencies were attributed to a rapid formation of a crust which prevents the oil from migrating to the surface where it may evaporate (Aghbashlo et al., 2013). Increasing the feed rate decreased encapsulation efficiency suggesting that larger particles have lower efficiencies.

Following this rationale, drying the particles faster could improve the encapsulation efficiency. Increasing the inlet temperature may be one way to achieve this, but due to the increasing volatility of the ester with increasing temperatures, this approach alone may not be sufficient to improve efficiency substantially, and increasing the inlet temperature also increases the outlet temperature, which should be kept below the boiling point of the ester. Replacement of water with more volatile evaporative component by increasing the ethanol concentration would allow the particles to dry faster at the same temperature and may improve the encapsulation efficiency. Decreasing the droplet size by adjusting atomization parameters could also improve encapsulation. However, noting the relationship between droplet size and encapsulation efficiency, to normalize the PAE content across size distribution would require determining process parameters that increase encapsulation efficiencies to the maximum capacity of the excipients to retain the ester. Alternatively, a monodisperse aerosol generator could be employed to ensure narrow size distributions and therefore comparable PAE content, though this will certainly increase production time.

MicroPEM data showed that aerosol concentrations within the chamber diminish after approximately 3 min, and that a subsequent dose could be administered at this time point without substantial concentration escalation. At a 4 μm cutoff, it also demonstrated that the dispersion device was suitable to disperse the powder efficiently for inhalation.

Scaling the impactor dose to 100 mg and adjusting for respiration rate and number of animals, over 3 min each animal would receive 11.13 μg POA and 1.41 μg PAE. If powder is administered at 3 min intervals for 5 intervals, this increases the dose to 55.67 μg POA and 7.03 μg PAE per animal. For tuberculosis studies of effectiveness of aerosols the animal of choice is the guinea pig for which there is a TB model (Padilla-Carlin et al., 2008). For a 350 g guinea pig, this would equate to 0.16 mg/kg POA and 0.02 mg/kg PAE per day. Ultimately, this dose would require confirmation by comparison to PK using known directly deposited doses achieved by endotracheal intubation of animals and insufflation directly to the lungs (Fiegel et al., 2008). Delivered locally, the lung concentrations would be much higher than at an oral dose of the same mass. The lower PAE/POA ratio observed in the impaction data may be a function of the volatility of the ester, which, while substantially reduced by the formulation, appears not to be completely mitigated. Inertial impaction results suggest the ester is incorporated at higher efficiency in smaller particles and is retained through the impaction process, which involves an airflow of 4 L (at 60 L/min), compared to the total air volume of the sampling impactor of 6 L (over 3 min). If the ester volatility was the main contributor to the difference in PAE/POA ratio observed in the sampling impactor, it would be expected that the smaller particles would have less PAE due to the higher surface area, and this is not reflected in the inertial impaction sizing data. This observation requires further investigation. Pharmacokinetic evaluation of animals in the chamber would arrive at a definitive dose, as airflow in the dosing chamber is minimal during the exposure. Future pharmacokinetic and pharmacodynamics studies were the justification for the performance of the experiments described here.

5. Conclusion

We have developed a respirable dry powder combining two forms of POA each of which offers potential for enhanced therapy for tuberculosis therapy, one from local delivery and the other from inherently higher potency. The use of this combination may improve outcomes in the treatment of multiple and extensively drug resistant tuberculosis. The aerodynamic particle size distribution is appropriate for inhalation therapy and for delivery to a small volume nose-only exposure chamber in high concentrations requiring minimal drug for the evaluation of efficacy in a guinea pig model of tuberculosis.

Acknowledgements

The authors acknowledge the support of the NC TraCS Institute of the University of North Carolina at Chapel Hill, through a 4D Pilot Grant to Drs. Braunstein and Hickey.

References

  1. Acosta CD, Dadu A, Ramsay A, Dara M, 2014. Drug-resistant tuberculosis in Eastern Europe: challenges and ways forward. Public Health Act. 4, S3–s12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Aghbashlo M, Mobli H, Madadlou A, Rafiee S, 2013. Fish oil microencapsulation. as influenced by spray dryer operational variables. Int. J. Food Sci. Technol 48, 1707–1713. [Google Scholar]
  3. Amdur MO, Mead J, 1958. Mechanics of respiration in unanesthetized Guinea pigs. Am. J. Physiol. Legacy Content 192, 364–368. [DOI] [PubMed] [Google Scholar]
  4. Chang Y-X, Yang J-J, Pan R-L, Chang Q, Liao Y-H, 2014. Anti-hygroscopic effect of leucine on spray-dried herbal extract powders. Powder Technol. 266, 388–395. [Google Scholar]
  5. Chew NY, Shekunov BY, Tong HH, Chow AH, Savage C, Wu J, Chan HK, 2005. Effect of amino acids on the dispersion of disodium cromoglycate powders. J. Pharm. Sci 94, 2289–2300. [DOI] [PubMed] [Google Scholar]
  6. Cipolla D, Shekunov B, Blanchard J, Hickey A, 2014. Lipid-based carriers for pulmonary products: preclinical development and case studies in humans. Adv. Drug Deliv. Rev 75, 53–80. [DOI] [PubMed] [Google Scholar]
  7. Cohen KA, Abeel T, Manson McGuire A, Desjardins CA, Munsamy V, Shea TP, Walker BJ, Bantubani N, Almeida DV, Alvarado L, Chapman SB, Mvelase NR, Duffy EY, Fitzgerald MG, Govender P, Gujja S, Hamilton S, Howarth C, Larimer JD, Maharaj K, Pearson MD, Priest ME, Zeng Q, Padayatchi N, Grosset J, Young SK, Wortman J, Mlisana KP, O’Donnell MR, Birren BW, Bishai WR, Pym AS, Earl AM, 2015. Evolution of extensively drug-resistant tuberculosis over four decades: whole genome sequencing and dating analysis of mycobacterium tuberculosis isolates from kwaZulu-Natal. PLoS Med. 12, e1001880. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cynamon MH, Klemens SP, Chou TS, Gimi RH, Welch JT, 1992. Antimycobacterial activity of a series of pyrazinoic acid esters. J. Med. Chem 35, 1212–1215. [DOI] [PubMed] [Google Scholar]
  9. Draper P, 1998. The outer parts of the mycobacterial envelope as permeability barriers. Front. Biosci. J. Virtual Lib 3, D1253–D1261. [DOI] [PubMed] [Google Scholar]
  10. Durand F, Jebrak G, Pessayre D, Fournier M, Bernuau J, 1996. Hepatotoxicity of antitubercular treatments. Rationale for monitoring liver status. Drug Saf. 15, 394–405. [DOI] [PubMed] [Google Scholar]
  11. Fiegel J, Garcia-Contreras L, Thomas M, VerBerkmoes J, Elbert K, Hickey A, Edwards D, 2008. Preparation and in vivo evaluation of a dry powder for inhalation of capreomycin. Pharm. Res 25, 805–811. [DOI] [PubMed] [Google Scholar]
  12. Fuchs M, Turchiuli C, Bohin M, Cuvelier ME, Ordonnaud C, Peyrat-Maillard MN, Dumoulin E, 2006. Encapsulation of oil in powder using spray drying and fluidised bed agglomeration. J. Food Eng 75, 27–35. [Google Scholar]
  13. Gill TE, Zobeck TM, Stout JE, 2006. Technologies for laboratory generation of dust from geological materials. J. Hazard. Mater 132, 1–13. [DOI] [PubMed] [Google Scholar]
  14. Goula AM, Adamopoulos KG, 2008. Effect of maltodextrin addition during spray: drying of tomato pulp in dehumidified air: II. Powder properties. Drying Technol. 26, 726–737. [Google Scholar]
  15. Heifets LB, Flory MA, Lindholm-Levy PJ, 1989. Does pyrazinoic acid as an active. moiety of pyrazinamide have specific activity against Mycobacterium tuberculosis? Antimicrob. Agents Chemother 33, 1252–1254. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hickey AJ, Misra A, Fourie PB, 2013. Dry powder antibiotic aerosol product development: inhaled therapy for tuberculosis. J. Pharm. Sci 102, 3900–3907. [DOI] [PubMed] [Google Scholar]
  17. Hickey AJ, Durham PG, Dharmadhikari A, Nardell EA, 2015. Inhaled drug. treatment for tuberculosis: past progress and future prospects. J. Control. Release published online November 2015. [DOI] [PubMed] [Google Scholar]
  18. Hoppentocht M, Hagedoorn P, Frijlink HW, de Boer AH, 2014. Technological and practical challenges of dry powder inhalers and formulations. Adv. Drug. Deliv. Rev 75, 18–31. [DOI] [PubMed] [Google Scholar]
  19. Karo B, Hauer B, Hollo V, van der Werf MJ, Fiebig L, Haas W, 2015. Tuberculosis treatment outcome in the European Union and European Economic Area: an analysis of surveillance data from 2002 to 2011. Euro Surveill. Bull. Eur. sur les maladies Transm. = Eur. Commun. Dis. Bull. 20 [DOI] [PubMed] [Google Scholar]
  20. McDermott W, Tompsett R, 1954. Activation of pyrazinamide and nicotinamide in acidic environments in vitro. Am. Rev. Tuberculosis 70, 748–754. [DOI] [PubMed] [Google Scholar]
  21. Mitchell W, Seidel W, 1974. Preparation of an alcohol-containing powder. Google Patents. [Google Scholar]
  22. Padilla-Carlin DJ, McMurray DN, Hickey AJ, 2008. The guinea pig as a model of infectious diseases. Comp. Med 58, 324–340. [PMC free article] [PubMed] [Google Scholar]
  23. Patton JS, Byron PR, 2007. Inhaling medicines: delivering drugs to the body through the lungs. Nature reviews. Drug Discov. 6, 67–74. [DOI] [PubMed] [Google Scholar]
  24. Peterson ND, Rosen BC, Dillon NA, Baughn AD, 2015. Uncoupling environmental pH and intrabacterial acidification from pyrazinamide susceptibility in mycobacterium tuberculosis. Antimicrob. Agents Chemother 59, 7320–7326. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Raynaud C, Laneelle MA, Senaratne RH, Draper P, Laneelle G, Daffe M, 1999. Mechanisms of pyrazinamide resistance in mycobacteria: importance of lack of uptake in addition to lack of pyrazinamidase activity. Microbiology (Reading, England) 145 (Pt 6), 1359–1367. [DOI] [PubMed] [Google Scholar]
  26. Roche N, Huchon GJ, 2000. Rationale for the choice of an aerosol delivery system. J. Aerosol Med. Off. J. Int. Soc. Aerosols Med 13, 393–404. [DOI] [PubMed] [Google Scholar]
  27. Rodes CE, Chillrud SN, Haskell WL, Intille SS, Albinali F, Rosenberger ME, 2012. Predicting adult pulmonary ventilation volume and wearing complianceby on-board accelerometry during personal level exposure assessments. Atmos. Environ 57, 126–137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Smaoui Fourati S, Mzid H, Marouane C, Kammoun S, Messadi-Akrout F, 2015. [Multidrug-resistant tuberculosis: epidemiology and risk factors]. Rev. Pneumol. Clin 71, 233–241. [DOI] [PubMed] [Google Scholar]
  29. Smith I, 2003. Mycobacterium tuberculosis pathogenesis and molecular determinants of virulence. Clin. Microbiol. Rev 16, 463–496. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. U.S. Pharmacopeial Convention, 2011. U.S. Pharmacopeia National Formulary 2011: USP 34 NF 29. United States Pharmacopeial. [Google Scholar]
  31. World Health Organization,, 2015. Global tuberculosis report 2015. World Health Organization. [Google Scholar]
  32. Young EF, Braunstein MS, Durham PG, Hickey AJ, Zhong L, Welch JT, 2015. A Comparison of the Anti-tuberculous Activity of Low Dose PZA and Aerosol Pyrazinoate Esters in a Guinea Pig Model of Infection. Host Response in Tuberculosis, Keystone Symposia. [Google Scholar]
  33. Yuen CM, Kurbatova EV, Tupasi T, Caoili JC, Van Der Walt M, Kvasnovsky C, Yagui M, Bayona J, Contreras C, Leimane V, Ershova J, Via LE, Kim H, Akksilp S, Kazennyy BY, Volchenkov GV, Jou R, Kliiman K, Demikhova OV, Vasilyeva IA, Dalton T, Cegielski JP, 2015. Association between regimen composition and treatment response in patients with multidrug-Resistant tuberculosis: a prospective cohort study. PLoS Med. 12, e1001932. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Zhang Y, Mitchison D, 2003. The curious characteristics of pyrazinamide: a review. Int. J. Tuberculosis Lung Dis. Off. J. Int. Union Against Tuberculosis Lung Dis 7, 6–21. [PubMed] [Google Scholar]
  35. Zhang Y, Shi W, Zhang W, Mitchison D, 2014. Mechanisms of pyrazinamide: action and resistance. Microbiol. Spectr 2 (Mgm2-0023-2013). [DOI] [PubMed] [Google Scholar]

RESOURCES