Development of PLGA dry powder microparticles by supercritical CO2-assisted spray-drying for potential vaccine delivery to the lungs

The Journal of Supercritical Fluids 128 (2017) 235–243

Contents lists available at ScienceDirect

The Journal of Supercritical Fluids journal homepage: www.elsevier.com/locate/supflu

Development of PLGA dry powder microparticles by supercritical CO2assisted spray-drying for potential vaccine delivery to the lungs

MARK

Márcia Tavaresa, Renato P. Cabrala, Clarinda Costaa, Pedro Martinsb, Alexandra R. Fernandesb, ⁎ ⁎ Teresa Casimiroa, , A. Aguiar-Ricardoa, a b

LAQV–REQUIMTE, Departamento de Química, Faculdade de Ciências e Tecnologia, Universidade NOVA de Lisboa, 2829-516 Caparica, Portugal UCIBIO–REQUIMTE Departamento de Ciências da Vida, Faculdade de Ciências e Tecnologia, Universidade NOVA de Lisboa, 2829-516 Caparica, Portugal

A R T I C L E I N F O

A B S T R A C T

Keywords: Dry powder inhalation Pulmonary drug delivery Poly(lactic-co-glycolic acid) Supercritical assisted atomization DOE

In this work, biocompatible and biodegradable poly(D-L-lactide-co-glycolide) (PLGA) composite microparticles with potential use as carrier for vaccines and other drugs to the lungs were developed using supercritical CO2assisted spray-drying (SASD). Bovine serum albumin (BSA) was chosen as model vaccine, and L-leucine as a dispersibility enhancer, and their effects on the particle characteristics were evaluated. The dry powder formulations (DPFs) were characterized in terms of their morphology and aerodynamic performance using an in vitro aerosolization study – Andersen cascade impactor (ACI) − to obtain data such as the fine particle fraction (FPF) with percentages up to 43.4%, and the mass median aerodynamic diameter (MMAD) values between the 1.7 and 3.5 μm. Additionally, pharmacokinetic and cytotoxicity studies were performed confirming that the produced particles have all the necessary requirements for potential pulmonary delivery.

1. Introduction The local application of drugs to the respiratory tract via inhalation facilitates a site specific treatment of lung diseases with higher treatment efficacy, lower systemic exposure and consequently, reduced side effects [1,2]. It is also an advantageous route for systemic drug delivery since it shows a high solute permeability which facilitates gas exchange via diffusion [3], due to its very thin absorption membrane (0.1–0.2 μm), to its elevated blood flow (5L/min) and to the highly vascularized alveolar epithelium constituted by a single layer of cells, which offers a large absorptive surface (80–100 m2) [4–6]. It also has an easy administration, shows early effects of drugs’ pharmacological actions and has no risk of drug decomposition [7,8]. Furthermore, unlike the oral route, it is not subject to first pass metabolism which is especially important to macromolecules (i.e. peptides and proteins) that are easily degraded by enzymes [9]. Controlled drug release systems composed of polymeric materials with particular characteristics, such as biocompatibility and degradability, have been shown to improve the pharmacokinetic and pharmacodynamic profiles of encapsulated drugs in the lung [1,7,10]. Vaccines are responsible for death prevention associated with numerous infection diseases every year, even though there is still a huge amount of children morbidity due to vaccine-preventable diseases [11]. Vaccines administered parenterally require expensive cold chain

⁎

transport and trained personnel, and they can induce needle-stick injuries with possible transmission of viruses [12,13]. To treat the infectious diseases that affect poor populations the vaccines should be simple, cheap, easy to produce and stable. This is all possible with pulmonary vaccination added to all the advantages for drug delivery already mentioned [14]. Above that is the fact that the lungs have gained a lot of attention given that the respiratory tract is the main entry of pathogens and also it has an extensive dendritic cell network lining the airway epithelium that facilitates a first line of defence for antigens [15]. To efficiently deliver powder formulation to the lungs, an inhaler should generate an aerosol of a suitable size with a reproducible drug dosing while ensuring chemical stability and activity [5]. The dry powder inhaler (DPI), presents all these characteristics, being also propellant-free [16,17]. As for the delivery of vaccines this device is very suitable once macromolecules (polysaccharides, proteins and peptides) tend to degrade when in a liquid solution and so are provided with a greater stability [12]. In order to understand where particles with different size deposit in the respiratory tract, the aerodynamic diameter (da) has to be taken into account and is defined by the equation

da = d g √

Corresponding author. E-mail addresses: [email protected] (T. Casimiro), [email protected] (A. Aguiar-Ricardo).

http://dx.doi.org/10.1016/j.supflu.2017.06.004 Received 5 March 2017; Received in revised form 6 June 2017; Accepted 7 June 2017 Available online 08 June 2017 0896-8446/ © 2017 Elsevier B.V. All rights reserved.

ρp ρo χ

The Journal of Supercritical Fluids 128 (2017) 235–243

M. Tavares et al.

Fig. 1. Schematic representation of the SASD apparatus: (CB) cryogenic bath; (LP) liquid pump; (HB) heating bath; (TC) temperature controller; (M) manometer; (S) saturator; (P) precipitator; (c) cyclone.

Where dg is the geometric diameter, ρ is the mass density of the particle, ρo is the unit density and χ the particle dynamic shape factor [4]. Though some conflicts regarding the range for optimal da to efficiently reach the deep lungs may exist, the most recent works state that the aerodynamic diameter of aerosol particles should be between 1 and 5 μm since this size range is the one that reaches to the deep lungs [18–20]. Poly(lactic-co-glycolic acid) (PLGA) is used as a biodegradable slowrelease polymer, it is physically strong, with low toxicity, highly biocompatible and bioabsorbable and also allows an effective encapsulation of all kinds of drugs, proteins and various other macromolecules such as DNA, RNA and peptides [21–23]. Additionally, it was approved by the FDA (Food and Drug Administration) for drug targeting in the year 2000 [23], being one of the first among all the new minimally invasive protein delivery system options, which makes it one of the most promising materials in medical and biotechnological research with high commercial interest [24,25]. More recently, alternative particle production methods using supercritical fluids (SCF), especially supercritical carbon dioxide (scCO2), have been proposed to overcome some limitations. Conventional methods usually require large amounts of organic solvents and thus involve additional extensive purification steps to remove the residual solvent. Carbon dioxide is the most used SCF, since it is nontoxic, nonflammable, inexpensive and readily available in high purity from a variety of sources [26–28]. Besides, it has a low critical point (31.1 °C and 73.8 bar), which is suitable to use with heat-sensitive materials, being able to reduce manufacturing complexity and energy [29]. Supercritical assisted atomization (SAA) or supercritical assisted spraydrying (SASD) is based on the solubility of controlled quantities of scCO2 in a liquid solution containing the drug and excipient. The mixing happens in a heated saturator with a large contacting surface, assuring a high residence time and a near-equilibrium solubilisation of the CO2 [30,31]. In this work, composite particles using PLGA as an excipient were produced using SASD. Bovine serum albumin (BSA) was used as a model antigen [32], and L-leucine as a dispersibility enhancer [33] in order to develop pulmonary vaccines. Furthermore, all the particles were properly characterized to make sure they were suitable for pulmonary delivery.

2. Experimental section 2.1. Materials Poly(D-L-lactide-co-glycolide) (PLGA, PURASORB® PDLG 5002A, molar ratio: 50/50, inherent viscosity 0.21 dL/g) was purchased from PURAC (Gorinchem, Netherlands). Bovine Serum Albumin (BSA, Mw 60 kDa), L-Leucine (> 98%) and acetonitrile were obtained from Sigma-Aldrich. Industrial carbon dioxide (purity ≥ 99.93%) was obtained from Air Liquide. 2.2. Design of experiment A design of experiment (DoE) approach was adopted using Statistica™ software version 10 (Statsoft, Bell, Tulsa, Oklahoma) to understand the effect of some of the formulation variables on the particles size [34,35]. In this case the absence or presence of both L-leucine and BSA as well as two different CO2-to-liquid solution volumetric flow ratios (8.3 and 5). All the experiments were made in triplicate and the statistical differences in between different formulations was obtained using ANOVA. The analysis was performed considering changes in Dv,50, MMAD and FPF. 2.3. Microparticles preparation A solution of 0.47% (w/v) PLGA, 0.09% (w/v) BSA and 0.02% (w/ v) L-leucine was prepared with 83% (v/v) acetonitrile and 16% (v/v) distilled water. While the PLGA was dissolved in part of the acetonitrile, both the BSA and leucine were dissolved in water. After full dissolution, the aqueous solution was added to the PLGA solution under stirring, with the rest of the acetonitrile added in the end. Particles were produced using a laboratory scale SASD apparatus, represented in Fig. 1 [36–38]. Two high-pressure pumps to deliver the liquid solution (HPLC pump 305 Gilson) and the CO2 (HPLC pump K501, Knauer) to the static mixer. Since the CO2 pump is suitable for liquid solutions, there is a need to liquefy the CO2 in a cryogenic bath. Before entering the static mixer, the CO2 is heated in an oil bath. The static mixer (3/16 model 37-03-075 Chemieer) is a high-pressure vessel with a 4.8 mm diameter, 191 mm length and 27 helical mixing 236

The Journal of Supercritical Fluids 128 (2017) 235–243

M. Tavares et al.

PerkinElmer DSC 7. Samples (10–30 mg) were placed in a holder where the temperature was raised from 25 °C to 300 °C in a heating rate of 10 °C/min in nitrogen.

elements that allows the solubilisation of the CO2 into the liquid solution due to its high surface packing. In order to control the temperature of the mixture, the static mixer is enrolled by heating tapes controlled by a Shinko FCS-13A temperature controller. The obtained mixture is then atomized into the precipitator in the form of a spray by the passage through a nozzle with an internal diameter of 150 μm. At the same time, a flow of previously heated compressed air enters the precipitator to evaporate the liquid solvent. The visual precipitator is an aluminium vessel that operates at near-atmospheric conditions heated by a set of heating tapes that are connected to a Setra pressure transducer connected to the top of it. After the precipitator, there is a heating tape set to make sure the flow remains at high temperatures, thus avoiding heat losses and consequent solvent condensation, connected to an Isopad TD 2000 heat controller. The particles will then be collected from a filter (FSI, nylon mono-filament, 1 μm) which is placed in a cyclone that helps with the separation between the particles and the CO2-solvent flow.

2.9. Fourier transform infra-red FT-IR was performed in order to study the interactions between the excipient and drug, and also to study the BSA’s second structure. The spectra were carried out on a PerkinElmer spectrum 1000 FTIR coupled with Opus Spectroscopy Software with potassium bromide (KBr) tablets containing 20% sample. 2.10. Powder flowability The aerodynamic performance of the formulations was assessed using an aluminium Andersen Cascade Impactor apparatus (ACI, Copley). The assays are performed gravimetrically using the European Pharmacopeia protocol [39]. Glass microfiber filters (80 mm, Filter Lab) were placed in all the stages and were weighted before and after the capsules release. Each capsule was drawn through the induction port into the ACI at a flow rate of 62 ± 2 L/min for 3.9 ± 0.2 s. The effective cut-off aerodynamic diameter for each stage of the ACI were: stage −1, 8.6 μm; stage 0, 6.5 μm; stage 1, 4.4 μm; stage 2, 3.2 μm; stage 3, 1.9 μm; stage 4, 1.2 μm; stage 5, 0.55 μm; stage 6, 0.26 μm. The fine particle fraction (FPF) was determined by the interpolation of the percentage of the particles containing less than 5 μm. The mass median aerodynamic diameter (MMAD) was determined as the particle diameter corresponding to 50% of the cumulative distribution.

2.4. Particle size distribution The particle size (PS) and the particle size distribution (PSD) were measured in a Morphologi G3 equipment from Malvern. Morphologi G3 has an automatic sample dispersion unit coupled to a microscope in which different objectives are controlled by a software. Each of more than 30000 particles are photographed and statistically analysed according to different geometrical parameters. Also, the span was calculated, a measure of the width of the particle distribution, and it considers the Dv,10, Dv,50 and Dv,90 (particle diameter in volume corresponding to 10, 50 and 90% of the population) as represented in Eq. (2):

Span =

Dv,90 − Dv,10 Dv,50

2.11. Pharmacokinetic studies The in vitro release studies were carried out in phosphate-buffered saline at pH 7.4 (PBS). 60 ± 2 mg of sample was introduced in a cellulose ester dialysis membrane (MWCO 100,000, 24 mm x 15 feet dry diameter, 1.8 mL/cm, Spectrum Labs) that was suspended in 5 mL of PBS. Samples of 0.5 mL were taken at predetermined time points and other 0.5 mL of PBS were added to the suspension and the samples. BSA was quantified by UV spectroscopy (PerkinElmer Lambda 35) at a wavelength of 280 nm.

(2)

2.5. Particles morphology The morphology of the produced particles was studied by Scanning Electron Microscopy (SEM) equipment from Hitachi, S-2400 instrument, with an accelerating voltage set to 15 kV and at magnifications of 5 K and 10 K. All the samples were dispersed on carbon tape previously attached to an aluminium stub and were coated with gold before analysis.

2.12. In vitro cytotoxicity In vitro cytotoxicity assays were performed using a lung adenocarcinoma cell line (A549), through the use of CellTiter 96® AQueous Non-Radioactive Cell Proliferation Assay (Promega, Madison, USA), a colorimetric method for determining the number of viable cells [40]. At 80% confluence, cells were harvested and centrifuged. The supernatant was discarded, and the cell pellet re-suspended in 2 mL of medium. Cell density was evaluated as the total number of viable cells within the grids on the hemacytometer (Hirschmann, Eberstadt, Germany) using trypan blue exclusion method. For this procedure 350 μL of medium was pipetted to a 2 mL eppendorf together with 50 μL of the 2 mL cell suspension followed by 100 μL of 0.4% (v/v) trypan blue solution (Sigma). The hemacytometer was loaded and examined immediately under the microscope at low resolution, and total number of viable cells determined. For growth inhibition assays, 0.75 × 105 cells/mL were plated into flat bottomed 96-well plates (Costar, Corning, NY) and incubated at 37 °C, 99% humidity and 5% CO2 (v/v) for 24 h. After this period, the samples of BSA, PLGA and PLGA/BSA/Leu were added (concentrations between 10 and 100 nM), and the cells were incubated for an additional 48 h. Subsequently a reaction mix of medium, MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt), and PMS (phenazine methosulfate) in a ratio of 100:19:1 was added to each well and further incubated for 45 min. During this period, MTS is bioreduced into formazan, by dehydrogenase enzymes present in

2.6. Bulk density Bulk density of the microparticles was determined in a helium pycnometer (Micromeritics AccuPyc 1330). Once inside the calibrated cell of the pycnometer the powder is maintained 24 h under vacuum to totally dry and degas the sample. Density is given by measuring the pressure change of helium in the calibrated volume. loaded with the powder samples. 2.7. X-ray diffraction XRD powder diffraction was performed in a RIGAKU X-ray diffractometer (model Miniflex II) with automatic data acquisition (Peak search for Windows v. 6.0 Rigaku) using CuKα radiation (λ = 0.15406 nm) and working at 30 kV/15 mA. Diffraction patterns were collected in the range 2θ = 2–55° with a 0.02° step size and an acquisition time of 1 min/step. 2.8. Differential scanning calorimetry The differential scanning calorimetry (DSC) patterns were obtained by treating the samples in a differential scanning calorimeter 237

The Journal of Supercritical Fluids 128 (2017) 235–243

M. Tavares et al.

Table 2 Properties of PLGA microparticles with encapsulated BSA and leucine.

PLGA/BSA PLGA/Leu PLGA/BSA/Leu

Dn,50 (μm)

Dv,50 (μm)

Span

Da (μm)

3.0 ± 0.2 2.6 ± 0.2 2.6 ± 1.2

9.6 ± 0.2 10.0 ± 0.1 10.6 ± 0.1

0.93 ± 0.03 0.92 ± 0.02 0.75 ± 0.04

4.2 4.4 4.7

Dn,50–Particle number diameter of 50% cumulative distribution; Dv,50- Particle volumetric diameter of 50% cumulative distribution; Span determined according to Eq. (2); DaAerodynamic diameter.

metabolically active cells, which in turn is susceptible of being measured at 490 nm absorbance by an Infinite F200 Microplate Reader (Tecan, Männedorf, Switzerland), directly from the 96-well assay plate, so that the quantity of formazan product measured is directly proportional to the number of metabolic active cells in culture. The cell viability results for each concentration assayed were normalized relatively to the control samples. The assays were performed in triplicate.

Fig. 2. Design of the BSA/leucine/PLGA assays. Table 1 ANOVA testing the effects of the parameters on MMAD, FPF and Dv,50 for a significance level of 5%. p-values marked with * mean p < 0.05; ** p < 0.01;***p < 0.001.

3. Results and discussion

p-value

BSA Leucine Ratio

MMAD

FPF

Dv,50

0.0753 0.0045** 0.0118*

0.1817 0.0559 0.0004***

0.0512 0.0067** 0.6691

To observe and confirm the effects of the addition of BSA and leucine, as well as the effect of CO2/solution ratio variation, a design of experiment was performed and the results tested for statistical significance. The variables studied were the MMAD, FPF and Dv50. In order to do a complete study, several assays were performed, including a central point (Fig. 2). With the analysis of Table 1 it is possible to assume that for a significance level lower than 5%, FPF is affected by the addition of leucine and Dv,50 by the change of CO2/liquid solution ratio, while MMAD is

MMAD − Mass median aerodynamic diameter of 50% cumulative distribution; FPF- Fine particle fraction; Dv,50- Particle volumetric diameter of 50% cumulative distribution.

Fig. 3. Effect of BSA, leucine and ratio change in the particles MMAD, FPF and Dv,50. In the 1st row from left to right: MMAD vs BSA; FPF vs BSA and Dv,50vs BSA. [At the x-axis: -1 represents absence of BSA (0% w/v), 1 the presence of BSA (0.09% w/v), 0 the central point (0.045%)]. 2nd row: MMAD vs Leu; FPF vs Leu and Dv,50 vs Leu. [x-axis: -1 represents absence of leu (0% w/v), 1 the presence of Leu (0.02% w/v), 0 the central point (0.01% w/v)]. 3rd row: MMAD vs Flow ratio, FPF vs Flow ratio; Dv,50 vs Flow ratio. [x-axis: -1 is the ratio of 8.3, 1 is the 5 ratio and 0 is the central point (6.7 ratio)].

238

The Journal of Supercritical Fluids 128 (2017) 235–243

M. Tavares et al.

Fig. 4. SEM images of PLGA microparticles (a 1500×, b 10,000× magnification) and the BSA/ leucine-loaded PLGA microparticles (c 5000x, d 10000x).

affected by both variables. Fig. 3 shows how the dependent variable is affected by the independent ones, and the graphics support the information given by the analysis of the p-value. By observation of the graphic concerning MMAD, the lowest obtained value requires the presence of leucine, and this variable also showed the lowest p-value. Also, there is a slight decrease in the value when a higher ratio is used, but with less significance. Meanwhile, the FPF is only affected by the increase in CO2/liquid ratio, and that is confirmed by the high and precise FPF values obtained. This variation is the highest observed, showing the high significance of this variable. At last, the Dv,50 is only affected by the addition of leucine, which is supported by the graphics where it is possible to see that in the other cases there is no significant

Fig. 5. X-Ray Diffraction. (a) unprocessed BSA; (b) BSA microparticles.

Fig. 6. X-Ray Diffraction. (a) PLGA microparticles; (b) BSA microparticles; (c) Leucine; (d) BSA-PLGA microparticles; (e) Leucine-PLGA microparticles; (f) BSA-Leu- PLGA microparticles.

239

The Journal of Supercritical Fluids 128 (2017) 235–243

M. Tavares et al.

Fig. 7. DSC of (a) PLGA microparticles; (b) raw BSA; (c) BSA-loaded PLGA microparticles; (d) Leucine-PLGA microparticles; (e) BSA/leucine-loaded PLGA microparticles.

Fig. 8. FTIR. (a) PLGA microparticles; (b) BSA; (c) BSA-PLGA microparticles; (d) BSA-Leu-PLGA- microparticles.

variation in the diameter values. With this study, it was proven that statistically, the addition of leucine as a dispersibility enhancer and the use of a higher CO2/liquid solution improve the particles characteristics and for that reason they were carried in the following studies. Even though PLGA has been used as a carrier for pulmonary delivery in several research papers, there is no literature of it being micronized by SASD. Therefore, the first step of the process was to optimize the PLGA microparticles until the best operating conditions were found. Considering the design of experiment study performed, the assays were carried using the parameters that led to best aerodynamic properties: using both the BSA and the dispersibility enhancer − leucine − and using a volumetric flow of CO2/liquid solution ratio of 8.3.

Table 3 Representation of the aerodynamic properties determined by ACI and DUSA for the BSAloaded PLGA microparticles.

PLGA/BSA PLGA/Leu PLGA/BSA/Leu

MMAD (μm)

FPF (%)

GSD

ED (%)

3.5 ± 0.9 1.7 ± 0.1 2.9 ± 0.8

25.4 ± 2.3 37.6 ± 2.4 43.4 ± 3.7

2.50 ± 0.04 2.3 ± 0.3 2.4 ± 0.2

95.3 ± 3.8 98.8 ± 1.7 99.6 ± 0.1

MMAD − Mass Median Aerodynamic Diameter; FPF- Fine particle fraction; GSD − Geometric Standard Deviation; ED − Emitted Dose.

240

The Journal of Supercritical Fluids 128 (2017) 235–243

M. Tavares et al.

Ideally, particles should have a high diameter associated with a low density, in a way that the macrophages would avoid them and still they would have the flowability that would allow them to reach the deep lung [41,42] . Slight improvements in the size and morphology of the particles were also noticed when comparing SEM images (Fig. 4) with and without encapsulated BSA and leucine (c and d) that show a more spherical form and particles with diameters smaller than 1 μm. The solid-state of all the produced particles was also studied. Both raw and processed BSA were represented in Fig. 5, and the two curves have two peaks at about 7° and 20°, which indicates a presence of a semi-crystalline compound. In Fig. 6 the XRD curves of all the produced formulations are presented and it is possible to understand how the encapsulation of both leucine and BSA affect the particles solid-state. When BSA is encapsulated (Fig. 6d) the peak at 7° is no longer noticed, which could indicate that all the BSA is within the polymer. When leucine is added (Fig. 6c) the pattern gains two sharp peaks, which are from leucine by observation of curves. The curve (c) represents the XRD for raw leucine, which shows a crystalline form, and that is confirmed by the appearance of its peaks (6.1°, 24.4° and 30.6°). However, when it is encapsulated in PLGA a new peak at about 20° appears which might indicate a structural change, while the peak at about 5° remains in the same position. When leucine is present in a small fraction, it tends to lose its crystalline form, which explains the fade of its peaks especially when BSA is encapsulated as well, because it has a very low concentration in this formulation. Nonetheless, the fact that the peaks are still noticed indicates that leucine does not lose entirely its crystalline form. The DSC curves give complementary information of the XRD as can be seen in Fig. 7. In all the DSC curves for PLGA, there is a peak near 50 °C corresponding to the Tg of the polymer. Curve (b) presents a broad peak around 70 °C, which is attributed to the melting point of raw BSA. Around this temperature, it is also expected to have a peak related to the reversible unfolding of protein chains. Also, there is a sharper peak at 225 °C due to the full denaturation of the protein. However, by analysing curve (c), this peak along with the broad one disappear suggesting the stability enhancement of the protein when entrapped in PLGA. Also, it might indicate the protein’s dispersibility in the polymer’s matrix. The peak related to the polymer’s Tg suffers a slight shift to the right (from 51.8 °C to 53.7 °C) which might be due to the overlap with the BSA’s broad endothermic peak. When leucine enters the formulation (d;e) a new peak appears at about 218 °C that is correspondent to the leucine’s decomposition. Unlike BSA, this peak appears when leucine is micronized with PLGA and with both PLGA/ BSA, which indicates the presence of a crystalline form, supporting the results obtained by XRD. The peak correspondent to PLGA suffers once again a shift (54.8 °C in (d) and 55.5 °C in (e) indicating once again one possible interaction with leucine that leads to an increase in the Tg. By the observation of the FTIR spectra in Fig. 8 it is possible to notice that when the powder is constituted by PLGA and BSA there is an appearance of two peaks at 1655 and 1543 cm−1, which correspond to the amide I (C]O stretching) and amide II (NeH bending) respectively. The PLGA peak correspondent to the OH suffered a shift from 3528 to 3302 cm−1. The rest of the BSA peaks are masked by the polymer’s peaks, which is understandable since the ratio of BSA and PLGA is near 1:10. When leucine is added to the PLGA formulation the FTIR spectra (Fig. 8d) presents two new peaks observed at 1586 and 1522 cm−1, which are associated to stretching vibrations of CO2. These peaks suffered a minor shift to the left when compared to the leucine spectrum probably due to some conformational change. Also, the peaks at 3000 and 2955 cm−1 show a larger band which may happen by the influence of leucine’s peaks at 2957 cm−1 which is due to asymmetric stretch of CH2 group. The same happens when both BSA and leucine is added, and that can be confirmed by comparing it with the PLGA-BSA spectrum.

Fig. 9. BSA release profiles from dry powder formulations with and without the presence of leucine. Experimental data was correlated using Peppas & Sahlin equation [49]. Table 4 Kinetic values obtained from the Peppas & Sahlin equation for the BSA-loaded and BSA/ Leu-loaded PLGA microparticles. Peppas & Sahlin

BSA/Leu BSA

R2

kd

kr

m

0.9792 0.9683

0.0165 0.0212

−0.0017 -7E-6

0.3389 0.4742

Fig. 10. Viability of A549 cells after 48 h of exposure to 100 nM BSA, PLGA and BSA/leuloaded PLGA microparticles. Results are expressed as mean ± SEM to controls from at least three independent experiments.

The temperature of the static mixer was set to 40 °C above the critical temperature of CO2 (31.1 °C). In terms of the drying temperature in the precipitator, considering that we are using water and acetonitrile (boiling temperatures 100 °C and 82 °C, respectively) and that the Tg of PLGA is around 45–50 °C, it was found that the best temperature was around 70 °C, with compressed air entering at nearly 100 °C, to obtain completely dry powders while preventing the formation of melted particles. As for the pressure, throughout the assays three different pressures were tested. A pressure of 9 MPa was used since it showed a slight improvement in the particle characteristics, such as their size and morphology. By analysing the diameters obtained in Table 2, and comparing the geometric (Dv,50) and numeric diameters (Dn,50), there is a substantial difference justified by the existence of a few, yet large, agglomerates. As for the Dv,50 there is no significant variance between formulations, and the value rounds the 10 μm, which indicates that the presence of agglomerates in the powder affects all formulations. In order to calculate the aerodynamic diameter (da), the bulk density has to be taken into account according to Eq. (1). The particles show a low density (0.19 ± 0.01 g/mL), which is a major advantage for its aerodynamic properties, caused by the CO2 expansion and consequent droplet burst, affecting directly the da, which is in the suitable range for inhalation.

241

The Journal of Supercritical Fluids 128 (2017) 235–243

M. Tavares et al.

dispersibility enhancer those problems can be mildly surpassed. It is also possible to conclude that the best obtained formulation for pulmonary vaccine delivery was PLGA/BSA/Leu, with a MMAD of 2.9 μm and a FPF of 43.4%, which present fully proper characteristics for inhalation.

Since both BSA and leucine present peaks between 1490 and 1650 cm−1 it is possible to see an increase of peaks in that area making it harder to identify them. As for the PLGA peak at 3528 cm−1, it disappears when leucine is added, which might mean that there is some interaction between these molecules. By testing the deposition of different formulations in the various stages of the Andersen Cascade Impactor it is possible to observe that they present different aerodynamic properties from one another. Table 3 shows the mass median aerodynamic diameter (MMAD) of the particles and a slight difference is noticed when in comparison with the da, probably meaning that there is some disaggregation while the particles are inhaled, leading to a smaller MMAD. Both this value and the fine particle fraction (FPF) tend to improve when leucine is present, which was already expected [43,44], but the highest percentage of FPF is obtained when both leucine and BSA are present. Even though this value of FPF is above the range obtained for other PLGA microparticles produced using spray drying or double-emulsion method [45,46], it is highly related to the presence of the aggregates [47,48]. It is known that DPIs of the first generation commonly exhibited a relatively low efficiency in terms of the FPF. The efficiency of the microparticles entrainment resides on the ability of the DPI to generate forces sufficient to deagglomerate the powder and create a high flow, and that becomes hampered when there is such a large amount of aggregates [29,46]. For that reason, and even though the used DPI might not be the most appropriate, the high existence of aggregates is the factor that mainly influences the aerodynamic results because the diminishing of agglomerates was accompanied with the improvement of both FPF and MMAD. Besides this, the powders fit in the proper range and are suitable for inhalation [20] . The BSA release experiments were performed to examine the mechanism of the BSA-loaded microparticles controlled release. By modelling the drug release using mathematical equations, important information regarding mass transport and chemical processes relatively to the drug delivery system can be acquired, coupled with the effect of the parameters, for instance particle geometry or drug loading [50]. The release curves were fitted using the Peppas and Shalin equation [51], as seen in Fig. 9, and the kinetic values obtained are in Table 4. Both diffusion exponents (n) lead to a Fick’s diffusional release, even though this value is characterized for n < 0.43, and the BSA release without leucine has n = 0.47, it is safe to admit that diffusion is the main present factor [47]. Although both formulations have different release rates, since the presence of leucine leads to a faster initial release, in both cases full BSA release was achieved within 60 h. To confirm that the microparticles were not cytotoxic, an in vitro viability assay was performed using lung adenocarcinoma cells as a model. By the analysis of the results represented in Fig. 10 it is clear that both BSA and PLGA are not toxic, once the percentages are all very close to the 100%, and that was already expected since they are known for their biocompatibility [23,52]. Besides the produced microparticles (BSA/Leu/PLGA) show 100% cell viability, confirming that the SASD technique does not affect the components toxicity.

Acknowledgements This work was supported by the Associate Laboratory for Green Chemistry LAQV-REQUIMTE which is financed by national funds from FCT/MCTES (UID/QUI/50006/2013) and co-financed by the ERDF under the PT2020 Partnership Agreement (POCI-01-0145FEDER−007265) and by the Unidade de Ciências Biomoleculares Aplicadas-UCIBIO-REQUIMTE which is financed by national funds from FCT/MCTES (UID/Multi/04378/2013) and co-financed by the ERDF under the PT2020 Partnership Agreement (POCI-01-0145-FEDER007728). T.C. also acknowledges FCT-Lisbon for the IF/00915/2014 contract. The authors also acknowledge N. Costa (Laboratório de Análises REQUIMTE – LAQV/UCIBIO) for technical support References [1] T. Takami, Y. Murakami, Development of PEG–PLA/PLGA microparticles for pulmonary drug delivery prepared by a novel emulsification technique assisted with amphiphilic block copolymers, Colloids Surf. B. Biointerfaces 87 (2011) 433–438. [2] V. Levet, R. Rosière, R. Merlos, L. Fusaro, G. Berger, K. Amighi, N. Wauthoz, Development of controlled-release cisplatin dry powders for inhalation against lung cancers, Int. J. Pharm. 515 (2016) 209–220. [3] I.M. El-Sherbiny, N.M. El-baz, M.H. Yacoub, Inhaled nano- and microparticles for drug delivery, Glob. Cardiol. Sci. Pract. 2 (2015) 1–14. [4] J.C. Sung, B.L. Pulliam, D. Edwards, Nanoparticles for drug delivery to the lungs, Trends Biotechnol. 25 (2017) 563–570. [5] G. Pilcer, K. Amighi, Formulation strategy and use of excipients in pulmonary drug delivery, Int. J. Pharm. 392 (2010) 1–19. [6] C.L. Pastoriza, J. Todoroff, R. Vanbever, Delivery strategies for sustained drug release in the lungs, Adv. Drug Deliv. Rev. 75 (2014) 81–91. [7] M.B. Broichsitter, C. Schweiger, T. Schmehl, T. Gessler, W. Seeger, T. Kissel, Characterization of novel spray-dried polymeric particles for controlled pulmonary drug delivery, J. Control. Release 158 (2012) 329–335. [8] T. Parumasivam, A.S. Ashhurst, G. Nagalingam, W.J. Britton, H.K. Chan, Inhalation of respirable crystalline rifapentine particles induces pulmonary inflammation, Mol. Pharm. 14 (2017) 328–335. [9] K. Koushik, U.B. Kompella, Preparation of large porous deslorelin-PLGA microparticles with reduced residual solvent and cellular uptake using a supercritical carbon dioxide process, Pharm. Res. 21 (2004) 524–535. [10] A. Srivastava, T. Yadav, S. Sharma, A. Nayak, A.A. Kumari, N. Mishra, Polymers in drug delivery, J. Biosci. Med. 4 (2016) 69–84. [11] W.C. Koff, D.R. Burton, P.R. Johnson, B.D. Walker, C.R. King, G.J. Nabel, R. Ahmed, M.K. Bhan, S.A. Plotkin, Accelerating next-generation vaccine development for global disease prevention, Science 340 (2013) 1232911–1232917. [12] J.P. Amorij, V. Saluja, A.H. Petersen, W.L.J. Hinrichs, A. Huckriede, H.W. Frijlink, Pulmonary delivery of an inulin-stabilized influenza subunit vaccine prepared by spray-freeze drying induces systemic, mucosal humoral as well as cell-mediated immune responses in BALB/c mice, Vaccine 25 (2007) 8707–8717. [13] T. Sou, E.N. Meeusen, M. de Veer, D.A.V. Morton, L.M. Kaminskas, M.P. McIntosh, New developments in dry powder pulmonary vaccine delivery, Trends Biotechnol. 29 (2011) 191–198. [14] B. Pulliam, J.C. Sung, D. Edwards, D esign of nanoparticle-based dry powder pulmonary vaccines, Expert Opin, Drug Deliv. 4 (2007) 651–663. [15] H.O. Alpar, S. Somavarapu, K.N. Atuah, V.W. Bramwell, Biodegradable mucoadhesive particulates for nasal and pulmonary antigen and DNA delivery, Adv. Drug Deliv. Rev. 57 (2005) 411–430. [16] S.A. Shoyele, A. Slowey, Prospects of formulating proteins/peptides as aerosols for pulmonary drug delivery, Int. J. Pharm. 314 (2006) 1–8. [17] A.S. Silva, M.T. Tavares, A. Aguiar-Ricardo, Sustainable strategies for nano-in-micro particle engineering for pulmonary delivery, J. Nanopart. Res. 16 (2014) 2602. [18] H.S. Choi, Y. Ashitate, J.H. Lee, S.H. Kim, A. Matsui, N. Insin, M.G. Bawendi, M.S. Behnke, J.V. Frangioni, A. Tsuda, Rapid translocation of nanoparticles from the lung airspaces to the body, Nat. Biotechnol. 28 (2010) 1300–1304. [19] J. Hu, Y. Dong, G. Pastorin, W.K. Ng, R.B.H. Tan, Spherical agglomerates of pure drug nanoparticles for improved pulmonary delivery in dry powder inhalers, J. Nanopart. Res. 15 (2013) 1560. [20] P. Wanakule, G.W. Liu, A.T. Fleury, K. Roy, Nano-inside-micro: disease-responsive microgels with encapsulated nanoparticles for intracellular drug delivery to the deep lung, J. Control. Release 162 (2012) 429–437. [21] H.K. Makadia, S.J. Siegel, Poly lactic-co-Glycolic acid (PLGA) as biodegradable controlled drug delivery carrier, Polymer 3 (2011) 1377–1397. [22] L.J. Cruz, P.J. Tacken, R. Fokkink, B. Joosten, M.C. Stuart, F. Albericio, R. Torensma, C.G. Figdor, Targeted PLGA nano- but not microparticles specifically

4. Conclusions The development of dry powders for inhalation stands as an outbreak solution to overcome drawbacks associated with drug delivery while the use of sustainable techniques represents a production substitute to surpass the environment issues that affect the industry nowadays. For that, this work represents a research field that has been increasingly growing and taken into consideration worldwide. Herein, it was proven that the SASD apparatus is capable of producing PLGA dry powder formulations suitable for inhalation. Throughout the obtained results it is possible to conclude that the particles manufacturing process has associated problems, since it is hard to establish a meeting point between ideal powder properties and SASD parameters. However, with the addition of an API or/and a 242

The Journal of Supercritical Fluids 128 (2017) 235–243

M. Tavares et al.

[23]

[24] [25] [26]

[27] [28]

[29] [30] [31]

[32] [33]

[34]

[35] [36]

[37]

[38] M.C. Silva, A.S. Silva, J.F. Lodeiro, T. Casimiro, C. Lodeiro, A. Aguiar-Ricardo, Supercritical CO2-assisted spray drying of strawberry-like gold-coated magnetite nanocomposites in chitosan powders for inhalation, Materials 10 (2017) 74. [39] Preparations for inhalation: aerodynamic assessment of fine particles. european pharmacopeia 7.0, European Pharmacopeia 7.0 (2010) 274–285 (Counc. Eur. 2.9.18). [40] C.R. Rodrigues, P.V. Baptista, F. Marques, D.P. Fernández, J.G. Varela, L. Sánchez, D.V. García, M.L. Torres, A. Fernándeza, J.J. Fernández, Heteroleptic mononuclear compounds of ruthenium(II): synthesis, structural analyses, in vitro antitumor activity and in vivo toxicity on zebrafish embryos, Dalton Trans. 45 (2016) 19127–19140. [41] D. Edwards, J. Hanes, G. Caponetti, J. Hrkach, A.B. Jebria, M.L. Eskew, J. Mintzes, D. Deaver, N. Lotan, R. Langer, Large porous particles for pulmonary drug delivery, Science 276 (1997) 1868–1871. [42] Z. Liang, R. Ni, J. Zhou, S. Mao, Recent advances in controlled pulmonary drug delivery, Drug Discov. Today 20 (2015) 380–389. [43] P.H. Moghaddam, V. Ramezani, E. Esfandi, A. Vatanara, M.N. Meibodi, M. Darabi, K. Gilani, R. Najafabadi, Development of a nano–micro carrier system for sustained pulmonary delivery of clarithromycin, Powder Technol. 239 (2013) 478–483. [44] R.S. Kaye, T.O.L.S. Purewal, H.O.Y.A. Alpar, Drug delivery simultaneously manufactured nano-in-micro (SIMANIM) particles for dry-powder modified-release delivery of antibodies, J. Pharm. Sci. 98 (2009) 4055–4068. [45] Y. Yang, N. Bajaj, P. Xu, K. Ohn, M.D. Tsifansky, Y. Yeo, Development of highly porous large PLGA microparticles for pulmonary drug delivery, Biomaterials 30 (2009) 1947–1953. [46] N. Sivadas, D. O’Rourke, A. Tobin, V. Buckley, Z. Ramtoola, J.G. Kelly, A.J. Hickey, S.A. Cryan, A comparative study of a range of polymeric microspheres as potential carriers for the inhalation of proteins, Int. J. Pharm. 358 (2008) 159–167. [47] P. Rogueda, D. Traini, The nanoscale in pulmonary delivery. Part 2: formulation platform, ExpertOpin. Drug Deliv. 4 (2007) 607–620. [48] V.A. Philip, R.C. Mehta, M.K. Mazumder, P.P. Deluca, Effect of surface treatment on the respirable fractions of PLGA microspheres formulated for dry powder inhalers, Int. J. Pharm. 151 (1997) 165–174. [49] D.Y. Arifin, L.Y. Lee, C.H. Wang, Mathematical modeling and simulation of drug release from microspheres: implications to drug delivery systems, Adv. Drug Deliv. Rev. 58 (2006) 1274–1325. [50] W. Lafayette, A simple equation for the description of solute release. III. Coupling of diffusion and relaxation, Int. J. Pharm. 57 (1989) 169–172. [51] P.L. Ritger, N.A. Peppas, A simple equation for description of solute release I. Fickian and non-fickian release from non-swellable devices in the form of slabs spheres, cylinders or discs, J. Control. Release 5 (1987) 23–36. [52] W. Zou, C. Liu, Z. Chen, N. Zhang, Studies on bioadhesive PLGA nanoparticles: a promising gene delivery system for efficient gene therapy to lung cancer, Int. J. Pharm. 370 (2009) 187–195.

deliver antigen to human dendritic cells via DC-SIGN in vitro, J. Control. Release 144 (2010) 118–126. F.S.T. Mirakabad, K.N. Koshki, A. Akbarzadeh, M.R. Yamchi, M. Milani, N. Zarghami, V. Zeighamian, A. Rahimzadeh, S. Alimohammadi, Y. Hanifehpour, S.W. Joo, PLGA-Based nanoparticles as cancer drug delivery systems, asian pac, J. Cancer Prev. 15 (2014) 517–535. N. Ibrišimović, H.A. Dubai, V. Kerleta, Quantification of biodegradable PLGA nanoparticles for drug targeting, Acta Med. Acad. 39 (2010) 118–131. A.G. Ding, S.P. Schwendeman, Determination of water-soluble acid distribution in poly(lactide-co-glycolide), J. Pharm. Sci. 93 (2004) 322–331. T. Barroso, M. Temtem, T. Casimiro, A. Aguiar-Ricardo, Development of pH-responsive poly(methylmethacrylate-co-methacrylic acid) membranes using scCO2 technology. Application to protein permeation, J. Supercrit Fluids 51 (2009) 57–66. E.J. Beckman, Supercritical and near-critical CO2 in green chemical synthesis and processing, J. Supercrit. Fluids 28 (2004) 121–191. A. Aguiar-Ricardo, V.D.B. Bonifacio, T. Casimiro, V.G. Correia, Supercritical carbon dioxide design strategies: from drug carriers to soft killers, Phil. Trans. R. Soc. A 373 (2015) 20150009. A.H.L. Chow, H.H.Y. Tong, P. Chattopadhyay, B.Y. Shekunov, Particle engineering for pulmonary drug delivery, Pharm. Res. 24 (2007) 411–437. G.D. Porta, E. Reverchon, Engineering powder properties by supercritical fluid for optimum drug delivery, BIOP Int. (2005) 54–60. L. Casettari, E. Castagnino, S. Stolnik, A. Lewis, S.M. Howdle, L. Illum, Surface characterisation of bioadhesive PLGA/chitosan microparticles produced by supercritical fluid technology, Pharm. Res. 28 (2011) 1668–1682. L. Chen, T. Okuda, X.Y. Lu, H.K. Chan, Amorphous powders for inhalation drug delivery, Adv. Drug Deliv. Rev. 100 (2016) 102–115. I. Alfagih, N. Kunda, F. Alanazi, S.R. Dennison, S. Somavarapu, G.A. Hutcheon, I.Y. Saleem, Pulmonary delivery of proteins using nanocomposite microcarriers, J. Pharm. Sci. 104 (2015) 4386–4398. R.P. Cabral, A.M.L. Sousa, A.S. Silva, A.I. Paninho, M. Temtem, E. Costa, T. Casimiro, A. Aguiar-Ricardo, Design of experiments approach on the preparation of dry inhaler chitosan composite formulations by supercritical CO2-assisted spraydrying, J. Supercrit. Fluids 116 (2016) 25–35. A. Aguiar-Ricardo, Building dry powder formulations using supercritical CO2 spray drying, Curr. Opin. Green Sust. Chem. 5 (2017) 12–16. R.B. Restani, A.S. Silva, R.F. Pires, R. Cabral, I.J. Correia, T. Casimiro, V.D.B. Bonifácio, A. Aguiar-Ricardo, Nano-in-Micro POxylated polyurea dendrimers and chitosan dry powder formulations for pulmonary delivery, Part. Part. Syst. Charact 33 (2016) 851–858. A.S. Silva, A.M. Sousa, R.P. Cabral, M.C. Silva, C. Costa, S.P. Miguel, V.D.B. Bonifácio, T. Casimiro, I.J. Correia, A. Aguiar-Ricardo, Aerosolizable gold nano-in-micro dry powder formulations for theragnosis and lung delivery, Int. J. Pharm 519 (2017) 240–249.

243