Dexamethasone palmitate (DXP, 99%) was provided by Interchain (France), Testosterone decanoate (TD, 98%) for use as internal standard (IS) (Figure 1), was provided by Sigma-Aldrich (France), 1,2-dipalmitoyl-sn-glycero-3-phos- phatidylcholine (DPPC, 99%) by Corden Pharma (Switzerland) and hyaluronic acid, sodium salt 95% (HA) (MW = 1000 kDa) by Acros Organics. HPLC-grade methanol and acetonitrile were purchased from Carlo Erba (Italy). All chemicals used, were of analytical grade. Water was purified using a RIOS/MilliQ system from Millipore (France).

An aqueous solution of HA was prepared dissolving 200 mg of HA into 150 ml of water upon magnetic stirring at room temperature. An ethanolic solution was prepared by dissolving 750 mg of DPPC and 50 mg of DXP into 350 ml of ethanol absolute. Ethanolic and aqueous solutions were then mixed at a ratio of 70/30 (v/v) prior to spray-drying and the mixture maintained under moderate stirring while fed into the spray-dryer. Spray drying was performed using a mini spray-dryer BÜCHI B-290 (France) equipped with a 0.7 mm diameter two-fluid nozzle, which operates in a co-current mode according to conditions detailed in Table 1 [12].

Sprague Dawley male rats with average weight of 300 g were obtained from Harlan Laboratory (Gannat, France). All animal experiments were carried out in accordance with the Principles of Laboratory Animal Care as adopted and propagated by the EU guidelines for Animal Experiments (86/609/EEC and 2010/63/EU) and Legislation in force in France (Decree No. 2013-118 of February 1, 2013) [13]. They were housed 4 per cage while on study in accordance to EEC guidelines. The light/dark cycle was 12 h/12 h. The temperature in the animal room was ambient room temperature of approximately 25˚C and the ambient humidity was in the range of approximately 35% - 60%. Animals were allowed access to food and water ad libitum throughout the duration of the study [14].

Rats were anesthetized with an intraperitoneal injection of a mixture of ketamine (100 mg∙kg⁻¹) and xylazine (10 mg∙kg⁻¹) and then intratracheally administered with 3 mg of DXP powder corresponding to 150 mg of DXP using a Microsprayer (Penn Century, Philadelphia, PA) [15] [16]. Powder was delivered through the insufflation device by rapidly pushing a 2 cm³ bolus of air through

the device. After administration, rats were supported vertically for 1 min. Bronchoalveolar fluid was collected at t = 30 min, 1 h, 2 h, 3 h, 4 h, 6 h, 18 h and 24 h after administration. At least 3 different rats were used for each time point. Bronchoalveolar lavage fluid (BALF) was collected by lavaging lungs three times with 1 mL of PBS per intratracheal lavage [17] [18]. The lungs were massaged and the fluid withdrawn immediately and collected in centrifuge tube. The rats were euthanized by cardiac puncture under deep isoflurane anesthesia before BALF collection. The trachea was cannulated using an 18-gauge needle adaptor for subsequent injection and retrieval of BALF. Then BALF collected containing the lung cells was frozen at −80˚C until further analysis [19].

Acetonitrile stock solutions of DXP (100 μg/mL) and TD (10 μg/mL) were prepared and stored at (−20˚C) until use. The DXP stock solution was diluted with acetonitrile to give a series of standard solutions with concentrations ranging from 1 μg/mL to 80 μg/mL. The calibration samples were prepared by diluting 1 volume of each standard solution with 1 volume of BALF. The final concentrations of calibration samples used to draw the calibration-curve ranged from 0.5 to 40 μg/mL (0.5, 2, 5, 10, 20 and 40 μg/mL). Standards for calibration were stored at −20˚C until the use [20] [21].

Extraction method of DXP in BALF was performed by using a mixture chloroform: methanol (9:1 v:v) [22] [23]. In an amber glass tube, 100 µL of BALF, 100 µL of different concentrations of standard solutions of DXP (0.5 - 40 µg/mL) and 100 µL of IS (2.64 µg/mL) were mixed with a vortex during 30 s. Then 3 mL of a mixture chloroform: methanol (9:1, v:v) were added in the tube. Sample was vortexed for 3 min to obtain protein precipitation then was centrifuged for 10 min at 10,000 rpm using a centrifuge (Jouan GR 412, France). The organic phase in the bottom was transferred into a clean amber vial and evaporated to dryness under a stream of nitrogen at 30˚C. The residue was then reconstituted into 100 µL of acetonitrile and vortexed prior to analysis using HPLC conditions described below [24] [25].

Determination of DXP concentration in BALF was performed by HPLC. A Water^s 2707 autosampler chromatographic system was employed equipped with a Waters^TM 1525 binary HPLC pump, a Waters^TM 2998 photodiode array detector, and a Waters^TM Breeze software. The analysis was performed at 240 nm using a Symmetry Shield^TM RP18 column (5 µm, 250 × 4.6 mm; Waters, France). Column temperature was maintained at 40˚C using a column Heater (Model 1500, Waters Corporation). The mobile phase consisted in a mixture of water: acetonitrile (15:85 (v: v)) and was pumped in isocratic mode for 40 min through the column with a flow rate of 1.2 mL/min. The sample volume injected was of 50 mL [26] [27].

The method was validated according to the criteria developed by the International Conference of Harmonization (ICH) [28]. The parameters evaluated to assess the reliability of the results consist of the linearity of the chromatographic response as a function of analyte concentration from 0.5 to 40 mg/mL, the repeatability of the chromatographic analysis of the reference solution at three (3) levels of concentration: 2, 10 and 40 mg/mL (n = 6), the repeatability of chromatographic analysis of a sample of BALF (n = 6), the repeatability of the procedure (n = 6). The limits of detection (signal-to-noise ratio of 3) and quantification (signal-to-noise ratio of 10) determined were assessed from serial dilutions of the standard solution of DXP [29] [30].

The linearity of the response of DXP was assessed from a scatter plot. The regression line was determined according to the least squares method. An analysis of variance (ANOVA) was performed to test the statistical significance and the overall slope of the regression line. The accuracy was assessed through the coefficient of variation calculated from the ratio of the standard deviation of the mean of each series of measurements. The level of statistical significance was set at 0.05 [31] [32].

The validated method was applied to a series of DXP in BALF samples collected from rats treated with DXP large porous powder and the measured concentrations were fitted using a non-compartment pharmacokinetic model [33] [34].

Chromatographic conditions used provided sufficient resolution with elution of DXP and IS respectively at 27.1 min and 23.9 min (Figure 2). There were no significant (≥10% of LLQ area) interferences at either of these retention times and the selectivity of the method is demonstrated by the representative HPLC chromatograms presented in Figure 2 [27]. A linearity range has been set to the concentration range of 0.5 to 40 µg/mL of DXP with a coefficient of determination R² = 0.999. The line of regression equation was established: Y = (0.262 ± 0.007)X − (0.071 ± 0.022), Y the chromatographic response and X, the concentration in µg/mL [30] [35]. The repeatability tests reported that relative standard deviations were less than 0.1% for the reference solutions and at 3.7% to a DXP in BALF sample for the chromatographic analysis and 2.7% for the whole procedure [36]. The overall recovery of DXP from BALF samples for concentrations 2, 10 and 40 µg/mL was 84.3% ± 1.6%. Consistent recovery values at low, medium and high concentrations indicate the extraction process is acceptable across this range. The high recoveries observed indicate analyte stability under the extraction conditions applied and good extraction. The limits of detection and quantification were, respectively, 0.03 and 0.1 µg/mL [37].

An example of chromatograms obtained from the extract of a DXP and IS (internal standard) in BALF samples from standard solutions and rats treated with DXP powder is given in Figure 2. DXP contents in BALF from rats treated, ranged from 31.5 to 0.1 µg/mL corresponding to DXP levels from 30 min to 24 h after lung administration. The results of DXP obtained in BALF were converted to epithelial lining fluid (ELF) concentrations according to the theoretical volume of ELF reported in the literature (30 µL/kg) and to the volume of PBS used for bronchoalveolar lavage (1 mL). Thus, for a rat of 300 g, the dilution factor determined by calculation was 333 (Table 2) [38] [39]. All the observed concentrations of DXP in BALF and ELF lie within the ranges predicted by the PK model for the dosing regimen. This suggests that the assay is performing well and that the current HPLC method can be used successfully for the measurement of DXP levels in BALF samples Results indicate a greater degree of sensitivity is required to adequately detect DXP levels 18h after administration in lung where the dose is low that the LOQ. Further work would include using a tandem mass spectrometer for greater selectivity and potentially increased detection sensitivity [40].