Aerosol-Based Efficient Delivery of Clarithromycin,a Macrolide Antimicrobial Agent,to Lung Epithelial Lining Fluid and Alveolar Macrophages for Treatment of Respiratory Infections

Materials and animals

CAM was purchased from Wako Pure Chemicals CO., Ltd. (Osaka, Japan). All other regents were commercially available and of analytical grade. Male SD rats (200–230 g) were purchased from Japan SLC (Shizuoka, Japan). The animal experimental plan used was been approved by the Committee of Laboratory Animal Center (No. 09-009), and conforms to the Guiding Principles for the Care and Use of Experimental Animals in Hokkaido Pharmaceutical University.

Animal experiments and pharmacokinetic analysis

CAM dissolved in phosphate-buffered saline solution (pH 7.4) was aerosolized into rat lungs at a dose of 0.2 mg/0.25 mL/kg using a Liquid MicroSprayer ^® (Model IA-1C, PennCentury, Inc., Philadelphia, PA, USA) under pentobarbital anesthesia. The delivery efficiency to lung surface of a Liquid MicroSprayer ^® was more than 97%. At each designated time point (0.25, 0.5, 1, 2, 4, 8, and 24 h), blood was collected from the jugular vein. In order to collect the bronchoalveolar lavage fluid, the trachea was immediately cannulated and the lungs were lavaged three times with 5-mL ice-cold phosphate-buffered saline solution (pH 7.4).⁽²⁹⁾ The bronchoalveolar lavage fluid was immediately centrifuged at 4°C (650×g, 5 min) to separate AMs from the ELF. AMs were extracted with 1 mL 0.1 M NaOH for HPLC analysis.

In order to calculate the concentrations of CAM in ELF, the apparent volume of ELF was estimated using urea as an endogenous marker of ELF dilution.⁽³⁰⁾ The mean ELF value estimated in the present study was 393 μL/215 g rat. In order to calculate the concentrations of CAM in AMs, the intracellular volume in AMs was determined by a velocity-gradient centrifugation technique using ³H-water,⁽³¹⁾ and this was estimated to have a mean value of 4.2 μL/mg cell protein. The concentration of CAM in each sample was measured by high-performance liquid chromatography (HPLC) as described below. The protein concentration in the AMs extracts was determined using Coomassie protein Assay reagent (Pierce Chemical Company, Rockford, IL, USA) with bovine serum albumin as a standard.⁽³²⁾

For the pharmacokinetic analysis, the area under the curve—time curve from time 0 to time 24 h (AUC) was calculated by the trapezoidal rule. Also, bioavailability (BA) in ELF and AMs was calculated from Equation (1)⁽³³⁾ using the data following the intravenous injection a dose of 50 mg/mL/kg in this study. \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} {\rm BA} = \frac {{\rm D}_{\rm iv \ injection} \times {\rm AUC}_{\rm formulation}} {{\rm D}_{\rm formulation} \times { \rm AUC}_{\rm iv \ injection}}\tag {1} \end{align*} \end{document}

where D_formulation is the dose of formulation, D_{iv injection} is the dose of intravenous injection, AUC_formulation is the AUC following administration of formulation, and AUC_{iv injection} is the AUC following intravenous injection. D_{iv injection} was 50 mg/kg wt, AUC_{iv injection} in ELF and AMs were 350 μg*h/mL and 15,441 μg*h/mL, respectively.

Pharmacokinetics/pharmacodynamics (PK/PD) analysis

The antibacterial effects of CAM in ELF and AMs following administration of the formulation were estimated by PK/PD analysis. The ratio of AUC/minimum inhibitory concentration at which 90% of isolates (MIC₉₀) (AUC/MIC₉₀) was calculated as the PK/PD parameters of an antibacterial effect. The MIC₉₀ values against microorganisms infected in ELF and AMs were taken from the literature.^(34–38) The estimated effective values of AUC/MIC₉₀ are larger than 100 h.⁽³⁹⁾ Also, the required dose for effective therapy (RD, a dose required for winning 100 h that is effective AUC/MIC₉₀ value) was calculated from Equation (2). \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} { \rm RD } = \frac { { \rm D } _ { \rm formulation } } { ( { \rm AUC / MIC } _ { 90 } ) / 100 } \tag { 2 } \end{align*} \end{document}

Determination of CAM by HPLC

The concentrations of CAM in samples were measured by HPLC with electrochemical detection. The samples (50 μL) and added roxithromycin solution (as an internal standard dissolved in acetonitrile, 50 μL) were mixed, and 50 μL aliquots were subjected to HPLC using a system involving a Mightysil RP-18GP column (4.6×150 mm, 5-μm pore size; Kanto Chemical Co., Tokyo, Japan). The mobile phase was 50 mM phosphate buffer (pH 6.8)/acetonitrile/methanol (5:4:1). The separation was performed at a flow rate of 1.0 mL/min at 40°C and the eluate from the column was monitored using an electrochemical detector (Coulochem II, ESA, Bedford, MA, USA). A dual analytical cell was used with the upstream potential set at + 650 mV and the downstream potential at + 800 mV.

Pharmacokinetics of CAM

The time-courses of the concentrations of CAM in ELF, AMs, and plasma after administration of the aerosol and oral formulations to rats are shown in Figure 1. The pharmacokinetic parameters are summarized in Table 1. The concentrations of CAM in ELF and AMs following administration of the aerosol formulation were lower than those of the oral formulation. However, the BA of the aerosol formulation in ELF and AMs were 42 and 28, respectively. These values were greater than those of the oral formulation. These findings suggest that the aerosol formulation would have pharmacokinetic efficacy of 42- and 28-fold that of the intravenous injection. Also, the concentrations of CAM in plasma following administration of the aerosol formulation were remarkably lower than those following administration of the oral formulation. This indicates that, in comparison with oral formulation, the aerosol formulation can avoid distribution of macrolides to blood.

OPEN IN VIEWER

FIG. 1.

Time courses of concentrations of CAM in ELF (A), AMs (B), and plasma (C) after administration of the aerosol (○) and oral (€) formulation to rats. The aerosol formulation (0.2 mg/kg) was administered to rat lungs using a Liquid MicroSprayer ^® . At each time point after administration, the plasma, ELF, and AMs were collected, and then concentrations of CAM in each sample were determined. Each point represents the mean±SD (n=4). Insert of C: the enlargement of time course of CAM in plasma after administration of aerosol formulation. Time courses of CAM after administration of the oral formulation (50 mg/kg) were taken from the submitted literature.⁽⁴⁵⁾

Macrolides given by the oral route distributes in the alveolus through vascular endothelial cells and alveolar epithelial cells from the blood side. The alveolar barrier consists of three layers, the capillary lumen, connected tissue, and alveolar epithelial cells.⁽⁴⁰⁾ The alveolar epithelial cells that are tightly connected by numerous zonulae occludins is considered to provide a significant barrier between the plasma and ELF.⁽⁴¹⁾ There are several efflux transporters expressed in alveolar epithelial type I cells in the human and rat lung, including P-glycoprotein and breast cancer-resistant protein.^(42–44) CAM is an MDR1/P-glycoprotein substrate and, thus, CAM can cross from the alveolar epithelium to ELF via an MDR1 transporter.⁽⁴⁵⁾ However, transport of CAM to ELF across the alveolar barrier after oral administration may not be as efficient. In contrast, aerosolized CAM was efficiently delivered to ELF and AMs because it was sprayed into the alveolus directly (Fig. 1 and Table 1). Thus, a reduction in the dose and avoidance of distribution to the blood is possible. The AUC ratios of plasma/ELF of CAM following administration of the aerosol formulation were 0.0014 (Table 1). According to our recent report, this significant asymmetry is based on the fact that transport of CAM from ELF to blood is inhibited by the MDR1 transporter on the alveolar epithelial type I cells.⁽⁴⁵⁾ These findings suggest that aerosolization can avoid systemic side effects as well as provide efficient delivery of macrolides to AMs and ELF.

The AUC of CAM in AMs following administration of the aerosol formulation is 29-fold greater than that in ELF (Table 1). This indicates that CAM is able to concentrate intracellularly in AMs. Recently, we have reported that uptake of CAM by AMs is mediated by active transport systems.⁽⁴⁵⁾ According to previous reports,⁽⁴⁶⁾ CAM may be transported to the intracellular region of AMs via active transport systems which require Ca²⁺ and protein kinase A-dependent phosphorylation, the same as other phagocytes such as human polymorphonuclear neutrophils and J774 murine macrophages, in addition to passive diffusion. In another study, we have also confirmed that CAM distributes to the entire intracellular area of AMs and not the local area (data not shown).

Estimated antibacterial effects of CAM

The estimated antibacterial effects of CAM in ELF and AMs following administration of the aerosol and oral formulation to rats are summarized in Tables 2 and 3. Recently, there has been increasing interest in the relationship between the PK and PD of antimicrobial agents and, therefore, the use of PK/PD parameters is now widespread.^(47,48) It is proposed that the PK/PD analysis of an antibiotic treatment is important for selecting a dose and optimizing the treatment of individual patients. The effects of antimicrobial agents are concentration- and/or time-dependent and the PK/PD parameters used generally are the maximum plasma concentration/MIC₉₀ (C_max/MIC₉₀), AUC/MIC₉₀ and the time above the MIC₉₀. Because the antibacterial effects of CAM depend on the AUC/MIC₉₀,⁽³⁹⁾ the AUC/MIC₉₀ values in ELF and AMs were calculated in the present study. Also, the RD was calculated to obtain information regarding an effective dose. In the case of the aerosol formulation, despite a low dose, the AUC/MIC₉₀ ratios of CAM against most pathogenic microorganisms infected in ELF and AMs were larger than 100 h that is the effective value. Only the AUC/MIC₉₀ against H. influenzae (3.7 h) was lower than the therapeutic value. However, the RD of the aerosol formulation was slightly 5.4 mg/kg (1/93th of the RD of oral formulation). The oral formulation also had an effective AUC/MIC₉₀ value against most pathogens except H. influenzae. Interestingly, the RD of the aerosol formulation against each pathogenic microorganism for effective therapy was markedly lower than those of the oral formulation. The AUC/MIC₉₀ and RD values showed that efficient antibacterial effects of CAM in ELF and AMs are obtained by aerosolization of a dose lower than that used orally. The antibacterial effect of CAM following aerosol formulation to animals used as models of respiratory infection should be investigated in the future because it was not examined in this study.

For sterilization of pathogenic microorganisms infected in ELF and AMs, it is required that CAM is stable in ELF and AMs following administration of aerosol formulation. We have shown that CAM was stable in ELF and AMs for 48 h at 37 ^oC in vitro (data not shown). Although AMs produce and secrete various bioactive substances, such as enzymes, cytokines, complements, proteins, lipids, and reactive oxygen species,^(49–54) these bioactive substances may not affect the stability of CAM. In addition, it is important that the aerosolized CAM do not injure lung tissues. The aerosol formulation of CAM was found to be nontoxic following administration because no release of lactate dehydrogenase from lung tissue was observed (data not shown). This indicates that the aerosolized CAM does not injure lung tissues, at least at the dose used in this study.