Drug disposition study on a levofloxacin dry powder formulation using the isolated perfused lung in rats

Abstract

In this study, using an artificially perfused and mechanically ventilated lung as an experimental model, levofloxacin disposition after bolus and pulmonary delivery has been compared. This model was used to investigate the absorption of levofloxacin through the airway into the blood. Krebs-Ringer buffer medium containing bovine serum albumin was perfused into the lungs at a rate of 4 ml.min^− 1. The lungs were ventilated with humidified air heated to 37 °C at 60 breaths per minute. Levofloxacin was administered into the perfusion circuit as a bolus injection and into inhalation air (dry powder). By using HPLC combined with UV detection, the concentration of levofloxacin (in efferent) was determined after lung perfusion. Levofloxacin transported from lung to blood and was detectable within an hour of the experiment, although due to rapid elimination it was no longer detectable 20 min after bolus administration. The normalized AUC_0−inf. after administration of dry powder and solution were 34209.17 ± 6446.1 and 329371.02 ± 94627.7 ng.min.ml^− 1.mg^− 1, respectively. At the end of the experiment, higher levels of levofloxacin were detected in the lungs via pulmonary administration. This ex vivo model resulted in localized delivery of levofloxacin to the lung compartment and measurable transfer into the perfusate.

Introduction

It is well known that the pulmonary circulation serves vital purposes besides gas exchange¹. Compared to other delivery methods such as injection or oral, pulmonary drug delivery offers numerous advantages. To treat respiratory diseases such as cystic fibrosis and asthma, drugs can be administered locally to the lungs. The lung is characterized not only by non-invasive delivery, but also by low proteolytic activity, high solute permeability, and large absorption area^2,3.

The principal cause of pulmonary morbidity in adults with cystic fibrosis (CF) is lung infections caused by Pseudomonas aeruginosa (PA)⁴. Infections caused by PA are difficult to treat because they can quickly develop antibiotic resistance⁵. A synthetic third generation of fluoroquinolone antibiotic, levofloxacin, works well against PA^6,7. It is ofloxacin’s L-isomer, which has a wide range of antibacterial activity⁸. In in vitro testing, levofloxacin was shown to retain its antipseudomonal activity in the sputum of CF patients, in contrast to some other antibiotics such as tobramycin, which decreased under the same circumstances⁶. The fluoroquinolone is widely distributed and can be detected in most body tissues at concentrations higher than the corresponding concentrations in plasma. With a mean volume of distribution of 1.1 L/kg, levofloxacin distributes widely throughout the body and penetrates effectively into most body tissues and fluids⁹.

Inhalation delivery of antibiotics must be sufficiently deposited and distributed in the lungs to achieve local concentrations above the minimum inhibitory concentration (MIC)⁴. Inhalational administration of antibiotics can deliver high concentrations locally in the lungs to treat persistent bacterial infections¹⁰. In addition, topical antibiotic administration improves lung function, reduces exacerbations and intrapulmonary bacterial load⁴.

Dry powder inhalers (DPI) have been around for a while, are considered as medical devices that use dry powder as their active ingredient¹¹. Due to the solid form and relative chemical stability of the powder, DPIs do not require cold chain storage¹². With high pulmonary deposition, they enable the delivery of higher drug dosages⁷. Using a dry powder formulation of levofloxacin with utilizing local delivery option can improve patient compliance, making administration simpler, and saving money¹³. DPIs are portable tools that make it simple for the patient to administer the formulation. DPIs do not need propellants. They are portable devices that enable relatively easy administration of the formulation by the patient¹². Dry powder with the particles that fall between 1 and 5 μm in aerodynamic diameter have the ability to deposit in both the central and respiratory zones. On the other hand, smaller particles are primarily exhaled, and larger ones affect the oro-pharynx zone⁷.

Since toxicological and pharmacological responses are more closely associated with concentration of drug at target organs than in plasma, characterizing the kinetic profile of drugs in particular organs or tissues is becoming more interesting^14,15. The disposition of drugs in organs or body tissues, such as the lung, has been studied using isolated tissue techniques, which were first developed in the field of physiology^16–18. For toxicological, biochemical, or pharmacological studies, the isolated perfused lung preparation is a versatile method¹⁹.Cells are not fragmented or dispersed in the perfused lung preparation; rather, they are kept in their normal anatomical and physiological relationships. In addition, unlike incubation and homogenate preparations, intracellular cofactors are not diluted. Perfused lung experiments avoid some drawbacks of whole animal studies, including metabolism by other organs, elimination of metabolites, and redistribution of test materials to other organs^20–22. Most pharmacokinetic studies report only the kinetic profile of drugs in plasma, while distribution data typically focus on a few drug concentrations in major body tissues¹⁴. The results of these experimental methods highlight the importance of tissue isolation and artificial perfusion in characterizing drug disposition, particularly in body regions of pharmacological or toxicological interest. The aim of the present work was to evaluate the inhaled administration of a new levofloxacin formulation using an isolated perfused lung model and to compare it with the systemic administration of levofloxacin.

Materials and methods

Materials

Levofloxacin was kindly provided by tehranchemie, Iran. The perfusion medium for the lung experiments was a modified Krebs-Ringer buffer comprising NaCl, KCl, CaCl₂, MgSO₄, NaHCO₃, KH₂PO₄, D-glucose and bovine serum albumin (BSA). All chemicals used for the perfusion medium, except BSA, were of analytical grade. All chromatographic solvents were HPLC grade.

Dry powder formulation of Levofloxacin

Dry powder formulation of levofloxacin, generated by spray drying, was used to deliver a large dose of antibiotics in a single inhalation. In the previous study³, levofloxacin was co-spray-dried with sodium chloride and leucine. Various conditions were studied for optimum inhalation properties with leucine as an auxiliary substance and metal cations as a dispersant. Briefly, the feed solution was prepared by dissolving levofloxacin, leucine, and sodium chloride in water. Following the addition of hydrochloric acid (for adjusting pH to a value of 5.98), a Büchi Mini Spray Dryer B-191 (Büchi Labortechnik, Switzerland) was used for open-loop spray drying of the feed solution. A spray dryer with a two-fluid flow atomizer nozzle of 0.7 mm diameter was equipped with a 10% feed rate setting, equivalent to 4 mL per minute. Air was used as a drying gas at a flow rate of 600 L/h. The inlet temperature was set at 150 °C, and the aspiration ratio was set to 90%. In this study, the median diameter (D50%) was between 1.31 and 2.89 μm, which was appropriate for deep lung deposition^19,23. The aerodynamic performance of this same formulation (including FPF) has been previously assessed in reference³, confirming its capability to yield a respirable fraction on dispersion.

Animals

Animal study were done in accordance with ARRIVE guidelines²⁴. The Animal Ethics Committee at the Tehran University of Medical Sciences approved the experiment (IR.TUMS.REC.1394.334). All methods were carried out in accordance with relevant guidelines and regulations²⁵. Male Wistar rats from the Laboratory Animals Centre at Tehran University of Medical Sciences weighing 200 ± 20 g were used in this study. The rats were divided into two groups according to randomization (n = 5) and kept under standardized temperature (25 °C), humidity (50–60%), and light (12:12 h light-dark) conditions. Water and food were readily available to the rats until they were anesthetized with a ketamine-xylazine mixture (ketamine 100 mg/kg and xylazine 10 mg/kg) administered intraperitoneally (IP). Heparin was administered to the animals intravenously to prevent blood clotting in the lungs (1000 IU/kg) after IP anesthesia. Rats were perfused in single perfusion mode for 60 min or re-circular perfusion mode for 20 min according to the study plan.

Isolated lung model

Using the technique described by Wang et al.⁶, isolated lung perfusion was performed. Nevertheless, we have modified the technique for our laboratory. Briefly, the procedure consisted of performing a tracheotomy and tracheal cannulation of rats, followed by mechanical ventilation under positive pressure. The pulmonary artery and vein were cannulated after tracheotomy. It was established that the inflow cannula would be connected to a mechanical pump before being inserted in order to initiate artificial perfusion of the lungs simultaneously with interruption of blood supply to the tissues.

Perfusion of the pulmonary circulation was performed using a Terumo infusion pump at a constant flow rate of 4 ml.min^− 1. The perfusion medium for the lung experiments was Krebs-Ringer buffer pH 7.4, consisting of NaCl (118 mM), KCl (4.7 mM), CaCl₂ (2.5 mM), MgSO₄ (1.2 mM), NaHCO₃ (24.9 mM), KH₂PO₄ (1.2 mM), D-glucose (5 mM) and 4.5% w/v BSA. The perfusion medium was oxygenated by bubbling a mixture of 95% oxygen and 5% carbon dioxide (CO₂) 10 min prior to the perfusion’s start and throughout the experiment. CO₂ was added to the circulating buffer to maintain the pH at the range of 7.35–7.45. We used a Harvard rodent ventilator (Model 683, USA) programmed to provide a 10 ml.kg^− 1 tidal volume at a respiratory frequency of 60 respirations.min^− 1 of atmospheric air with an end-expiratory pressure of 2 cmH₂O. Air was conditioned before being supplied to the rats by connecting ventilator to a double-jacketed chamber with water at 37 °C to moisturize the air before delivery. Instead of supplying non-conditioned atmospheric air to the ventilator, the air was taken from this chamber.

Several criteria were used to evaluate the viability of lung tissue: firstly, to verify that all areas of the lung have been adequately perfused and no tissue edema had developed, the appearance of lung tissue was visualized at the beginning of the infusion and followed by the experimental procedure. The local areas which quickly lost blood or developed translucent areas during the experiment were a criterion for the non-viability of the experimental preparation; Secondly, since lactate dehydrogenase (LDH) increases in serum following lung injury, LDH was measured in perfusate of isolated lungs as an indicator of lung damage^18,26.

Samples were collected from the efferent fluid at specific time points after dose injection. Sampling times were as follows: every 1 min for the 0–6 min; every 2 min for the 6–20 min, and every 10 min for the 20–60 min.

Samples were centrifuged for 10 min at 13,000 rpm in Eppendorf tubes. Supernatants were transferred to new tubes and stored at − 20 °C until analysis. After the lung was dissected from the trachea and surrounding tissue, it was stored at − 20 °C for analysis.

Drug injection

Intratracheal administration of a dry powder formulation of Levofloxacin

In order to reduce the breath frequency prior to administering the powder formulation (The 3 mg formulation contains 2.5 mg of levofloxacin), the breath frequency was reduced from 60 to 20 breaths per minute. Through a needle that was placed above the bifurcation of the trachea into the main bronchi, the dose was actuated into the trachea. As soon as the dose was administered, the breath frequency was increased to 60 breaths per minute. To assess the exact dose delivered, the residual dose in the tracheal catheter was determined and included in the calculations. Single (for 60 min) and recirculation -mode (for 20 min) perfusions were used in intratracheal administration. In recirculation or closed-loop, the solution continuously circulated through the cannula from the reservoir (inflow) and exited from the lung via the efferent (outflow) back to the reservoir. In Single-pass or open-loop, fresh perfusate was delivered from a reservoir and collected once from the efferent without returning to the reservoir. In the recirculation setup, the perfusate volume was 100 mL, whereas in the single-pass configuration it was 240 mL.

Vascular administration of Levofloxacin solution

In a single mode perfusion, levofloxacin was administered as a bolus injection through the Y device of the inflow cannula as a 0.5 mg dose of levofloxacin dissolved in a perfusion medium without albumin (2.5 mg.ml^− 1).

Sample pretreatment and analytical technique

A Smartline isocratic reversed phase high performance liquid chromatography (HPLC) system was used to analyze levofloxacin concentrations in efferent fluid and lung tissue samples (Berlin, Germany). This apparatus featured an online degasser, a variable UV detector, a manual injection valve, and Chrom Gate software in addition to a stainless steel or inert Smartline S-1050 pump. The stationary phase was a Hector C18 column (150 × 4.6 mm, 5 μm). The column temperature was kept constant at ambient temperature. The mobile phase consisted of water, acetonitrile, triethylamine, and phosphoric acid mixed in the volumetric proportions 255:42:0.9:2.1 (v/v), then adjusted to pH 3.0 and used at a flow rate of 1 mL·min⁻¹. All measurements were carried out at 296 nm.

A 1 mg.mL^− 1 levofloxacin stock solution was made in methanol. The range of calibration standard solutions (diluted in Krebs-Ringer medium or blank lung tissue) was 20- 2500 ng.ml^− 1. For lung tissue, each sample was first weighed and mixed with saline at a 1:2 (w/v) ratio. The tissue was then homogenized using a high-speed homogenizer (IKA T25 ULTRA-TURRAX, Germany), and the supernatant from the homogenate was collected for analysis. To pretreat standard and samples (perfusate or lung tissue) before injection into HPLC, 25 µl of 70% trichloroacetic acid and 50 µl of internal standard solution (IS) (ciprofloxacin, 14 µg.ml^− 1) were added to 500 µl of the desired samples. Samples were centrifuged at 13,000 rpm for 10 min after vortexing for some minutes.

Finally, the supernatant was collected and 100 µl of it was injected into the HPLC. The retention times for levofloxacin and IS were 9.2 ± 0.5 and 10.5 ± 0.5 min, respectively. Selectivity, linearity, limit of quantification (LOQ), precision, accuracy, and recovery were all analyzed to ensure the validity of method based on Food and Drug Administration (FDA) guideline²⁷.

Calculations and statistical analysis

During this study, samples were collected at specific time points from the efferent fluid in single-mode perfusion and the reservoir in recirculation mode to allow calculation of levofloxacin concentrations. The average values ​​of 5 experiments were shown as data points in the graphs and tables. The statistical moments theory was used to analyze drug concentration curves in efferent fluid²⁸. According to the theory of statistical moments, the area under the curve (AUC) and mean transit time (MTT) in the isolated lung were calculated using the following equations derived from the analysis of the outflow curves:

Sigma Plot was used to do a student’s t-test at a significance level of 0.05 (Version 11, USA). Data were shown as means ± standard error (SE).

Results

Blank samples were analyzed for interference at the levofloxacin and IS retention times, and none was identified (Fig. 1). Thus, at 296 nm, there was an adequate response to levofloxacin and IS signals. The chromatographic run could be completed in less than 15 min. Plotting the area ratios of levofloxacin and IS versus the concentration of levofloxacin yielded the calibration curve. Over the concentration range of 20 to 2500 ng.ml^− 1, the linearity of the standard curve was demonstrated to be good (R² = 0.9934). At low, intermediate, and high concentrations, acceptable recovery for standard samples were observed. All intra- and inter-assay values for accuracy and precision were within acceptable range.

Fig. 1.

Chromatograms of (A) blank sample in perfusion medium; (B) spiked sample with 750 ng.mL⁻¹ levofloxacin, (C) perfusate sample after pulmonary delivery of dry powder, time = 30 min.

To compare levofloxacin concentrations after pulmonary (2.5 mg) and bolus (0.5 mg) administration, concentrations were dose-normalized. To facilitate comparison of different routes of administration, mean levofloxacin efferent concentrations were presented on a logarithmic scale as a function of perfusion time in Figs. 2 and 3. Ten minutes after administration of the solution as a bolus injection in single pass perfusion, the levofloxacin concentration in the efferent fluid fell below the limit of detection. However, 60 min after delivering the powder, concentrations were still detectable in the efferent fluid.

Fig. 2.

Mean efferent concentrations as a function of perfusion time after solution (bolus) and powder (intratracheal delivery) of levofloxacin after single dose of 0.5 mg and 2.5 mg, respectively, in a single pass perfusion (mean ± SE, n = 5 per groups).

Fig. 3.

Mean accumulated efferent concentrations as a function of perfusion time after powder (intratracheal delivery) of levofloxacin after a single dose of 2.5 mg in a recirculation perfusion (mean ± SE, n = 5).

Table 1 shows dose normalized values used to compare the efficacy of levofloxacin administration by various routes. After powder administration, the maximum concentration of levofloxacin in the perfusate was 1028.04 ± 306.9 ng.ml^− 1. In contrast, after solution administration, the maximum concentration of levofloxacin (at first sampling time) in the perfusate was 22876.29 ± 2507.6 ng.ml^− 1. The AUC_0−inf. after dry powder and solution administrations were 34209.17 ± 6446.1 and 329371.02 ± 94627.7 ng.min.ml^− 1, respectively. After 60 min of administration, the levofloxacin lung concentration in intratracheal method was more than 4 times than bolus, indicating that a significant proportion of levofloxacin was retained in lung tissue (P-value < 0.05). The concentration of levofloxacin in lung tissue in recirculation mode (powder administration) was 15567.7 ± 5030.7 ng.ml^− 1.

Table 1.

Levofloxacin dose normalized parameters after bolus and pulmonary administration at a dose of 0.5 and 2.5 mg, respectively, in a single pass perfusion in rats (mean ± SE, n = 5).

Routes of administration	Lung concentration* (ng.ml− 1. mg− 1)	AUC0−inf. (ng.min.ml− 1. mg− 1)	MTT (min.)	Cmax (ng.ml− 1. mg− 1)	Drug accumulation (%)
Intra tracheal (dry powder)	21644.63 ± 2919.4	34209.17 ± 6446.1	–	1028.04 ± 306.9	23.2
Bolus (solution)	4886.22 ± 1083.0	329371.02 ± 94627.7	0.51 ± 0.4	22876.29 ± 2507.6**	1.8

^*Levofloxacin concentration in lung tissue at the end of study (60 min).

^**For bolus administration is the concentration at first time of sampling.

Discussion

The disposition profile of levofloxacin in the lung has clinical significance as it is used in clinical practice for respiratory infections. To evaluate the pulmonary delivery of levofloxacin, the isolated, perfused and ventilated rat lung model was selected. The usefulness of isolated perfused lung in pharmacokinetic studies has been confirmed in previous studies^10,16. It is possible to achieve rapid onset of action, reduced systemic exposure, and fewer side effects by delivering locally acting drugs directly to the lungs. Systemic delivery of drugs via inhalation route is also possible due to the highly vascularized respiratory mucosa, the large absorptive surface area, and relatively low level of enzymatic activity in the lung²⁹.

The inhaled levofloxacin aqueous formulation, originally referred to as MP-376 (commercially known as Quinsair in Europe and Aeroquin in the USA), is a fluoroquinolone formulation developed for the long-term treatment of Pseudomonas aeruginosa lung infections in patients with CF. Quinsair formulation relies on the complexation of levofloxacin with divalent cations, such as Mg²⁺, to increase aqueous solubility and decrease epithelial permeability, resulting in decreased systemic absorption and increased pulmonary residence time. This vehicle provides a controlled exposure to CF patients with limited plasma exposure, supports high concentrations in the sputum, and has been tested in clinical studies in CF patients³⁰.

In the present study, pulmonary administration of levofloxacin was evaluated and compared with its bolus administration via single-pass and recirculation modes.

Compared to recirculation experiments, which require a volume of approximately 100 milliliters, the single-pass mode has the drawback of requiring a larger perfusate volume to maintain the isolated lung perfusion during prolonged single-pass experiments. Additionally, a major disadvantage of the recirculation mode, particularly with lipophilic solutes, is the rapid partitioning of these solutes from the perfusate into the more lipoid lung tissue. In consequence, a drug may reach to an unreal equilibrium with lung tissue as it accumulates in the perfusate. As a result, clearance of the drug from the lungs can be artificially slowed at significant lung retention levels. Due to the continuous elimination of drug from the lung via the perfusate or blood, this is unlikely to occur with single-pass isolated lung perfusion³¹.

This study was to investigate whether spray dried levofloxacin powder could easily cross the lung barrier from the air-to-blood side by collecting samples from the perfusate before (time 0) and at predetermined times during one hour after dosing. After administration of the powder, the concentration in the efferent fluid reached its maximum after 6.8 min in single-pass mode, indicating rapid absorption through the lungs, followed by a slow descending phase showing a continuous transfer from lung to perfusate media. In recirculation mode, however, levofloxacin rapidly appeared in efferent and the concentration increased at an almost constant rate (R² = 0.9867), which could be due to very high concentration of the drug in lung side which in turn could be a result of fast dissolution and distribution of engineered powder into lung tissue⁷. When developing drugs for pulmonary delivery, it is essential to consider not only the permeability of the drug, but also its retention in lung tissues²⁹.

Pulmonary retention is an indication of local deposition in the airways. The spray-dried levofloxacin/NaCl/leucine formulation was formulated to provide a high mass of drug in one inhalation by producing respirable, highly dispersed, high drug load particles. Leucine is used as a surface modifier, which increases aerosolization by decreasing particle cohesiveness, and sodium chloride is used to enhance de-agglomeration and increase the fine-particle fraction^3,32,33. In addition, inhalation delivers a high amount of levofloxacin directly to the airway surface, where it produces a high concentration that will be sustained at the site of deposition. This is primarily due to the method of administration and the properties of aerosol: while intravenous bolus dosing results in systemic distribution of the drug, inhalation leads to a local depot within the lung lumen.

Compared to bolus injection, pulmonary administration in both single and recirculation perfusions resulted in higher levofloxacin levels in lung tissue at the end of study, indicating a superior rout of administration as far as higher drug concentration in lung tissue is concerned. Since the physicochemical properties of the drugs play an important role in pulmonary accumulation, drugs with moderate to high lipophilicity and a basic amine component (pKa > 8.0) appear to have greater pulmonary accumulation¹. At physiological pH, levofloxacin has three different zwitterion groups: a carboxylic group with a pKa of 5.5, a piper-azinyl group with a pKa of 8.0, and another proton acceptor function with a pKa of 6.8 ± 0.3^34,35. Considering that levofloxacin has an intrinsic partition coefficient equal to 0.74^35,36, indicating a relatively high affinity for the lipidic phase. The study by Lee et al. evaluated the penetration of levofloxacin into lung tissue after oral administration to subjects undergoing lung biopsy or lobectomy. At every time point over the 24-hour sampling period, the concentration of levofloxacin in lung tissue exceeded that in plasma¹⁴. Having a partition coefficient equal to about 0.7 for levofloxacin, we may expect high concentration of the drug in lung tissue after 60 min in bolus mode. Thought this is true, we still have much higher lung concentration after pulmonary administration (Table 1), which again confirms that pulmonary administration could be a better delivery route to keep high lung drug concentration.

Conclusion

The isolated perfused lung is an excellent model for understanding the pulmonary disposition of inhaled substances to a considerable degree. Pulmonary dosing in this model resulted in detectable appearance of levofloxacin in the perfusate over time, confirming that the drug can cross the airway surface into the circulation. Based on this study, levofloxacin could be delivered locally to the lungs, increasing the concentration in the lung and reducing the systemic concentration.

Author contributions

Dr. Rouini and Dr. Toliyat are responsible for all content of the manuscript, including the data and analysis. MR, TT, MD, and MGK contributed to the conception and design of study. AB prepared the formulation. MD, MG, HL, BS, KM performed animal section. HL was analyzed the samples by HPLC. MR and MD interpreted the data. MD wrote the first draft of the manuscript. The revised manuscript was composed by incorporating all authors.

Funding

This study was supported by National Institute for Medical Research Development (NIMAD).

Data availability

Data availability will be obtained on request by MR Rouini (rouini@tums.ac.ir).

Declarations

Competing interests

The authors declare no competing interests.