Research paper
The permeability of large molecular weight solutes following particle delivery to air-interfaced cells that model the respiratory mucosa

https://doi.org/10.1016/j.ejpb.2008.09.006Get rights and content

Abstract

The transepithelial transport rates of compounds after deposition as aerosolised particles onto respiratory cell layers and allowing dissolution in the cell surface secretions has not been reported in a comprehensive manner to date. Here, the twin-stage impinger (TSI) was used to deposit potentially respirable particles (aerodynamically <6.4 μm) of varying molecular weight dextrans labelled with fluorescein isothiocyanate (FITC-dex) onto Calu-3 cells, a model of the bronchial epithelium. The TSI functioned as a particle size segregator, with >96% of the deposited particles being geometrically <6.4 μm (as measured by microscopy) and the particles being deposited discretely with a uniform distribution. Cell layers tolerated particle deposition at an air flow of 60 L/min. A small reduction in transepithelial electrical resistance (TER) of <10% occurred initially, but the original TER was recovered within 10 min and there was no significant effect on apparent permeability (Papp) of FITC-dex 4 over 4 h. Interleukin 8 (IL-8) secretion in the apical and basolateral directions over 24 h was not increased by exposure to the TSI and particle deposition. The rate of FITC-dex 4 (4 kDa) transport across the cell layer after deposition and dissolution of the particles in the cell surface secretions was ∼20-fold higher (P < 0.05) than if applied as a solution. The volume of cell surface secretions was estimated by tracer dilution (3.44 ± 1.90 μl, mean ± SEM) and this value was used to calculate the Papp of compound once deposited as a particle. The Papp value was found to be similar to that obtained when the compound was applied in solution (P < 0.05). Thus, the increased transport rate was attributable to the differences in donor chamber solute concentration rather than any change in the permeability of the cell layer itself. Following particle deposition, transport of FITC-dex with molecular weights between 4 and 70 kDa correlated well (r2 = 0.918) with reported in vivo canine pulmonary clearance after intratracheal instillation of dextrans of similar molecular weight. The use of the TSI and the Calu-3 cell line for the assessment of compound dissolution and transport rates after particle deposition may allow more realistic analyses to be made with respect to the in vivo situation.

Introduction

Most in vitro investigations of respiratory epithelial permeability or toxicity involve the application of test materials to layers of pulmonary epithelial cells as solutions or suspensions [1], [2], [3], [4], [5], [6], [7], [8], [9]. Within the lung, however, compounds are presented to the epithelial lining fluid as aerosolised solid particles or liquid droplets. Unless cleared by the mucociliary transport system or macrophages (bioresistant particles), particles deposited on the mucosal surface undergo dissolution followed by solute transport across the epithelium (biodegradable particles). Thus the process of particle dissolution in vivo presents a significantly different situation from the application of a solution to respiratory epithelial cells. Some of the major differences in presenting a compound to a cell layer as a particle as opposed to a solution are summarised in Table 1.

Compounds in solution present a uniform concentration over the surface of the cell layer. However, in the case of compounds administered as an aerosol, dissolution into the cell layer lining fluid (or spreading in the case of a deposited liquid droplet) can be envisaged to produce a concentration gradient. A saturated system may be produced immediately adjacent to the particle, with a decreasing concentration as a function of distance from the particle, which is dependent on the dynamics of lateral diffusion in the cell layer lining fluid and partition into the cell layer itself. Non-uniform exposure across the surface of the cell layer may be expected to affect toxicity and absorption profiles, but to date, there have been few attempts to evaluate this in vitro.

Although bespoke apparatuses have been developed to apply particles to cell-layers, they are highly technical and costly to setup [10], [11]. The most relevant studies to date have been performed using adaptations of relatively simple and commercially available apparatuses such as the twin-stage impinger (TSI), Andersen cascade impactor (ACI) and multi-stage liquid impinger (MSLI) [12], [13], [14]. For example, the MSLI has been used to deposit porous microparticles impregnated with sodium fluorescein (flu–Na) onto Calu-3 cell layers [12]. Once deposited, particles were immediately solubilised by the addition of a fluid to allow transepithelial electrical resistance (TER) to be measured and a starting concentration to be known. However, because of the addition of fluid immediately after particle deposition, particles did not dissolve within the cell surface fluid and the dissolution-transfer process that occurs in vivo was not modelled.

In this study, the TSI was adapted to accommodate a Calu-3 cell layer (an air-interface cell culture model of the bronchial epithelium which secretes a glycoprotein rich mucus layer [15]) in the lower chamber where respirable particles of an aerodynamic diameter <6.4 μm were deposited. This system allowed deposition of aerosols onto the surface of air-interfaced cell layers, producing a more representative exposure scenario for the assessment of mucosa-particle interaction in terms of toxicology and dissolution limiting permeability to that occurring in vivo. The objective of this work was to determine the extent to which the administration of compounds in particulate form would reproduce the transport kinetics observed in vivo and to evaluate these in relation to the permeability of the same compounds applied as aqueous solutions.

Section snippets

Materials

All materials were obtained from Sigma–Aldrich (Dorset, UK) unless otherwise stated. Cell culture flasks (75 cm2 with ventilated caps) and Transwell cell culture systems (0.33 cm2 polyester, 0.4 μm pore size) were from Costar (through Fisher Scientific, Leicestershire, UK). Cell culture reagents included trypsin/ethylene diamine tetraacetate sodium (EDTA) solution (2.5 g/L trypsin, 0.5 g/L EDTA), Hank’s balanced salt solution [HBSS, no phenol red, including NaHCO3 at 0.33 g/L with HEPES buffer (0.01 

Apparatus

Light microscopy was performed with a Wilovert S inverted microscope (Hunt Wetzlar, Germany). Particle size measurements were performed with a Nikon Labophot light microscope (Tokyo, Japan), linked to an Acorn personal computer with particle size analysis software designed in-house. TER was measured using chopstick electrodes and an EVOM voltohmmeter (STX-2 and Evom G, World Precision Instruments, Stevenage, UK). Cell layers were disrupted using a Vibracell 400 cell sonicator (Sonics and

Adaptation of the TSI for particle delivery

A glass TSI was assembled as described in the British Pharmacopeia [16], apart from the absence of solution in the lower chamber (Fig. 1a). The adapter piece (B) was removed and parafilm wrapped around the base of the connecting tube to produce an attachment surface for the Transwell insert. The Transwell insert was then pushed onto the connecting tube until the tapered internal walls of the insert fastened firmly onto the parafilm (Fig. 1b). Thus, air flowing down the connecting tube was

Particle deposition to cell layers

Light microscopy of representative particles (FITC-dex 4 and FITC-dex 40) showed that >96% of the particles which were deposited in the Transwell were <6.4 μm in geometric diameter (data not shown). The microscopy confirmed that the TSI was effective in permitting only particles of a respirable size to travel to the lower chamber and that particles deposited discretely with an even distribution over the surface of a cell-free Transwell. TER was measured after deposition of particles onto Calu-3

Discussion

In this study, a system for application of respirable aerosols to layers of epithelial cells modelling the bronchial region of the lung is reported. This system caused no measurable adverse effects to the cell layer and allowed measurements of permeability of the deposited compound to be obtained. To date there have been few published attempts to replicate the aerosol deposition that occurs in vivo using in vitro models. Classically, in toxicological and transport studies, compounds of interest

Acknowledgments

The authors are grateful to Unilever plc for the funding of this work and to David Lockley for his helpful guidance. The authors are also grateful for the help of Grace Grainger in the drawing of the Transwell insert.

References (28)

  • M.E. Cavet et al.

    Transepithelial transport of the fluoroquinolone ciprofloxacin by human airway epithelial Calu-3 cells

    Antimicrob. Agents Chemother.

    (1997)
  • J.P. Evans et al.

    Transport of a series of d-phenylalanine-glycine hexapeptides across rat alveolar epithelia in vitro

    J. Drug Target.

    (1998)
  • S. Boland et al.

    Diesel exhaust particles are taken up by human airway epithelial cells in vitro and alter cytokine production

    Am. J. Physiol. Lung Cell. Mol. Physiol.

    (1999)
  • M. Aufderheide et al.

    A method for the in vitro exposure of human cells to environmental and complex gaseous mixtures: application to various types of atmosphere

    ATLA—Alt. Lab. An.

    (2002)
  • Cited by (76)

    • Air-liquid interface (ALI) impact on different respiratory cell cultures

      2023, European Journal of Pharmaceutics and Biopharmaceutics
    View all citing articles on Scopus
    View full text