For virtual screening experiments, molecules with maximal tissue exposure (AUC) in the airways after inhalation were identified by using combinations of logPn and pKa as input parameters in a multiscale, cell-based lung transport model (). For weak bases, lower lipophilicity and higher pKa promoted intracellular retention and led to greater local exposure relative to the alveoli (). The calculated airway/alveoli exposure ratio () ranged from 100 to 700 and increased with lowered logPn (increasing hydrophilicity) and higher pKa (greater positively charged fraction at physiological pH) Essentially, cell-permeant, hydrophilic molecules harboring a fixed positive charge showed the greatest accumulation and retention in the cells of the upper airway relative to the alveoli, following IT administration.
Virtual screening of monobasic compounds based differential tissue distribution in the airways and alveoli.
To probe the role of the route of administration, simulations were also performed by independently varying logPn and pKa to calculate the mass deposition pattern in the airways and alveoli under steady state conditions after IV administration (). In this manner we established the relationship between the physicochemical properties of small molecules and absolute and relative mass distribution in the airways () and alveoli (). Following IV administration, the majority of the mass was deposited in the alveoli irrespective of the physicochemical properties of the molecules (); the airways held less than 20% of total drug mass in the lungs. Compounds with low logPn and high pKa tended to exhibit the largest airway to alveoli mass ratios, which paralleled the results obtained after IT administration.
In order to validate the results of these virtual screening experiments, two fluorescent bioimaging probes, MTR and Hoe, were selected for more detailed analysis. MTR is a cell-permeant, hydrophilic cation, and Hoe is a cell permeant, hydrophobic weak base. Based on the screening results () and more detailed simulations (), the concentration profiles of these two fluorescent molecules in the airways and alveoli were markedly different after IT () and more similar after IV () administration. When given IT, the predicted MTR concentration, 40 to 60 min after administration, was nearly 10-fold higher in the airways than in the alveoli (). Conversely, the predicted concentration of Hoe in the airways was two-fold higher in alveoli than in airway (). When given IV, the predicted concentration of MTR in the airways was almost the same as that in alveoli (). However, the predicted concentration of Hoe in the airways was higher in alveoli (). Thus, MTR should be retained in the airways specifically after IT administration, whereas Hoe should not be retained in airways relative to alveoli regardless of the route of administration.
Simulations of local pharmacokinetics of MTR and Hoe after IV an IT administration.
Next, cell based assays were used to establish the intracellular retention of MTR and Hoe at a site of absorption. For this purpose, a transwell insert system with micro-fabricated pores was constructed. After seeding MDCK epithelial cells on the patterned pore arrays and adding hydrophobic fluorescent compounds in the basolateral side of cell monolayer, the time course dye uptake in the cells sitting above the pores and the kinetics of lateral transport from the cells lying on top of the pore to the neighboring cells was visualized by fluorescence microscopy.
Three hours after the addition of Hoe to the basolateral compartment, only cells that were within close vicinity of pores were stained, indicating that the cells formed a tight seal with the pores such that each pore fed almost exclusively into cells that were in immediate contact with the pores (). Monitoring of the cell-to-cell diffusion of Hoe over time, indicated that the pores served as point sources of sustained dye supply to the adjacent cells () and for cells grown on membranes with pores spaced by 80 µm () or 160 µm (), each pore could be considered as the single point source of dye molecules. Quantitative image analysis revealed that the rate of staining rapidly decreased as the distance of the cells from the pores increased (). Remarkably, only cells in the vicinity of each pore were labeled.
Probing the intracellular retention of Hoe along the plane of a cell monolayer.
As controls, cells were stained with Hoe plus BCECF-AM from the basolateral compartment (). BCECF-AM is a nonfluorescent cell-permeant ester, which generates a cell-impermeant, fluorescent molecule upon intracellular hydrolysis. While the extent of Hoe diffusion was dependent on the distance from the pores (), the green fluorescence of the hydrophillic ester hydrolysis product (BCECF) was exclusively restricted to the first layer of cells that were in direct contact with pores ().
Similar to the Hoe staining pattern, MTR also exhibited a highly constrained diffusion pattern with most of the staining restricted to the vicinity of each pore (). After two-hours of staining from the basolateral compartment with both Hoe () and MTR (), only cells within 60 microns of the pore being stained with both probes (). The normalized fluorescence intensity of MTR and Hoe were similar in the first and second layers of cells, but MTR showed higher penetration into the third layer ().
Probing the intracellular retention of MTR along the plane of a cell monolayer.
In the transversal direction, the absorption and retention of MTR and Hoe across multiple layers of cells was also assessed in primary NHBE cells differentiated as multilayers in ALC (). For the experiments, MTR and Hoe were simultaneously added in the apical side of the cells and intracellular accumulation was assessed using 3D reconstructions of the cell multilayers (). As a positive control, LTG was also included in the apical HBSS buffer. Thirty minutes after the addition of probes to the apical compartment, both MTR and Hoe staining were constrained to the first, outer surface layer of cells (, left). The cells beneath the surface layer of cells were stained with LTG (, right), indicating that the limited penetration of both MTR and Hoe. Different transport patterns of MTR, Hoe and LTG across the cell multilayers were verified by image quantitation using MetaMorph® software in the multiple Z-stack images of NHBE cell multilayers. Approximately 96%±2.76% of MTR or 96%±2.48% Hoe of the dye was retained in the surface cell layer whereas 50%±15.62% of LTG fluorescence was associated with the surface cell layer. Tukey's multiple comparison test following ANOVA (one-way analysis of variance) test showed statistically significant difference between MTR and LTG (p-value<0.0001) and also between Hoe and LTG (p-value<0.0001), but not between MTR and Hoe with p-value larger than 0.05 (α
Fluorescent confocal images of NHBE cell multilayers on the porous membrane with Z-stacks.
As an ultimate test of the results of in silico virtual screening experiments, mice were administered a mixture of MTR and Hoe by either IV tail vein or IT instillation and the distribution of the molecules in the lungs was assessed by fluorescent microscopy (). Hoe distributed throughout the lungs regardless of route of administration () with fluorescence in both alveoli and airways ()). Following IV administration, MTR also distributed throughout the lung in both airways and alveoli (). Conversely, IT administered MTR resulted in highly uneven fluorescence distribution (). Most importantly, the airway regions showed comparable MTR fluorescence in airway vs. alveoli after IV () but higher MTR fluorescence intensity in airways compared with the alveoli following IT delivery ().
Tiled fluorescent micrographs of coronal cryosections obtained from the left lungs of mice.
To confirm these observations quantitative image analysis was performed to compute background subtracted integrated intensity of alveolar and airway regions, to quantify the relative, differential fluorescence intensity distribution of Hoe and MTR in airway and alveoli. The fluorescence MTR/Hoe ratio ranged from 2.42 to 3.27 for IT administration. For MTR and Hoechst, the mean (± s.d.) percent airway delivery was 23.9%±5.8% and 8.8%±2.7%, respectively (based on 422 region measurements from a single lung). For IV administration, the fluorescence MTR/Hoe ratio ranged from 0.95 to 1.45. The mean (± s.d) percent airway delivery for MTR and Hoe were 7.5%±2.5% and 7.1%±1.8%, respectively (based on 383 region measurements from a single lung). The images and measurements were consistent with local intracellular retention of MTR in the airways compared with Hoe, following IT (but not IV) instillation. These in vivo results paralleled the in silico simulation results ().
In order to identify the most important parameters that might explain the differences in local retention of MTR and Hoe, a parameter exchange analysis was performed using computational simulations. For this purpose, individual parameters of the airway were exchanged with those of the alveoli, one at a time, and the simulations were rerun to calculate the exposure (AUC) of MTR and Hoe. Based on the results of this simulation analysis () the volume of interstitial smooth muscle cells together with the volume of mitochondria were the primary factors determining the retention of MTR in the upper airways relative to alveoli. Secondarily, the surface areas of epithelial and endothelial cell layers were important, affecting retention in opposite directions. Taken together, these results suggest that the mitochondrial density per unit absorption surface area is the key histological organization parameter responsible for the higher retention of MTR in upper airways after IT administration.
Results of parameter exchange analysis.