(720e) In Vitro and In Silico Characterization of Human Nasal Epithelial Pathophysiology in Cystic Fibrosis Airway Disease

Serrano Castillo, F., University of Pittsburgh
Corcoran, T., University of Pittsburgh
Bertrand, C. A., University of Pittsburgh
Confer, W. J., University of Pittsburgh
Shapiro, M. E., University of Pittsburgh
Parker, R. S., University of Pittsburgh
Introduction: Cystic Fibrosis (CF) is a recessive genetic disease caused by mutations in the CFTR gene, which codes for a widely expressed anion channel located in the apical membrane of epithelial cells. Loss of CFTR function results in an osmotic imbalance due to defective ion and water transport. Human nasal epithelial (HNE) cell cultures (HNE) are a novel experimental model of CF and show the same transepithelial transport defects observed in the CF airway. HNEs are easily harvested, can be cultured following well established protocols, and have the potential to allow for matched in vitro studies analogous to common clinical measurements. In this work, we implement a combination of experimental and computational methods with the aim of elucidating physiological and mechanistic differences in the ion and liquid transport mechanisms in HNEs harvested from non-CF (healthy), biological parents of a CF patient (carriers of a single CF-causing allele), and CF donors.

Methods: We have performed extensive characterization on well-differentiated non-CF (n=11), carriers of CF (n=2), and CF (n=5) HNE cell cultures; recruiting of donors for this study is still continuous with the aim of increasing of cohorts’ sample size. The study includes epithelial short-circuit current measurements, to assess CFTR channel activity, apical volume regulation and paracellular flux assays (as measured by Tc99m-DTPA (DTPA)), to study response to volume and osmotic challenges, and standard protein expression to assess changes across the three cohorts. We have also used this data to aid in the development of a biologically meaningful, compartmental model of airway epithelial transport physiology. The model is centered on a cellular compartment interfacing with two extracellular compartments representing the apical and basolateral fluid volumes. The model allows for ion and liquid transport both trans- and paracellularly. Individual terms are written for each major transporter and ion channel, as dictated by electroosmotic driving forces, as well as for intracellular volume regulation, electrolyte concentrations, membrane potentials, osmotic-driven water flux, and both diffusive and convective paracellular ion fluxes. Appropriate parameters ranges were identified through Latin hypercube sampling of the biophysically-relevant parameter hyperspace, and evaluation of the simulated results in order to guarantee model adherence to biological and physical constraints. Preliminary population-level fits were carried out using APT-MCMC, a Python-based Markov Chain Monte Carlo (MCMC) optimization package that combines affine-invariant samplers with parallel tempering to allow for the exploration of objective functions with multiple local minima and significant parameter correlation (Zhang, et al, Comput. Chem. Eng., 2017). Physiologically-motivated constants used in the model were obtained from previous literature on the modelling of human bronchial and nasal epithelial electrophysiology (Sanfur, et al, PNAS, 2017; Falkenberg, et al, Biophys. J., 2010).

Results: Our studies showed that non-CF HNE cultures have a higher stimulated CFTR current than CF carriers (p<o.o5) and CF (p<0.01) HNE cultures. This trend matches the results obtained from whole cell mature CFTR protein expression levels seen between the three cohorts. Similarly, thin film studies revealed significantly decreased apical DTPA retention (p < 0.05) in CF HNE cultures, as well as apical volume at 12 and 24 hrs following an apical volume challenge (p<0.01) when compared to non-CF or CF carrier samples. Preliminary results from model simulations suggest that an increase in the paracellular ionic and liquid permeabilities in CF samples may be a cause of these changes. No significant difference was found in the apical volume, or paracellular transport between non-CF and CF carriers. We have also shown that the liquid hyperabsorptive behavior seen in CF HNEs can be corrected following an apical hyperosmotic challenge to levels comparable to those of non-CF and CF carrier samples. DTPA absorption rates are also decreased following a hyperosmotic challenge for all cohorts, but non-CF cultures appear to be more sensitive than CF ones as measured by the change in normalized apical DTPA retention at 24 hrs post volume challenge (p<0.05).

Conclusion: CF HNEs have higher liquid absorption rates than both non-CF HNEs and CF carriers HNEs. This behavior may be correlated with the lack of functional CFTR expression. It is of particular interest that cultures from CF carrier donors, even with significantly lower CFTR current and protein expression levels, still show transepithelial water transport dynamics similar to those of non-CF donors. This might suggest that in CF carriers, the epithelium favors different transport dynamics in order to accommodate the decreased CFTR availability. This notion is supported by simulations that predict a shift in the ratio of convective to diffusive ionic paracellular flux in CF carriers compared to non-CF donors. We intend to explore the nature of these physiological differences with our model with the aim of defining parametric regions that characterize the three cohorts.