Advanced pathophysiology mimicking lung models for accelerated drug discovery

Abstract

Background: Respiratory diseases are the 2nd leading cause of death globally. The current treatments for chronic lung diseases are only supportive. Very few new classes of therapeutics have been introduced for lung diseases in the last 40 years, due to the lack of reliable lung models that enable rapid, cost-effective, and high-throughput testing. To accelerate the development of new therapeutics for lung diseases, we established two classes of lung-mimicking models: (i) healthy, and (ii) diseased lungs – COPD. Methods: To establish models that mimic the lung complexity to different extents, we used five design components: (i) cell type, (ii) membrane structure/constitution, (iii) environmental conditions, (iv) cellular arrangement, (v) substrate, matrix structure and composition. To determine whether the lung models are reproducible and reliable, we developed a quality control (QC) strategy, which integrated the real-time and end-point quantitative and qualitative measurements of cellular barrier function, permeability, tight junctions, tissue structure, tissue composition, and cytokine secretion. Results: The healthy model is characterised by (i) continuous tight junctions, (ii) physiological cellular barrier function, (iii) a full thickness epithelium composed of multiple cell layers, and (iv) the presence of ciliated cells and goblet cells. Meanwhile, the disease model emulates human COPD disease: (i) dysfunctional cellular barrier function, (ii) depletion of ciliated cells, and (ii) overproduction of goblet cells. The models developed here have multiple competitive advantages when compared with existing in vitro lung models: (i) the macroscale enables multimodal and correlative characterisation of the same model system, (ii) the use of cells derived from patients that enables the creation of individual models for each patient for personalised medicine, (iii) the use of an extracellular matrix proteins interface, which promotes physiological cell adhesion and differentiation, (iv) media microcirculation that mimics the dynamic conditions in human lungs. Conclusion: Our model can be utilised to test safety, efficacy, and superiority of new therapeutics as well as to test toxicity and injury induced by inhaled pollution or pathogens. It is envisaged that these models can also be used to test the protective function of new therapeutics for high-risk patients or workers exposed to occupational hazards. Graphical Abstract: [Figure not available: see fulltext.]