Application of a Renal Proximal Tubule Microphysiological System for Drug Safety Assessment and Disease Modeling

Abstract

A combination of in vitro assays and animal models are used to evaluate the properties of novel drug molecules, including safety and efficacy, throughout the drug development process. The data produced by in vitro assays and animal models guide the selection of the optimal compounds to advance in the drug development process and form the basis for initiation of human clinical trials. This screening methodology combining in vitro assays and animal studies to gain a wholistic picture of a drugs’ pharmacologic, toxicologic, and pharmacokinetic profile has enabled companies to bring molecules to market that successfully treat human disease and improve patient lives. However, the current drug development paradigm suffers from crucial limitations such as interspecies differences in animal models and poor cell phenotype of in vitro assays, which can cause the drug screening systems to fail to accurately imitate drug behavior in humans. As such, 85% of molecules that enter clinical trials do not make it to market because of a failure to demonstrate their safety and/or efficacy. In recognition of these shortcomings, the National Institutes of Health funded the development of tissue chips, which are 3-dimensional microphysiological devices that emulate tissue-specific microenvironments that support drug safety assessment, disease modeling, and efficacy testing. To that end, we previously developed a proximal tubule microphysiological system (PT-MPS) that recapitulates key metabolic, regulatory , and transport activities of renal proximal tubule cells in vivo. In this dissertation, we applied this PT-MPS to assess the safety of compounds with two antisense oligonucleotide backbone chemistries and to model the effect of glomerular dysfunction and subsequent proteinuria on the phenotype of the proximal tubule. We showed that compounds with phosphorodiamidate morpholino chemistry exhibited a more favorable safety profile while retaining pharmacodynamic activity compared to compounds with phosphorothioate chemistry. Together, these data support the choice of compounds with phosphorodiamidate morpholino chemistry for targeting renally-expressed genes such as cytochrome P450 3A5. We characterized the effect of proteinuria on proximal tubule cell phenotype and showed that human serum, but not albumin, causes a regenerative, secretory phenotype that is driven by a gain of nuclear factor kappa light chain enhancer of activated B cells (NF-κB) and activator protein 1 (AP-1) and loss of hepatocyte nuclear factor 4 alpha (HNF4α) and peroxisome proliferator activated receptor alpha (PPARα) epigenomic and transcriptional signatures. These findings are important because they largely reproduce the observations of other groups on the effect of proteinuria on proximal tubule epithelial cells (PTECs) and experimentally-derived animal models with glomerular dysfunction. Moreover, the response mirrors the changes observed from other insults suggesting that there is a conserved regulatory program that is activated by a variety of stressors and may represent a therapeutic target. Therefore, the PT-MPS can be used to evaluate novel strategies to prevent development of a secretory phenotype by PTECs during proteinuria, which is associated with fibrosis in progressive renal diseases . Taken together, the PT-MPS demonstrates utility in drug safety assessment and disease modeling, two areas where tissue chips could improve the drug development process.