Abstract
This study aimed to assess the nephrotoxicity associated with VRP-034 (novel formulation of polymyxin B [PMB]) compared to marketed PMB in a three-dimensional (3D) kidney-on-a-chip model. To model the human kidney proximal tubule for analysis, tubular structures were established using 23 triple-channel chips seeded with RPTEC/hTERT1 cells. These cells were exposed to VRP-034 or PMB at seven concentrations (1-200 µM) over 12, 24, and 48 h. A suite of novel kidney injury biomarkers, cell health, and inflammatory markers were quantitatively assessed in the effluent. Additionally, caspase and cytochrome C levels were measured, and cell viability was evaluated using calcein AM and ethidium homodimer-1 (EthD-1). Exposure to marketed PMB resulted in significantly elevated levels (P < 0.05) of four key biomarkers (KIM-1, cystatin C, clusterin, and OPN) compared to VRP-034, particularly at clinically relevant concentrations of ≥10 µM. At 25 µM, all biomarkers demonstrated a significant increase (P < 0.05) with marketed PMB exposure compared to VRP-034. Inflammatory markers (interleukin-6 and interleukin-8) increased significantly (P < 0.05) with marketed PMB at concentrations of ≥5 µM, relative to VRP-034. VRP-034 displayed superior cell health outcomes, exhibiting lower lactate dehydrogenase release, while ATP levels remained comparable. Morphological analysis revealed that marketed PMB induced more severe damage, disrupting tubular integrity. Both treatments activated cytochrome C, caspase-3, caspase-8, caspase-9, and caspase-12 in a concentration-dependent manner; however, caspase activation was significantly reduced (P < 0.05) with VRP-034. This study demonstrates that VRP-034 significantly reduces nephrotoxicity compared to marketed PMB within a 3D microphysiological system, suggesting its potential to enable the use of full therapeutic doses of PMB with an improved safety profile, addressing the need for less nephrotoxic polymyxin antibiotics.