Species differences in cellular processing of siRNA‐Lipid complex

Abstract

Lipid‐based nanocarriers provide a means to deliver nucleic acid therapeutics such as short interfering RNA (siRNA); a major challenge is the low success rate in preclinical‐to‐clinical translation. siRNA gene knockdown depends on its ability to reach the intracellular target RISC (RNA Induced Silencing Complex, degrades the targeted mRNA), and is hence determined by the kinetics of multiple cellular processes (see below). The present project used experimental and computational studies to investigate the species differences in the processing of siRNA‐lipid complex (Lipoplex).Experiments were performed using human breast MCF7 tumor cells transfected with Rab5, a marker for early endosomes (EE). Cells were treated with Lipoplex (with fluorescence‐labeled siRNA and lipids), in the presence of fetal bovine serum, mouse plasma, or rat plasma. Pilot studies showed that mouse and human plasma had similar effects on Lipoplex destabilization, cellular uptake and gene knockdown. Time lapse live cell confocal microscopy was used to monitor Lipoplex cellular binding, endocytosis, and siRNA release, every 10 min for 2 h. The results were analyzed using quantitative image analysis for the amounts of siRNA and lipids in individual regions of interest. Over 95% of intracellular siRNA+lipid signals in the first 10 min were co‐localized with Rab5, confirming Lipoplex was endocytosed into EE. Some intracellular siRNA+lipid signals were no longer associated with Rab5 after 30 min, indicating Lipoplex progression through the endocytic pathway. Of the Rab5‐co‐localizaing siRNA+lipid signals, the siRNA‐to‐lipid ratio (Ratio) decreased with time (up to 3‐fold between 10 to 90 min), whereas Ratio in the signals not co‐localizing with Rab5 was up to 4‐fold lower. These findings indicate continuous release of free siRNA from Lipoplex while in EE. A quantitative pharmacology or QP model was developed to describe the first order kinetics of Lipoplex transfer between extracellular fluid/cell membrane/intracellular endocytic organelles and siRNA release into the cytosol. The model assumptions were (a) extracellular Lipoplex destabilization due to interactions with plasma/serum proteins, (b) rapid degradation of free siRNA released from Lipoplex, (c) two pools of membrane‐associated Lipoplex (internalized and not internalized), and (d) Lipoplex degradation/elimination in late endosomes or lysosomes. Fitting the QP model to experimental results yielded the rate constants.We found significant species differences, i.e., 2‐ to >15‐fold variations in rate constants for Lipoplex binding to cell membrane, internalization, release of siRNA into cytosol, and elimination. These results suggest a need of accounting for species differences in cellular processing of Lipoplex during animal‐to‐human translation. This study further demonstrates the utility of QP in difficult‐to‐measure subcellular kinetic processes.Support or Funding InformationSupported in part by research grants (RO1CA158300, RO1CA163015, RO1EB015253) from NCI and NIBIB, and by Mosier Chair of Pharmaceutical Sciences.