Abstract
Advances in RNA interference technology have established it as a powerful therapeutic tool with important future potential. The design and the chemical modifications of the siRNA nucleotide backbone have greatly enhanced stability, durability, and pharmacokinetics while minimizing tolerability risks. The optimal combination of these modifications depends on the target gene, tissue, and RNA sequence, necessitating an iterative, experimental approach that currently relies heavily on animal models. To reduce the reliance and number of (humanized) animals required, we developed a human long-term liver 3D spheroid model designed for screening GalNAc-conjugated siRNAs which captures the process of uptake, potency, and durability for early in vitro screening. These liver spheroids remain viable in culture for at least 5 weeks while maintaining expression of the asialoglycoprotein receptor to facilitate GalNAc mediated uptake. siRNA was efficiently internalized by the spheroids without the need for transfection reagents, and its durable silencing efficiency was assessed by monitoring AHSA1 target gene expression over time. Target gene silencing in the spheroid model persisted up to 5 weeks post-treatment. Fluorescently labeled siRNA enabled visualization of uptake and distribution within the spheroid, revealing somewhat reduced siRNA accumulation in pericentral CYP3A4+ hepatocytes accompanied with somewhat reduced ASGR1 expression. No signs of hepatotoxicity were observed under the conditions used. By varying the number of phosphorothioate modifications in the siRNA backbone, distinct differences in silencing efficiency and durability were observed which were principally similar as obtained in vivo in mice. We propose that this long-term human liver spheroid model provides a valuable preclinical platform for evaluating siRNA-based therapeutics with respect to uptake, durability, and silencing efficiency, and could refine early in vitro screening and accelerate drug development.