The impact of lipid compositions on siRNA and mRNA lipid nanoparticle performance for pulmonary delivery
Abstract
Lipid nanoparticles (LNPs) are of significant interest as delivery systems for various RNA therapeutics, not least due to their outstanding success in applications including the COVID-19 vaccines and the siRNA therapeutic Onpattro®. As LNPs consist of different lipids, the lipid composition determines key properties of these particles. This study examines how lipid composition, especially helper and PEG-lipids, and RNA cargo (siRNA and mRNA) affect LNP performance in pulmonary delivery. By comparting two different helper and two different PEG-lipids, we assessed the impact on fusogenicity and endosomal escape, in vitro transfection efficiency, and subsequently protein corona formation. Their in vitro performance was assessed in the air-liquid interface (ALI) cell culture model, a sophisticated in vitro model of the lungs. Our results demonstrated that transfection efficiency and stability differ between the helper lipids DOPE and DSPC, depending on the RNA cargo. These differences can be attributed to the structural differences of the lipids and the different properties of the RNA molecules. Our investigations further demonstrated successful mucus penetration of all LNPs and 24-42% gene silencing in vitro. We also explored mucus proteins/LNP interactions in human lung mucus, finding distinct protein corona formation for DSPC- and DOPE-containing LNPs. This comprehensive analysis highlights the critical role of helper lipids in combination with RNA cargo in determining LNP properties, efficiency, and in vitro performance, providing valuable insights for optimizing RNA delivery systems.
Introduction
Over the past few years, lipid nanoparticles (LNPs) have established themselves as a delivery system for various types of RNA. The FDA approval of Onpattro®, the first small interfering RNA (siRNA)-based drug, followed by the mRNA vaccines from BioNTech/Pfizer and Moderna, has paved a promising path for the application of these RNA therapeutics utilizing LNPs as a delivery platform (Akinc et al., 2019, Schoenmaker et al., 2021). LNPs function for both cargos as delivery systems, although the structural properties, such as length, charge density, and stability differ significantly between siRNA and mRNA. mRNA is single-stranded and typically consists of more than 1,000 bases, making it relatively large and complex in structure. In contrast, siRNA is double-stranded and consists of only about 21 base pairs, making its structure comparatively small, less bulky, and less complex but rigid (Yin et al., 2014). Whereas siRNA can be used to selectively silence specific genes, mRNA enables the expression of desired proteins within cells.
Both approaches have significant therapeutic relevance. Exploiting these two mechanisms enables the treatment of numerous diseases associated with increased or decreased protein expression (Yin et al., 2014). A substantial proportion of currently untreatable diseases are represented by respiratory diseases. In 2017, chronic respiratory diseases were responsible for 3.9 million deaths globally (Labaki and Han, 2020). The application of RNA-based therapies in this context shows great promise. Delivery of RNA via LNPs for direct administration to the lungs is an ideal approach for treating respiratory diseases. Pulmonary delivery offers numerous advantages, including non-invasiveness, localized delivery resulting in reduced dosage and side effects, a large alveolar surface area, and minimal nuclease activity (Shaffer, 2020, Kandil and Merkel, 2019). However, pulmonary delivery also presents challenges for LNPs as a delivery system. One of the biggest hurdles that LNPs must overcome to reach the target cells is the mucus layer covering the pulmonary epithelium and the mucociliary clearance (Tafech et al., 2024).
To address the challenges associated with pulmonary delivery, this study evaluated the impact of variations in helper lipids, PEG-lipids and RNA cargo on the physicochemical properties and in vitro performance of LNPs. LNPs typically consist of four different lipid components: an ionizable lipid, a cholesterol (derivative), a helper lipid, which is often a phospholipid, and a PEG-lipid (Hald Albertsen et al., 2022). For our study, we based our formulations on the established Onpattro® composition. The LNP formulations consistently contained the ionizable lipid Dlin-MC3-DMA (MC3), essential for encapsulating RNA, and cholesterol, which is crucial for the structure and stability of the LNPs (Hald Albertsen et al., 2022). Additionally, the formulations incorporated one of two different helper lipids, namely 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE) or 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC), and one of two PEG-lipids, namely 1,2-dimyristoyl-rac-glycero-3-[methoxy(polyethylene glycol)-2000] (PEG-DMG) or 1,2-dimyristoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (PEG-DMPE). Helper lipids play a vital role within LNPs, significantly contributing to their overall structure, stability, and functionality (Kulkarni et al., 2019). As demonstrated by Kauffmann et al. they can also have an impact on transfection efficiency (Kauffman et al., 2015). Due to their diverse chemical structures, different helper lipids can influence interactions with cellular membranes, such as the endosomal membrane, substantially affecting cytosolic release of RNA and, therefore, transfection efficiency (Kulkarni et al., 2017). Helper lipids are also important in protein corona formation, potentially influencing biodistribution, cellular uptake and targeting (Zhang et al., 2021). PEG-lipids are primarily responsible for LNP colloidal stability. By forming the outer shell of the LNPs with the acyl chain anchored in the LNP and the hydrophilic PEG chain sticking out, they prevent aggregation or coalescence of the LNPs (Suzuki et al., 2020). While this is one of the most important roles of PEG-lipids, they can also influence other LNP properties such as immunogenicity by shielding the particles from opsonization and affecting the formation of the protein corona, which in turn influences in vivo distribution (Knop et al., 2010).
As discussed above, key properties of LNPs that contribute to transfection efficiency are influenced by their lipid compositions. To further investigate this observation, we first evaluated the physicochemical properties of various LNP lipid compositions containing siRNA or mRNA. Next, we compared the performance of these formulations in terms of cellular uptake, gene silencing or expression efficiencies, respectively, using a submerged cell culture model. As extensively reported in the literature, most RNA-based LNPs taken up by the cells remain trapped in the endosome, limiting their biological activity. To better understand how different lipids influence this process, we also assessed their endosomal escape ability. Since submerged models have limitations in replicating the complexity of lung barriers and the pulmonary environment, we further investigated LNP performance using an air-liquid interface (ALI) model. Cells cultured at the ALI form a pseudostratified epithelium, allowing for their differentiation into a mucociliary phenotype with mucus production and tight-junction formation (Baldassi et al., 2022). These characteristics better represent lung conditions, providing a more realistic model to assess the capacity of the LNPs to overcome key lung barriers such as mucus. Additionally, we investigated the formation of the protein corona on LNPs with regards to pulmonary delivery and their lipid composition. This was accomplished by analyzing the adsorption of proteins from lung mucus onto the surface of the LNPs.
Download the ful article as PDF here The impact of lipid compositions on siRNA and mRNA lipid nanoparticle performance for pulmonary delivery
or read it here
LNP Preparation
The siRNA-loaded and mRNA-loaded LNP formulations were based on the clinically approved Onpattro® formulation. In total, six lipids were employed in this study: (6Z,9Z,28Z,31Z)-heptatriaconta-6,9,28,31-tetraen-19-yl 4-(dimethylamino)butanoate (Dlin-MC3-DMA (MC3), MedChemExpress, Monmouth Junction, USA) as ionizable lipid, cholesterol (Sigma Aldrich, Taufkirchen, Germany), as helper lipids either 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC, Sigma Aldrich, Taufkirchen, Germany) or 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE, Lipoid GmbH, Ludwigshafen, Germany) and as PEGylated lipid either 1,2-dimyristoyl-rac-glycero-3-[methoxy(polyethylene glycol)-2000] (PEG-DMG 2000, Lipoid GmbH, Ludwigshafen, Germany) or 1,2-dimyristoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (PEG-DMPE, Lipoid GmbH, Ludwigshafen, Germany). RNA cargos that were encapsulated in this study were siRNA against the green fluorescent protein (GFP) gene (siGFP) and a scrambled negative control siRNA (siNC), both were obtained from Merck (Darmstadt, Germany). siRNA targeting human GAPDH (siGAPDH) was purchased from Integrated DNA Technologies (Leuven, Belgium), and mRNA encoding for eGFP (mGFP) was purchased from RiboPro (Oss, the Netherlands).
Stina Rademacker, Simone Carneiro, Müge Molbay, Federica Catapano, Ignasi Forné, Axel Imhof, Richard Wibel, Christoph Heidecke, Peter Hölig, Olivia Merkel, The impact of lipid compositions on siRNA and mRNA lipid nanoparticle performance for pulmonary delivery, European Journal of Pharmaceutical Sciences, 2025, 107182, ISSN 0928-0987, https://doi.org/10.1016/j.ejps.2025.107182.
If you missed our webinar, you can watch the recording here:
Smarter Stabilization for Lipid Nanoparticles: Formulating Biopharmaceuticals with Sucrose
