(W1030-06-39) Impact of Ionizable Lipid Source on the In Vitro Performance of siRNA Loaded Lipid Nanoparticles

Purpose: DLin-MC3-DMA (MC3), a widely used ionizable lipid and key component of lipid nanoparticle (LNP) systems, plays a critical role in siRNA encapsulation, endosomal escape, and the overall safety profile of siRNA-LNPs. While some efforts have been made to identify potential impurities of MC3, their impact on the quality attributes of siRNA-LNPs remains largely uncharacterized. This study aims to elucidate the relationship between MC3 purity and the physicochemical characteristics and in vitro performance of siRNA-LNPs. The findings are expected to offer valuable insights into impact of lipid source for MC3, supporting the development of generic LNP formulations.

Methods: MC3 lipids from three different vendors were evaluated for purity using liquid chromatography (LC) coupled with a charged aerosol detector (CAD), photodiode array (PDA), and mass spectrometry (MS). siRNA-LNPs were formulated with each MC3 source using the NanoGenerator Flex-M system and the Onpattro® composition, followed by dialysis into phosphate-buffered saline (PBS). Particle size distribution and surface charge were assessed via dynamic light scattering (DLS) and electrophoretic light scattering, respectively. Apparent pKa was determined using the 2-(p-toluidino)-6-naphthalenesulfonic acid (TNS) binding assay. LNP morphology and internal structure were characterized by cryo-electron microscopy (cryo-EM) and small-angle X-ray scattering (SAXS), respectively. siRNA encapsulation efficiency was evaluated using the Ribogreen assay. Thermal properties of formulations were assessed by differential scanning calorimetry (DSC). Functional activity was assessed in HepG2 cells by measuring transthyretin (TTR) mRNA and protein knockdown via qPCR and ELISA, respectively. Cell viability in HepG2 cells and immunogenicity in peripheral blood mononuclear cells were determined using the CellTiter-Glo® assay and a Luminex cytokine assay, respectively.

Results: Purity values determined by the LC-CAD method were 96.03, 86.30, and 98.63% for vendor 1, 2, and 3, respectively. The most abundant impurity, eluting at 7 minutes and present at 2, 7, and 0.8% of total peak area in vendor 1, 2, and 3 respectively, was identified by LC-MS as an oxidized and desaturated form of MC3 (m/z 656.5988, C₄₃H₇₇NO₃, mass error < 1 ppm). Figure 1 illustrates the UV chromatograms of MC3 lipids and the corresponding lipid nanoparticles prepared from each MC3 source, recorded at 270 nm. Since neither MC3 nor other LNP components exhibit UV absorbance at this wavelength, all peaks observed in the LC-PDA chromatograms are attributed to impurities present in the MC3 lipid. LC-CAD/PDA analysis of LNPs confirmed that all impurities in MC3, including low-abundance or volatile impurities (such as a unique peak at 10.9 minutes in vendor 3), carried over into the final siRNA-LNPs with profiles remaining largely unchanged post-formulation (Figure 1). These findings highlight that MC3 impurities can be retained throughout the formulation process, underscoring the importance of raw material purity in ensuring consistent LNP quality. Physicochemical and structural characterization of siRNA-LNPs prepared with each MC3 source showed slightly larger z-average particle size for vendor 2 (62.6 ± 3.2 nm) compared to vendor 1 (55.7 ± 3.5 nm) and vendor 3 (53.6 ± 1.9 nm), likely due to increased unsaturated lipid content affecting compaction. Despite this, all formulations had uniform size distribution (polydispersity index 0.08–0.18) and near-neutral zeta potential (–4.8 to –2.6 mV). Encapsulation efficiencies were consistent (~80%), and TNS assay-derived apparent pKa values (~5.9) showed no significant variation. SAXS and cryo-EM revealed similar internal structures across all samples, though DSC analysis showed lower phase transition temperature for vendor 2. The altered thermal profile may be due to impurity incorporation. All LNPs achieved ~83% TTR mRNA knockdown and ~62% protein knockdown in HepG2 cells at 5 nM siRNA, with no significant differences across MC3 sources (Figure 2). Cytotoxicity was negligible at both 5 nM and 30 nM doses. Immunogenicity screening in PBMCs treated with 30 nM siRNA-LNPs showed no notable variation in cytokine or chemokine induction.

Conclusion: This study examined the impact of MC3 lipid purity from different vendors on the characteristics and performance of siRNA-LNPs. Notable differences in impurity profiles were observed, with the most abundant impurity—an oxidized and desaturated form of MC3—ranging from 0.8 to 7% and persisting in the final LNP formulations. Despite these variations, the siRNA-LNPs made from the different MC3 lipids demonstrated no significant difference in physicochemical properties, in vitro target gene and protein knockdown, in vitro cytotoxicity and cytokine assay. These results suggest that, within the tested impurity range, such MC3-related impurities do not significantly impact in vitro performance of siRNA-LNP. This work offers insights to inform quality control strategies and define the potential impact of impurities in the development of generic LNP formulations.

Acknowledgements: The authors would like to acknowledge the U.S. Food and Drug Administration for financial support of this research (contract number: 75F40123C00118).The views expressed in this abstract do not reflect the official policies of the U.S. Food and Drug Administration or the U.S. Department of Health and Human Services; nor does any mention of trade names, commercial practices, or organization imply endorsement by the United States Government.


Figure 1. HPLC-PDA chromatograms (at 270 nm) of MC3 procured from three vendors and the siRNA-LNPs prepared by each MC3 source. As indicated by the red, green, and blue arrows, the impurity peaks observed in the individual MC3 lipids from vendor 1, 2, and 3, respectively, are carried over to the corresponding siRNA-LNP formulations.


Figure 2. TTR mRNA knockdown (48 h) and protein knockdown (72 h) at 5 nM siRNA concentration show no statistically significant differences across MC3 sources. Cell viability in HepG2 cells at 30 nM siRNA (48 h) remains comparable to media-treated controls. Complement C5a, TNF-α, and IL-6 secretion in PBMCs also indicate no impact of MC3 source on formulation immunogenicity. Data is shown as mean  SD, n=3-4. * P  0.05, **** P  0.0001.