From in vitro to in vivo: The Dominant role of PEG-Lipids in LNP performance

Abstract

Lipid nanoparticles (LNPs) are widely employed for delivering nucleic acids, such as mRNA, in both vaccines and therapeutic applications. These LNPs generally include an ionisable lipid (typically ∼ 50 %) to facilitate nucleic acid encapsulation and a PEG-lipid (typically ∼ 1.5 %) to enhance nanoparticle stability. To examine how the choice of PEG-lipid impacts LNP performance, we investigated the physicochemical characteristics and potency of LNPs prepared using two PEG-lipids with different acyl chain lengths: DMG-PEG 2000 and DSG-PEG 2000, containing 14 and 18 carbon tail lengths, respectively. These were combined with three commonly used ionisable lipids (ALC-0315, DLin-MC3 and SM-102). We evaluated the efficacy of these LNPs both in vitro (HeLa cells) and in vivo in mice after intramuscular (IM), subcutaneous (SC), and intravenous (IV) administration. In vitro studies showed that all LNP formulations primarily enter cells via clathrin-mediated endocytosis. Irrespective of the choice of ionisable lipid, DMG-PEG LNPs demonstrated higher in vitro mRNA transfection efficacy than DSG-PEG LNPs. These in vitro results aligned with the in vivo outcomes across all routes of administration tested. Our findings emphasise that despite the low percentage content of PEG-lipid, its selection critically influences LNP efficacy across different administration routes, with DMG-PEG-based LNPs outperforming DSG-PEG LNPs, regardless of the ionisable lipid used.

Introduction

Lipid nanoparticles (LNPs) are typically composed of four main lipid components: distearoylphosphatidylcholine (DSPC), an ionisable lipid, cholesterol and a PEG-lipid [1]. Among these, the ionisable lipid (which comprises approximately 50 % of the LNP composition) plays a critical role, facilitating nucleic acid complexation within the LNPs.

First described by Semple et al. in 2001 for nucleic acid delivery [2], ionisable lipids are designed to change charge in response to the solution’s pH, enabling the formation of particles that encapsulate the nucleic acid payload. In an acidic pH, these lipids acquire a positive charge, enabling high encapsulation efficiency of the negatively charged nucleic acids. During the particle formation process, the pH is subsequently raised, neutralising the ionisable lipid, rendering it more hydrophobic and thereby driving the ionisable lipid with mRNA into the interior of the lipid nanoparticles. As a result of this process, the potency of an LNP formulation is often reported to be linked to its pKa, with the most effective ionisable amino lipids having a pKa of around 6.5 [3]. Currently, lipid nanoparticle-based mRNA products approved for human use include Comirnaty™ [4], SpikeVax™ [5,6], and mRESVIA® [7], all of which are vaccines administered via the intramuscular route. In contrast, Onpattro® is an LNP-based RNA interference therapeutic used to treat hereditary transthyretin-mediated amyloidosis and is given via the intravenous route. In terms of choice of ionisable lipid, ComirnatyTM uses ALC-0315 ((4-hydroxybutyl)azanediyl)bis(hexane-6,1-diyl)bis(2-hexyldecanoate), SpikeVaxTM and mRESVIA® incorporates SM-102 (heptadecan-9-yl 8-((2-hydroxyethyl) (6-oxo-6-(undecyloxy) hexyl) amino) octanoate) and DLin-MC3-DMA (heptatriaconta-6,9,28,31-tetraen-19-yl-4-(dimethylamino)butanoate) is the ionisable lipid contained in Onpattro®. All three ionisable lipids have a common structure: an ionisable amine head group, a linker group, and hydrophobic tails, the latter of which confer a characteristic conical shape to the lipid. This conical shape is reported to facilitate LNP transfection to cells, as the broad shape of the tail region disrupts the endosomal membrane, rendering it easier for the mRNA to enter the cytosol [8].

Whilst representing only a small (1.5 %) proportion of the lipid formulation, PEG-lipids are another essential component of LNP formulations, providing stability by enhancing the hydrophilicity of the LNPs [9,10]. During LNP formation, PEG-lipids orient themselves with their hydrophilic head groups facing the LNP exterior, thereby improving the stability of LNPs during synthesis and storage and preventing LNP aggregation. However, the so-called “PEG dilemma” arises from the dual effect PEG can have on the fate of nanoparticles in vivo. While PEGylation of particles can extend the in vivo circulation time of particles by reducing particle opsonisation and clearance by the mononuclear phagocytic system, it can also decrease endosomal escape and LNP internalisation [11]. Therefore, to overcome the “PEG dilemma”, LNP formulations typically include a low concentration of PEGylated lipid (e.g., 1–2 %) [11,12]. Nonetheless, to address this “PEG dilemma” effectively, it is essential to consider the desorption dynamics of PEG-lipids. PEG desorption, the process by which PEG-lipids detach from the surface of lipid nanoparticles, is necessary to expose the nanoparticle surface to proteins, facilitating cellular uptake and transfection while maintaining a balance between stability and functional delivery. Importantly, PEG desorption is influenced by the length of the hydrophobic carbon tails, with the rate being inversely proportional to the length of the PEG-lipid hydrophobic anchor [13]. PEG-lipids with short lipid chains, such as DMG-PEG (C14), quickly dissociate in serum and are replaced by a protein corona, including the liver-specific ApoE [14]. The combined effect of PEG desorption and rapid protein corona formation leads to improved LNP internalisation and expression.

Among the approved LNP formulations currently available, those designed for vaccination are administrated intramuscularly into the deltoid muscle [4,15], while those designed for therapy (i.e. Onpattro®) are administered via intravenous infusion [16]. For liver-targeting LNPs, intravenous administration is preferred because the liver is the primary organ of LNP accumulation [17]. Nevertheless, liver accumulation also occurs after intramuscular administration [18,19]. In fact, following intramuscular administration, mRNA-LNPs can express not only at the injection site but also in the liver and draining lymph nodes [20,21], whereas mRNA-LNPs injected subcutaneously are generally confined to the site of injection [22]. This suggests that the choice of PEG lipids with different alkyl chain lengths is critical in LNP design and should be considered in combination with the route of administration.

Research on the effect of the PEG-lipid on LNP behaviour has been widely reported. Several studies have explored how different PEG-lipid anchor lengths impact LNP behaviour, including their pharmacokinetics [11,12], antibody production [23], transfection efficiency [24], and selective organ targeting [25]. However, the impact of combining PEG-lipids of varying alkyl chain lengths with ionisable lipids on LNP behaviour requires further investigation. To address this gap, we investigated the effect of using DMG-PEG (14 carbon tails) versus DSG-PEG (18 carbon tails) in combination with three commonly used ionisable lipids (ALC-0315, DLin-MC3, or SM-102) on the in vitro and in vivo fate of mRNA-LNPs. We further evaluated the in vivo efficacy of these ionisable lipid/PEG-lipid combinations when given by the intramuscular (IM), subcutaneous (SC), or intravenous (IV) route. This work intends to be a resource for selecting the appropriate ionisable lipid/PEG-lipid combination, considering the intended administration route and target site.

Download the full article as PDF here From in vitro to in vivo: The Dominant role of PEG-Lipids in LNP performance

or read it here

Materials

The PEG-lipids 1,2-distearoyl-rac-glycero-3-methylpolyoxyethylene (DSG-PEG 2000) and 2-dimyristoyl-rac-glycero-3-methoxypolyethylene glycol-2000 (DMG-PEG-2000) were purchased from Avanti Polar Lipids (Alabaster, AL, USA), while 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC) was purchased from Lipoid (Ludwigshafen, Germany). The ionisable lipids (4-hydroxybutyl) azanediyl)bis (hexane-6,1-diyl)bis(2-hexyldecanoate) (ALC-0315), (heptatriaconta-6,9,28,31-tetraen-19-yl-4-(dimethylamino)butanoate) (DLin-MC3-DMA) and (heptadecan-9-yl 8-((2-hydroxyethyl) (6-oxo-6-(undecyloxy) hexyl) amino) octanoate) (SM-102) were purchased from Broad Pharm, USA. EZ-Cap-modified firefly luciferase-mRNA (5-moUTP) was acquired from APExBIO (USA). Polyadenylic acid (PolyA), citric acid, sodium citrate tribasic dihydrate, Triton X-100 and cholesterol were purchased from Sigma Aldrich (St. Louis, MO, USA). Agarose, MOPS 10 X, Millennium RNA marker and RNase-free PBS 10X were purchased from Invitrogen. Quant-iT Ribogreen RNA assay, 1,1-dioctadecyl-3,3,3,3-tetramethylindotricarbocyanine iodide (DiR), formaldehyde loading dye, SYBR green stain II and sodium acetate were procured from Thermo Fisher Scientific (MA, USA). Other chemicals were used at analytical grade, and an in-house system provided RNA-se free water. DMEM cell culture media, TrypLE express, and L-glutamine were purchased from Gibco Life Technologies. Antibiotics penicillin/streptomycin and amphotericin B were purchased from Sigma Aldrich. The One-Glo Luciferase assay system and Vivo Glo Luciferin were bought from Promega. Endocytic pathway inhibitors chlorpromazine hydrochloride, cytochalasin D, and filipin complex ready-made solution were purchased from Sigma Aldrich. NucBlue live-ready probe was purchased from Invitrogen. All other chemicals were of analytical grade.

Ankita Borah, Valeria Giacobbo, Burcu Binici, Ranald Baillie, Yvonne Perrie, From in vitro to in vivo: The Dominant role of PEG-Lipids in LNP performance, European Journal of Pharmaceutics and Biopharmaceutics, Volume 212, 2025, 114726, ISSN 0939-6411, https://doi.org/10.1016/j.ejpb.2025.114726.