Rationally designed negatively charged lipid nanoparticles reduce systemic reactogenicity of mRNA vaccines while preserving immunogenicity

Abstract

Messenger RNA (mRNA) vaccines have emerged as versatile platforms for the prevention of infectious diseases and therapeutic cancer vaccination. However, strong reactogenicity remains a key challenge, with severe adverse events such as myocarditis. These concerns may limit the applications of mRNA vaccines. As lipid nanoparticles (LNP) are the primary driver of inflammation, optimizing LNP composition is a promising approach to reduce mRNA vaccine reactogenicity. However, the effects of tuning the physicochemical properties of LNP, particularly their surface charge, remain insufficiently characterized.

In this study, we aimed to engineer negatively charged LNPs by incorporating anionic phospholipids and evaluate their biodistribution, immunogenicity, and adverse reactions in mice. Negatively charged LNPs significantly reduced systemic inflammatory responses and fever compared to a conventional LNP. Moreover, cardiac troponin I—a marker of myocarditis—was reduced in negatively charged LNPs. Despite reduced systemic reactogenicity, negatively charged LNPs maintained antigen-specific immunogenicity comparable to a conventional LNP. The negatively charged LNPs did not attenuate local inflammation at the injection site.

This local inflammation depended on the fatty acyl chain of phospholipids in the LNP and was independent of the head group structure or negative charge. Surface charge modulation also reshaped biodistribution to distal organs, such as the spleen and liver. Correlation analysis linked reduced reactogenicity to decreased hepatic delivery of LNPs, and prior datasets supported this relationship. Taken together, our results establish phospholipid-driven surface-charge control as a promising approach to reduce systemic reactogenicity while maintaining the immunogenicity of mRNA vaccines.

Introduction

The development of messenger RNA (mRNA)-based medicines has rapidly expanded, enabling their application to diverse diseases [1]. For example, mRNA vaccines can induce robust humoral and cellular immune responses, and have become an important vaccine platform. Following the success of COVID-19 vaccines, mRNA vaccines have been developed against a broad range of pathogens, including respiratory syncytial virus and influenza virus. In addition to preventing infectious diseases, mRNA vaccines are being actively explored as cancer vaccines to induce tumor antigen-specific T cell responses [1].

Despite their strong immunogenicity, mRNA vaccines frequently cause systemic adverse reactions (e.g., fever and fatigue) and local reactogenicity (e.g., pain and swelling at the injection site) [2], [3]. Notably, the frequency and severity of these reactions tend to increase after booster vaccination, along with enhanced antigen-specific immune responses [4]. Moreover, rare but serious events, such as myocarditis, have been reported following mRNA vaccination [5]. Such reactogenicity remains a major hurdle for therapeutic cancer vaccines, which frequently require higher and repeated doses than infectious disease vaccines, and their combination with other immunotherapies, such as immune checkpoint inhibitors, may further exacerbate systemic inflammatory responses [6]. In addition, studies in humans and animal models have suggested that a fraction of intramuscularly administered mRNA vaccines can distributed beyond the injection site into the circulation and other organs [7], [8], yet the biological consequences of vaccine components present in off-target tissues remain incompletely understood. Taken together, substantial reactogenicity and a limited mechanistic understanding of biodistribution may contribute to vaccine hesitancy and hinder broader applications of mRNA vaccines [9]. Therefore, the reduction of adverse reactions is a critical goal.

Lipid nanoparticles (LNP), which encapsulate and deliver an mRNA payload, may contribute substantially to the reactogenicity of mRNA vaccines through activation of innate immune pathways [10], [11]. Crucially, “empty” LNPs without mRNA can also trigger inflammatory cytokine production (e.g., interleukin-6 (IL-6), interleukin-1β (IL-1β), and tumor necrosis factor-α (TNF-α)), local inflammation at the injection site, and systemic responses such as fever [12], [13], [14], [15], [16]. We previously demonstrated that these specific cytokines, particularly IL-6 and TNF-α, are the primary mediators of fever following mRNA vaccination [15]. Thus, optimization of LNP formulations is a promising strategy for reducing reactogenicity.

LNP typically comprise four major components: ionizable lipid, phospholipid, cholesterol, and polyethylene glycol (PEG)-conjugated lipid. Ionizable lipids have been extensively optimized because they influence mRNA encapsulation and endosomal escape, and can also drive innate immune stimulation [17], [18], [19], [20]. We also demonstrated that employing a biodegradable ionizable lipid reduced inflammatory responses and caused fewer adverse reactions with robust antigen-specific immunity compared to a conventional formulation [21]. In addition to ionizable lipids, modifying “non-ionizable” components—phospholipids, cholesterol, and PEG-lipids—can tune LNP-driven immunogenicity and reactogenicity, although such studies remain limited [22], [23]. Among the non-ionizable components, phospholipids are structurally diverse molecules comprising hydrophobic fatty acyl chains and hydrophilic head groups. Variations in the chain length, degree and position of unsaturation, and head group chemistry can markedly influence both the physicochemical properties of LNP and their biological responses [23]. For example, increasing the proportion of 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC) in LNP has been reported to enable B-cell-selective delivery to the spleen after intravenous administration [24]. We further showed that replacing the DSPC of LNP with an unsaturated phospholipid—1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC)—reduced systemic adverse reactions and attenuated delivery to distal organs, such as the liver [23].

Importantly, phospholipid engineering provides a practical means for modulating the surface charge of LNP. Particle surface charges can alter their fate in the body, and charge control enables organ-selective delivery [25]. For instance, incorporation of anionic phospholipids, such as phosphatidic acid or phosphatidylserine, into LNP for intravenous administration can reduce liver delivery and enhance delivery to secondary lymphoid tissues, such as the spleen and lymph nodes, whereas the inclusion of cationic lipids, such as 1,2-dioleoyl-3-trimethylammonium-propane (DOTAP), can promote delivery to the lung [26], [27]. However, these insights are largely based on intravenous delivery paradigms developed for gene therapy. The effect of charge-modified LNPs on the biodistribution, immunogenicity, and reactogenicity of intramuscularly administered mRNA vaccines remains poorly characterized.

In this study, we aimed to generate negatively charged LNPs by modifying phospholipid components and comprehensively evaluate their biodistribution, immunogenicity, and adverse reactions following intramuscular administration as mRNA vaccines in mice.

Download the full article as PDF here Rationally designed negatively charged lipid nanoparticles reduce systemic reactogenicity of mRNA vaccines while preserving immunogenicity

or continue reading here

Formulation of LNP

SM-102 was obtained from Cayman Chemical (Cat. #33474; Ann Arbor, MI, USA). DSPC, DOPC, DSPA, DSPS, DOPS, and PEG2k-DMG were sourced from NOF Corporation (Tokyo, Japan; Cat. #MC-8080, MC-8181, MA-8080LS, MS-8080LS, MS-8181LS, and GM-020). DOPA was purchased from Avanti Polar Lipids (Alabaster, AL, USA; Cat #840875P), and cholesterol was supplied by Sigma-Aldrich (St. Louis, MO, USA; Cat #C8667). The LNPs were assembled on a NanoAssemblr Ignite microfluidic device (Precision Nanosystems, Vancouver, Canada). Lipids were dissolved in ethanol, while mRNA was diluted in 6.25 mM acetate buffer (pH 5.0).

For the standard LNP formulation, SM-102, DSPC, cholesterol, and PEG2k-DMG were combined at a molar ratio of 50:10:38.5:1.5, and alternative formulations were derived by modifying this baseline composition. The L/R ratio (lipid nmol/mRNA μg) was set at 33.3. The two phases were combined at a volumetric ratio of 1:3 (organic:aqueous) and a total flow rate of 2 mL/min, yielding LNP at an N/P ratio of 5.5, defined as the molar ratio of ionizable lipid amines to mRNA phosphates. Immediately after mixing, the suspension was diluted threefold with 20 mM MES buffer (pH 5.5) and subsequently transferred into PBS by ultrafiltration using an Amicon Ultra-15 device with a 100-kDa Ultracel-PL membrane (Merck, Darmstadt, Germany; Cat. #UFC810096). The hydrodynamic diameter, polydispersity index, and zeta potential of the mRNA-loaded lipid nanoparticles were determined using dynamic light scattering (Zetasizer Nano ZS, Malvern Panalytical, Malvern, UK). The proportion of encapsulated mRNA and total RNA content was evaluated using the Quant-iT RiboGreen RNA assay (Invitrogen, Carlsbad, CA, USA; Cat. #R11491).

Tatsuya Karaki, Wakana Inoue, Tang Gis, Shintaro Shichinohe, Tokiko Watanabe, Toshiro Hirai, Yasuo Yoshioka,
Rationally designed negatively charged lipid nanoparticles reduce systemic reactogenicity of mRNA vaccines while preserving immunogenicity, Journal of Controlled Release, Volume 400, Part A, 2026, 115490, ISSN 0168-3659, https://doi.org/10.1016/j.jconrel.2026.115490.

See our next webinar:

Elevate process, maximize flexibility: advanced QbD and Continuous Manufacturing for moisture-protective coatings