Synthesis and Assembly of a New-generation Bifunctional Lipid Nanoparticle for Selective Delivery of mRNA to Antigen Presenting Cells

Abstract

Described here is the synthesis and assembly of a new-generation mRNA-bifunctional lipid nanoparticle (mRNA-BLNP3) for selective delivery of mRNA to antigen-presenting cells (APCs). Compared to mRNA-BLNP1 and BLNP2 (BioRxiv,doi.org/10.1101/2023.12.26.572282), the mRNA-BLNP3, incorporating a glycolipid with the lipid moiety designed to target the CD1d receptor on dendritic cells (DCs) and the sugar head group targeting the mannose-binding receptors on APCs is more selective with stronger target-specific immune responses. It was shown that vaccination in mice with mRNA-BLNP1 elicited enhanced cytokine induction and antibody responses compared to traditional mRNA-LNPs, and the mRNA-BLNP2 vaccine, incorporating a mannose-glycolipid, further improved DC targeting. However, BLNP3, incorporating the glycolipid with an aryl-mannose head group targeting the mannose receptor on APCs and the same lipid moiety targeting the Cd1d receptor on DCs, showed superior lymph node targeting in vivo with reduced liver accumulation, enhanced mRNA expression in DCs and macrophages, and increased DC maturation. Immunization in mice with mRNA-BLNP3 elicited enhanced humoral and cellular immune responses compared to mRNA-BLNP1 and mRNA-BLNP2, with higher antigen-specific IgG titers and granzyme B–producing CD8⁺ T cells, demonstrating that BLNP3 is a promising bifunctional lipid nanoparticle for delivery of mRNA vaccines with improved efficacy and safety.

Introduction

While many vaccine candidates against SARS-CoV-2 have been investigated in the past years (1), the speedy development of effective mRNA vaccines demonstrated the most powerful approach to control the pandemic.(2–5) The success of mRNA vaccines depends on the fast design of stable mRNA with nucleotide modification and the subsequent manufacture and encapsulation in lipid nanoparticles (LNPs) for vaccination. Current mRNA vaccines (4, 5) encoding the full length spike protein, contains pseudo-uridine and two proline substitutions to reduce immunogenicity and stabilize the prefusion state, respectively and is formulated with a mixture of four lipids, including a ionizable lipid, a phospholipid, a cholesterol and a PEG-conjugated lipid to form an mRNA-LNP for immunization (6). Vaccination with mRNA is attractive because the mRNA can rapidly translate to the corresponding protein antigen in the cytoplasm after immunization (7, 8) and provide adjuvant activities via toll-like receptors (TLRs) to enhance immune response (7). The success of mRNA vaccines against SARS-CoV-2 has stimulated interests in developing new mRNA vaccines against other diseases and new methods for targeted delivery of mRNA (9) to improve efficacy and safety (10–13).

Among the mRNA delivery systems, non-viral vectors with low immunogenicity represent a simpler and safer alternative to viral vectors. With the advances in development of new biocompatible materials and innovative fabrication approaches, non-viral vectors have become the preferred vehicles for mRNA delivery, and a variety of non-viral vectors have been explored, including lipids and polymers (14, 15). Recent development of LNPs has been focused on encapsulation of mRNA to reduce its degradation by nucleases and provide better transfection efficiency (16) and selective delivery to target specific subsets of antigen-presenting cells (APCs) (17). However, the issue of selectivity remains a challenge (18). Therefore, it is important to develop new LNPs with better selectivity for APCs to enhance immune responses and minimize undesirable translation of mRNA in different cells (19–20). We have previously developed a bifunctional LNP (BLNP1) containing the glycolipid C34 as a selective delivery component to target the CD1d receptor on dendritic cells (DCs) and the T cell receptor on iNKT cells to optimize the Th1 and Th2 pathways in CD4+ and CD8+ T cell responses (21). In addition, we modified the glycan moiety of C34 (22, 23) with mannose to generate BLNP2 with improved selectivity. Herein, we have developed an aryl-mannose glycolipid to target CD1d and DC-SIGN on dendritic cell (DC) and combined with other lipids to generate mRNA-BLNP3 for selective delivery of mRNA to APCs. The spike-mRNA of wild-type SARS-CoV-2 formulated as mRNA-BLNP3 vaccine was shown to elicit stronger immune responses with higher levels of spike protein-specific IgG antibodies than the traditional mRNA-LNP and other mRNA-BLNP vaccines.

Results

In our first approach, we reduced the four-component LNP commonly used in mRNA formulation to a two-component lipid nanoparticle to simplify the formulation process. We first prepared a guanidinyl lipid and a zwitterionic lipid to form mRNA-GZ-LNP (Fig. 1). In general, LNPs include a helper lipid and a cationic lipid. The helper lipid was used to stabilize the mRNA-LNP delivery system, such as 1,2-dioctadecanoyl-sn-glycero-3-phosphocholine (DSPC)(24) or 1,2-dioleoyl-sn-glycerol-3-phosphoethanolamine (DOPE)(25). These helper lipids not only stabilize the particles but also enhance mRNA delivery efficiency (26). Ionizable lipids are of great importance for delivery of mRNAs because their positively charged head groups can interact with the negatively charged phosphate of the mRNA molecule (27), and protonation of the ionizable components destabilizes the anionic membrane and facilitates nanoparticle disassembly, leading to the release of mRNA to the cytosol (28). To facilitate the mRNA encapsulation and delivery, the ionizable lipid is the most abundant component in LNP formulation. However, the phosphate groups with negative charge in helper lipids allow selective delivery to the spleen for protein expression in spleen and lymph nodes (29). Therefore, we introduce a higher number of zwitterion groups into the formulation of mRNA-LNPs to ensure a more precise mRNA delivery. By connecting a guanidine head with a tertiary amine and a phosphate group in one molecule (L5), the guanidine group is positively charged at physiological and acidic conditions, and the phosphate group is negatively charged at physiological conditions and neutral at acidic conditions. The combination of these functional groups increased the overall phosphate ingredient on lipid nanoparticles (IZ-LNP) to facilitate targeting.

Figure 1.

(A) Structure of synthetic lipids, (B) constructs of different mRNA-LNPs and mRNA-BLNPs preparations with bifunctional glycolipid for APC-targeted delivery

To improve target delivery, we then prepared different bifunctional glycolipids with the sugar head targeting APCs and the lipid moiety targeting CD1d on DCs (Fig. 2) and include the bifunctional lipid in the mRNA-GZ-BLNP, mRNA-IZ-BLNP and mRNA-BLNP formulations. Using these BLNPs, we can synthesize nanoparticles in sizes of one to two hundred nm with high transfection efficiency in HEK293 cells (BioRxiv, doi.org/10.1101/2023.12.26.572282). Although it is practical to use these BLNP formulation systems for in vitro or in vivo study, the long-term storage and serum instability of these BLNPs was still a concern due to the absence of cholesterol and PEGylated lipids. When mRNA-GZ-BLNP1 was incubated with 10% FBS at 37 °C for 6–48 h, we found a gradual increase of particle size as the incubation time increased (BioRxiv, doi.org/10.1101/2023.12.26.572282). Interestingly, inclusion of L3 showed a slightly stabilization of LNP, while inclusion of cholesterol and PEGylated lipid improved the transformation of LNPs to multilayer nanoparticles. Therefore, for in vivo studies, 1.5% PEG lipid and 38.5% cholesterol were included in mRNA-GZ-BLNP1 or mRNA-IZ-BLNP1 to improve stability. In the assembly of mRNA with GZ-LNP/IZ-LNP and mRNA mixing with L3, we observed an average particle size of mRNA-GZ-BLNP1 or mRNA-IZ-BLNP1 at around 174 or 140 nm by DLS, respectively with spherical morphology revealed by TEM and CryoEM (BioRxiv, doi.org/10.1101/2023.12.26.572282). We then immunized mice with spike-encoding mRNA-LNPs and a significantly higher level of neutralizing antibodies in the mRNA-GZ-BLNP1 group was found. In addition, the IgG titer in mice immunized with mRNA-GZ-BLNP1 of WT, Delta, or Omicron spike protein was determined, and it was found that the WT mRNA-BLNP1 vaccine elicited high IgG titers against the WT spike protein, but the titer was reduced against the Delta and Omicron spike proteins. However, though the WT mRNA vaccine with deletion of glycosites in the stem (WT-deg-CD/HR2)-GZ-BLNP1) induced a similar level of IgG titers against the WT spike protein, it generated significantly higher IgG titers against the Delta and Omicron spike proteins (30). On the other hand, when we incorporated the bifunctional lipid L3 into a conventional LNP formulation (Fig. 1B), we observed a strong induction of both cytokines IL-4 and IFN-γ (BioRxiv, doi.org/10.1101/2023.12.26.572282) and a selective uptake of mRNA-BLNP1 by BMDCs but not by the non-BMDCs cells in flow cytometry analysis. The other mRNA-BLNP2 with mannose as the head group also showed a similar phenomenon, indicating that the BMDC uptake was mediated by the lipid moiety. In a separate study, the formulated mRNA–LNPs were incubated with dendritic cells (DCs), B cells, and T cells to evaluate cell-type specificity. Both mRNA-BLNP1 and mRNA-BLNP2 exhibited a pronounced preference for DCs compared with B cells and T cells. Notably, mRNA-BLNP2 demonstrated the highest DC targeting efficiency, with 38.3% FITC-positive cells, which is likely attributed to dual targeting mediated by CD1d and mannose-binding receptors on DCs.

Figure 2.

Preparation of glycolipid L3, L4 and L6. Reagents and conditions: (a) 11-bromoundecanoic acid, PPh₃, toluene, reflux, 24h, 76%; (b) 4-(4-fluorophenoxy)benzaldehyde, LHMDS, THF, 0°C to rt, 12h, 69%; (c) H₂, Pd/C, MeOH, 2h, quant; (d) phytosphingosine, HBTU, DIPEA, THF, 12h, 80%; (e) 2,2-DMP, PTSA, 2h, quant; (f) 4-nitrophenyl chloroformate, Et₃N, THF, 12h. Compounds I(22), II(31), III(31) were synthesized according to the reference respectively.

With this promising result, we decided to incorporate L6 (Fig. 2), which contains the aryl-mannose head as a better ligand for DC-SIGN (31), into the design of new-generation BLNP (BLNP3). We next evaluated whether the mRNA delivery specificity and enhanced innate immune responses observed with CD1d- and mannose-receptor-targeted BLNPs in vitro could be recapitulated in vivo. C57BL/6 mice were intramuscularly (i.m.) immunized in the leg muscle with luciferase (Luc) mRNA-loaded BLNP3 or conventional LNP, and luciferase expression was monitored by in vivo bioluminescence imaging at 1, 6, and 24 h post-injection (Fig. 3A). Both formulations primarily transfected the injection site and, to a lesser extent, other organs. Because restricting mRNA transfection to the administration site and draining lymph nodes are critical to minimizing systemic toxicity, we analyzed the biodistribution of LNPs in major organs and inguinal lymph nodes (iLNs). The Ex vivo luminescence imaging revealed no significant accumulation in the heart, spleen, lung, or kidney for either BLNP3 or conventional LNP, with signal detected mainly in the liver and iLNs (Fig. 3B). However, BLNP3 administration resulted in a marked reduction of hepatic radiance compared with conventional LNP, where liver signal accounted for 75% of total intensity, confirming the superior lymph node-targeting capacity of BLNP3 (Fig. 3C).To investigate which cell types in the lymph node were transfected by LNPs, eGFP-encoding mRNA was delivered and detected by flow cytometry. It was found that BLNP3 successfully delivered eGFP mRNA to antigen-presenting cells, including macrophages and dendritic cells (DCs), with DCs (2.6-fold) and macrophages (1.4-fold) exhibiting higher mRNA expressions compared to LNPs (Fig. 4B and C). Since enhanced expression of mRNA in APCs is important for the activation of adaptive immunity and targeting CD1d or DC-SIGN can stimulate DC maturation (32), we analyzed matured DCs (CD80+CD86+) using flow cytometry and found that mRNA-BLNP3 markedly increased the percentage of matured DCs in lymph node (Fig. 4D and E). In contrast, LNP only modestly induced DC maturation, suggesting that the mRNA-BLNP3 formulation can enhance the selective delivery of mRNA to APCs and activate DCs. All these observations were further supported with cellular uptake assay (Fig. S1) and flow cytometry analysis (Fig. S2–S5).

Figure 3.

Selective mRNA delivery with mRNA-BLNP3 formulation. (A) In vivo bioluminescence imaging at 1, 6 and 24 h post-treatment of Luc mRNA-loaded LNP vs BLNP3. Mice were i.m. injected with Luc mRNA-loaded LNP or Luc mRNA-BLNP3 (5 μg mRNA per mouse). Total luminescence was quantified at different time points. (B) Ex vivo luminescence imaging. Mice were i.m. injected with Luc mRNA-LNP or Luc mRNA-BLNP3 (5 μg mRNA per mouse) 1, 6, 24 h before they were sacrificed. Major organs and iLNs were collected for ex vivo imaging. (C) Quantification of luminescence signals in B.

Figure 4.

(A) Mice were i.m. injected with eGFP mRNA-LNP and eGFP mRNA-BLNP3 (5 μg mRNA per mouse) 6 h before they were sacrificed, and iLNs were collected for FACS analysis. The error bar around each data point is the SD. Flow cytometry analysis of (B) DCs (C) macrophages, and matured DCs: (D) CD80⁺CD11c⁺ cells and (E) CD86⁺CD11c⁺ cells in iLNs (n = 3). (F) Cellular immune responses induced by H1N1 mRNA LNP vaccine: scheme of the prime, boost vaccination and the analysis of the humoral and T cell responses. C57BL/6J mice were i.m. immunized twice with 4, 1 or 0.25 μg of H1N1-mRNA LNP or H1N1-mRNA BLNP3 vaccine on days 0 and 14. (G) H1N1-specific IgG antibody titers in the mice treated with H1N1-mRNA-LNP or H1N1-mRNA-BLNP3 at different dosages. Analysis of cytotoxic T cell immunity induced by H1N1-mRNA vaccines in mice (H) The percentage of GrzB-producing cells in CD3⁺CD8⁺ T cell populations. Mean ± standard error of the mean (SEM) was plotted, and the statistical significances compared to PBS group were determined by one-way ANOVA with Prism 10.4.1.

Next, we evaluated humoral and T cell responses in immunized mice. Vaccination followed the timeline depicted in Fig. 4F, with a prime dose on day 0 and a boost on day 14. Although no significant difference was observed at the higher dose (4 μg), H1N1-mRNA-BLNP3 elicited higher anti-H1 IgG titers than conventional H1N1-mRNA-LNP at 1 μg and 0.25 μg doses (Fig. 4G). Moreover, immunization with H1N1-mRNA-BLNP3 elicited a 2.65-fold increase in the frequency of granzyme B–producing CD3⁺CD8⁺ T cells compared to H1N1-mRNA-LNP (Fig. 4H). These results indicate that the mRNA-BLNP3 formulation enhances both humoral and cellular adaptive immune responses compared to the clinically relevant mRNA-LNP vaccine. Collectively, these findings demonstrate that the mRNA-BLNP3 formulation can significantly improve the immunogenicity of mRNA vaccines, underscoring their translational potential.

Discussion

In this study, we compared a series of bifunctional lipid nanoparticle (BLNP) formulations for enhancing the specificity and immunogenicity of mRNA vaccines and simplified the conventional LNP systems into guanidinium/zwitterion-based BLNPs. We successfully achieved selective targeting of antigen-presenting cells, particularly dendritic cells (DCs), via CD1d and mannose-binding receptor pathways. Among them, the mENA-BLNP3 formulation exhibited superior lymph node targeting while minimizing off-target liver accumulation, and experiments in vivo demonstrated significantly enhanced mRNA expression in DCs and macrophages, increased DC maturation (CD80⁺CD86⁺), and elevated cytokine responses, including IFN-γ and IL-4. Importantly, the mRNA-BLNP3 formulation also promoted stronger humoral and cellular adaptive immune responses, including higher IgG titers and increased granzyme B–producing CD8⁺ T cells, especially at lower antigen doses. These findings highlight the potential of bifunctional glycolipids for designing next-generation mRNA delivery systems that not only improve cellular targeting but also vaccine efficacy. Taken together, the bifunctional BLNP3 represents a promising candidate for the development of mRNA-BLNP3 vaccines with enhanced immunological performance and safety profiles.

Materials and Methods

All study and synthesis (Schemes S1 and S2) data are included in the article and/or SI Appendix.