Visualizing lipid nanoparticle trafficking for mRNA vaccine delivery in non-human primates

Abstract

mRNA delivered using lipid nanoparticles (LNPs) has become an important subunit vaccine modality, but mechanisms of action for mRNA vaccines remain incompletely understood. Here, we synthesized a metal chelator-lipid conjugate enabling positron emission tomography (PET) tracer labeling of LNP/mRNA vaccines for quantitative visualization of vaccine trafficking in live mice and non-human primates (NHPs). Following intramuscular injection, we observed LNPs distributing through injected muscle tissue, simultaneous with rapid trafficking to draining lymph nodes (dLNs). Deltoid injection of LNPs mimicking human vaccine administration led to stochastic LNP delivery to three different sets of dLNs. LNP uptake in dLNs was confirmed by histology, and cellular analysis of tissues via flow cytometry identified antigen-presenting cells as the primary immune cell type responsible for early LNP uptake and mRNA translation. These results provide insights into the biodistribution of mRNA vaccines administered at clinically relevant doses, injection volumes, and injection sites in an important large animal model for vaccine development.

Keywords: PET-CT imaging; lipid nanoparticles; mRNA vaccines; non-human primate; vaccine biodistribution.

Graphical abstract

Figure 1

Radiometal-chelating LNPs enable loading of PET tracers while retaining mRNA delivery function (A) Dynamic light scattering (DLS) analysis of LNPs with or without DOTA-lipid. (B) LNP quality control metrics table for LNPs with or without DOTA-lipid. (C) CryoTEM imaging of LNPs, DOTA-LNPs, and non-radioactive Cu-loaded DOTA-LNPs. Scale bars, 50 nm. (D–F) C2C12 cells were incubated with 10 μg/mL mCherry-encoding mRNA delivered by LNPs or non-radioactive Cu-loaded DOTA-LNPs for 24 h, then analyzed by flow cytometry for mCherry expression. Shown are histograms of mCherry fluorescence (D), percentages of mCherry-positive cells (E), and mean fluorescence intensities of transfected cells (F). Statistical significance was determined by one-way ANOVA followed by Tukey’s post hoc test. ∗p < 0.05; ∗∗∗∗p < 0.0001. All data show means ± SEM. (G) BALB/c mice (n = 5 animals/group) were immunized i.m. with 10 μg of LNPs or non-radioactive Cu-loaded DOTA-LNPs that were labeled with DiD. Popliteal LNs were harvested at 12 h and analyzed for the percentage of DiD⁺ cells among monocytes, DCs, and subcapsular sinus macrophages (SSMs). (H and I) BALB/c mice (n = 4–5 animals/group) were immunized i.m. with 10 μg LNPs or non-radioactive Cu-loaded DOTA-LNPs encapsulating mCherry mRNA. Popliteal LNs were harvested 24 h post-injection and assessed for mCherry expression by flow cytometry. Shown are representative flow plots of mCherry expression in live cells recovered from negative control lymph nodes, LNP-immunized, and non-radioactive Cu-loaded DOTA-LNP-immunized lymph nodes (H), and the percentages of mCherry-positive cells (I).

Figure 2

PET-CT imaging reveals LNPs primarily distribute at injected muscle and immediate draining lymph node in mice (A) PET-CT study timeline. DOTA-LNPs encapsulating 5 μg mCherry mRNA were administered i.m. into the gastrocnemius muscle of BALB/c mice (data pooled from two studies, one with n = 3 animals/group and one with n = 5 animals/group). (B) PET-CT projections of mice over the imaging time course. Two different thresholds/viewing angles are shown; upper panels show a higher sensitivity view to visualize lower LNP signals detected in iliac LNs, and lower panels show a lower sensitivity view to visualize the injection site and popliteal draining LNs. Scale bars, 50 mm. (C–E) ROI analyses of PET signal at injection site (gastrocnemius muscle, C), draining popliteal LNs (D), and liver (E) for animals receiving free ⁶⁴Cu or ⁶⁴Cu-loaded DOTA-LNPs. (F) Ex vivo tissue gamma counter measurements of ⁶⁴Cu signal from DOTA-LNPs compared with blank and free ⁶⁴Cu controls. Statistical significance was determined by multiple unpaired t tests followed by Holm-Sidak post hoc test. ns, p > 0.05; ∗p < 0.033; ∗∗p < 0.002; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001. All data show means ± SEM.

Figure 3

LNPs rapidly reach draining lymph nodes following i.m. administration in rhesus macaques (A) Schematics of PET-CT timelines for mCherry (left) and N332-GT2 (right) NHP studies. Animals received injections of 50 μg mRNA per site i.m. (B) PET-CT projections of one representative animal from the mCherry study over time. (C) PET-CT projections of a second representative animal from the mCherry study over time. (D) PET-CT projections of one representative animal from the N332-GT2 study over time. (E) PET-CT projections of a second representative animal from the mCherry study over time. Scale bars in B–E represent 50 mm.

Figure 4

LNPs access draining lymph nodes but show very limited systemic distribution following i.m. injection in rhesus macaques (A–D) ROI analyses of PET signal at quadriceps muscle injection site (A), deltoid muscle injection site (B), quadriceps-draining lymph nodes (C), and deltoid-draining lymph nodes (D). (E and F) ROI analyses of PET signal in the deltoid-draining lymph nodes (apical, pectoral, and central axillary) (E) and the quadriceps-draining lymph node (iliac) (F) with different line colors corresponding to individual animals. (G–I) ROI analyses of PET signal at liver (G), spleen (H), and heart (I). Statistical significance was determined by one-way ANOVA followed by Tukey’s post hoc test. ns, p > 0.05; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001. All data show means ± SEM.

Figure 5

Ex vivo tissue analysis reveals LNPs and mRNA are persistent in draining lymph nodes of NHPs (A) NHP draining and non-draining (contralateral) lymph node samples taken at 40 h post-immunization. LNP signal via diD in green, tissue stain in gray. Imaging was conducted on 28 lymph node tissue samples collected from all quadriceps and deltoid-draining lymph node sites; shown are representative lymph node images. Magnification at 25× with left panel showing merged signal and right panel showing only LNP signal. Scale bars, 400 μm. (B) 63× magnification comparing LNP signal (green) between selected draining (white box indicating zoomed in region of view in 25× image) and non-draining lymph node samples. Scale bars 115 μm. (C) Expression of N332-GT2 mRNA in lymph nodes from ipsilateral (right) and contralateral (left) side measured by qRT-PCR using GAPDH as a reference gene. The experiment represents apical, central axillary, and pectoral lymph nodes obtained from a single animal 24 h post-injection. Three technical replicates were performed for each sample. (D) Expression of N332-GT2 mRNA in sorted PBMCs at 4 h and 24 h post-injection measured by qRT-PCR using GAPDH as a reference gene. These experiments represent PBMCs obtained from four animals. Three technical replicates were performed for each sample. Statistical significance was determined by Kruskal-Wallis test. ns, p > 0.05; ∗p < 0.0332; ∗∗p < 0.0021; ∗∗∗p < 0.0002; ∗∗∗∗p < 0.0001. All data show means ± SEM.