Comprehensive Optimization of a Freeze-Drying Process Achieving Enhanced Long-Term Stability and In Vivo Performance of Lyophilized mRNA-LNPs

Abstract

The success of mRNA vaccines against SARS-CoV-2 has prompted interest in mRNA-based pharmaceuticals due to their rapid production, adaptability, and safety. Despite these advantages, the inherent instability of mRNA and its rapid degradation in vivo underscores the need for an encapsulation system for the administration and delivery of RNA-based therapeutics. Lipid nanoparticles (LNPs) have proven the most robust and safest option for in vivo applications. However, the mid- to long-term storage of mRNA-LNPs still requires sub-zero temperatures along the entire chain of supply, highlighting the need to develop alternatives to improve mRNA vaccine stability under non-freezing conditions to facilitate logistics and distribution. Lyophilization presents itself as an effective alternative to prolong the shelf life of mRNA vaccines under refrigeration conditions, although a complex optimization of the process parameters is needed to maintain the integrity of the mRNA-LNPs. Recent studies have demonstrated the feasibility of freeze-drying LNPs, showing that lyophilized mRNA-LNPs retain activity and stability. However, long-term functional data remain limited. Herein, we focus on obtaining an optimized lyophilizable mRNA-LNP formulation through the careful selection of an optimal buffer and cryoprotectant and by tuning freeze-drying parameters. The results demonstrate that our optimized lyophilization process maintains LNP characteristics and functionality for over a year at refrigerated temperatures, offering a viable solution to the logistical hurdles of mRNA vaccine distribution.

Keywords: freeze-drying/lyophilization; mRNA-LNPs; thermostable vaccines.

Figure 1

(A) Shelf temperature and vacuum setpoints used during the freeze-drying cycle. (B–D) Comparison of the physicochemical parameters of fresh (4 °C) and freeze-dried LNPs in two different buffers (PBS or Tris 5 mM) using sucrose or maltose as lyoprotectants. (B) Encapsulation efficiency of mRNA (%). (C) Particle size obtained by DLS and (D) the Z potential values. Dashed line indicates separation of the PBS and Tris groups. (E,F) Transfection efficiency of freeze-dried LNPs normalized with control LNPs in HeLa and 293T (F) cells. Vertical dashed line indicates separation of the PBS and Tris groups, horizontal dashed lines indicate the basal transfection efficiency respective to control non-lyophilized LNPs. Data are presented as the geometric mean of at least three independent replicates, and error bars indicate the standard deviation (±SD).

Figure 2

(A) Shelf temperature and vacuum setpoints used during optimized freeze-drying. (B) Transfection efficiency of freeze-dried LNPs normalized with control samples at 4 °C in 293T cells. Horizontal dashed lines indicate the basal transfection efficiency respective to control non-lyophilized LNPs. (C,D) Comparison of the physicochemical parameters of fresh (4 °C) and freeze-dried LNPs obtained by the initial and modified methods, in 20% sucrose or 20% maltose. (B) Particle size obtained by DLS. (C) Encapsulation efficiency of mRNA (%). (E) Average luminescence radiance (p/s) and bioluminescence images of mice treated with mRNA-LNPs. Mice were intramuscularly injected with LNPs at a dose of 1 µg of LUC-encoding mRNA/animal, and bioluminescence images were taken four hours post inoculation using the IVIS Lumina XRMS Imaging System. For graphs (B–D), data are presented as the geometric mean of at least three independent replicates, and error bars indicate the standard deviation (±SD). For the graphs in (E), data are the geometric mean of independent duplicates (two mice injected per sample), and error bars indicate the standard deviation (±SD). * p < 0.05, ** p < 0.01, *** p< 0.005, ns = not significant.

Figure 3

Physicochemical characterization of frozen (−80 °C) and freeze-dried LNPs stored at 4 °C, 25 °C, and 37 °C for up to 60 weeks. (A–F) Analysis of physicochemical properties: (A) particle size, (B) polydispersity index, (C) Z potential, (D) encapsulation efficiency (%) of mRNA, (E) total mRNA concentration obtained by RiboGreen assay, and (F) mRNA integrity (%) obtained by capillary electrophoresis. (G). Cryo-TEM images of liquid and lyophilized LNPs freshly prepared or stored at 4 °C, 25 °C, and 37 °C for 60 weeks. For each image, size distribution histograms with Gaussian fitting curves (solid lines) are depicted, obtained from the Cryo-TEM images (N = 150/image).

Figure 4

Functional results of frozen (−80 °C) and freeze-dried LNPs stored at 4 °C, 25 °C, and 37 °C for up to 60 weeks. (A,B) Transfection efficiency of the indicated LNPs in HeLa (A) and 293T (B) cells. (C,D) Average luminescence radiance (C) and bioluminescence images (D) of mice treated with mRNA-LNPs. Mice were intramuscularly injected with LNPs at a dose of 1 µg of LUC-encoding mRNA/animal, and bioluminescence images were taken four hours post inoculation using the IVIS Lumina XRMS Imaging System (software version 4.8.2).