(T1530-06-36) DoE-Based Optimization of RNA Nanosac Lyophilization for Cold-Chain-Free Storage and Transportation

Purpose: Nanosac is a polyphenol nanocapsule, developed for systemic delivery of RNA therapeutics. Nanosac is produced by coating cationized mesoporous silica nanoparticles (MSNs) with siRNA and polydopamine sequentially and removing the MSN core. Polydopamine nanocapsule is negatively charged at physiological pH and binds serum albumin to interact with cells. Its flexibility helps Nanosac navigate through the endothelium and tumor environment, delivering siRNA to solid tumors via systemic administration.¹ The objective of this study is to optimize the lyophilization of Nanosac for its cold-chain free storage and transportation. We employ different freezing rates and lyoprotectants, such as sucrose, trehalose, mannitol, and their combinations. The lyoprotectant formulation was optimized using a response surface Design of Experiments (DoE) approach, with the size of reconstituted Nanosac as the response. Preliminary screening identified quench freezing and a combination of sucrose and mannitol as preferable conditions for Nanosac lyophilization. The response surface DoE design was implemented to determine the optimal solid content% (w/v) (concentration of solids in the formulation, including Nanosac and lyoprotectants, divided by the volume of the dispersion) and the ratio of sucrose to mannitol for the final lyophilization formulation with and without siRNA payload.

Methods: For the preliminary screening, two combinations of lyoprotectants were tested: sucrose and mannitol (sucrose/mannitol), and trehalose and mannitol (trehalose/mannitol). For both combinations, the solid content was fixed at 1.5% (w/v), with the ratio of sucrose or trehalose to mannitol ranging from 25% to 75%. For the freezing rate screening, Nanosac was frozen by quench freezing with liquid nitrogen or shelf-ramped freezing at a rate of -1°C per minute, with sucrose/mannitol as a lyoprotectant. The preliminary screening found the sucrose/mannitol combination, with sucrose as the major component, and quench freezing to be optimal. For further optimization, the solid content% (w/v) and the sucrose to mannitol ratio were varied according to the response surface DoE. The samples were lyophilized using a Labconco manifold freeze-dryer for 72 hours and reconstituted in water for 15-20 seconds. The reconstituted Z-average was measured using a Malvern Zetasizer Ultra (Red). The response surface was plotted using JMP Pro 16 software.

Results: The preliminary screening of lyoprotectants was conducted with a fixed solid content of 1.5% (w/v) and varying ratios of sucrose or trehalose to mannitol (25%, 50%, and 75%). With the sucrose or trehalose ratio of 25% or 50%, the reconstituted Nanosac aggregated significantly, resulting in size ranging from 10,000 to 30,000 nm (Figure 1a). On the other hand, with 75% sucrose or trehalose, the reconstituted Nanosac showed smaller Z-averages of 267.6 ± 12.9 nm or 321.6 ± 14.6 nm respectively, with sucrose/mannitol yielding smaller size than the trehalose/mannitol combination (p < 0.05, t-test). A potential explanation is that mannitol can crystallize and damage the structure of the Nanosac, resulting in particle aggregation. Therefore, we chose the sucrose/mannitol combination as the preferred formulation. Nanosac frozen with liquid nitrogen quench freezing was dispersed in water better than those with shelf-ramped freezing at -1°C per minute (Figure 1b). This suggests that, during shelf-ramped freezing, water may have formed large ice crystals, causing a freeze concentration of Nanosac. In contrast, quench freezing may have produced smaller ice crystals, undergoing less freeze concentration and aggregation. The lyoprotectant formulation was further optimized using a response surface DoE design, where the solid content% (w/v) and the sucrose/mannitol ratio were varied, with and without a siRNA payload (Table 1). Irrespective of the presence of siRNA, the Z-averages showed a saddle surface, with the smallest particle size observed at 87.5% sucrose (Figures 2a and 2b). This suggests that a 75% sucrose to 25% mannitol ratio was insufficient for protecting Nanosac during lyophilization. Conversely, 100% sucrose (no mannitol) also resulted in an increased particle size, highlighting the crucial but unclear role of mannitol. Additionally, increasing the solid content resulted in smaller particle sizes upon reconstitution, suggesting that higher viscosity of lyoprotectant solution limited particle mobility and aggregation. Overall, the smallest particle size upon reconstitution was achieved with a sucrose/mannitol ratio of 87.5:12.5 and a total solid content of 4.6 % (w/v) for both blank and siRNA-loaded Nanosac.

Conclusion: The choice of lyoprotectants, the composition, and the quantity, as well as the freezing rate during the lyophilization process are crucial for maintaining the size of lyophilized Nanosac. A response surface DoE design helps optimize the formulation for Nanosac lyophilization. The optimized formulation was 87.5% sucrose and 12.5% mannitol at an overall solid content of 4.6 % (w/v) for both blank and siCD47-containing Nanosac.

References: 1. Kim, Hyungjun, et al. "Nanosac, a noncationic and soft polyphenol nanocapsule, enables systemic delivery of siRNA to solid tumors." ACS nano 15.3 (2021): 4576-4593.

Acknowledgements: The authors acknowledge the support provided by the Ronald W. Dollens Graduate Scholarship in the Life Sciences, the Purdue Research Foundation Proof of Concept Fund, and the Purdue Institute for Drug Discovery and the Purdue Institute of Inflammation, Immunology, and Infectious Disease as part of the Drug Discovery in Infectious Disease Training Program.

Table1. Response surface DoE design for lyoprotectant combination (sucrose/mannitol) formulations.

Figure 1. Preliminary Screening of Lyoprotectants and Freezing Conditions. (a) Screening of combinations of sucrose and mannitol (sucrose/mannitol) or trehalose and mannitol (trehalose/mannitol) for lyophilization of Nanosac. The combinations in varying ratios (sucrose or trehalose%) were used as a lyoprotectant, and their effects on the size of the reconstituted Nanosac were determined. The dashed line represents the Z-average of freshly prepared Nanosac (180 nm). All groups had a solid content of 1.5 % (w/v) and were quench frozen in liquid nitrogen and lyophilized with a Labconco freeze dryer. (b) Comparison of quench freezing in liquid nitrogen and shelf-ramped freezing at a rate of -1°C per minute. The dashed line represents the Z-average of freshly prepared Nanosac (183 nm). Sucrose/mannitol mixtures in indicated ratios were used as a lyoprotectant, with a solid content of 2.75 % (w/v).

Figure 2. Response surface plots of the Z-averages for (a) blank Nanosac and (b) siRNA-loaded Nanosac, lyophilized with varying solid content% (w/v) and sucrose ratio (%) in the sucrose/mannitol combination.