(W1030-03-16) Predicting the Aggregation Behavior of Liposomes Using Thermal Analysis

Purpose: Aggregation of liposomal drugs is a critical stability attribute and is affected by various parameters such as lipid composition, payload, and charge. The goal of this study is to demonstrate that it is possible to predict aggregation behavior in liposomes using solid-state DSC.

Methods: Three liposome formulations were made using a lyoprotectant (sucrose) at pH 7 in Histidine buffer; liposomes without protein, and liposomes with entrapped protein A (L-A) or with entrapped protein B (L-B). All three liposomes were in the same formulation and were made using the same processes (lipid-blend, formulation, extrusion, and lyophilization cycle). The only difference was the presence/absence of protein and the molecular weight of the protein entrapped in liposomes. Protein A and protein AB had Mw of 76 and 134 KDa, respectively. All freeze-dried liposomes were analyzed by solid-state differential scanning calorimetry to determine melting behavior of lipids and sugar. After reconstitution, zeta potential and average particle size of liposomes were characterized by Malvern Zetasizer and Dynamic Light Scattering (DLS), respectively. The effect of storage temperature and time on aggregation behavior of the liposomes were evaluated at time zero (t0) and after 0.5, 1- and 3-months storage at 5°C, 25°C and 40°C. The aggregation rate constant at each temperature was calculated from the slope of linear regression lines of plots of average hydrodynamic diameter (n=3) of liposomes against storage time at each temperature.

Results: The DSC thermograms (Table 1) reveal that all liposomes showed two endothermic peaks and an exothermic peak, corresponding to the melting of the lipid-phase (Tm1) and lyoprotectant sucrose (Tm2) and re-crystallization of sucrose (Tc), respectively. A correlation was found between the molecular weight of the entrapped protein, and the melting temperatures of lipids (Fig 1a), and sugar (Fig 1b), suggesting that the larger the protein (Mw), the lower the melting temperature of liposomes and sugar in the solid-state. No aggregation was observed for liposomes without protein across 5°C to 40°C, as expected (these liposomes had a phase transition temperature at 60°C (±4°C) and no change in their size was expected by time at lower temperatures than 50°C). A correlation was found between the molecular weights of entrapped proteins and the aggregation rate of liposomes (Fig 3a). Aggregation rate for liposomes increased as the molecular weight of the entrapped protein increased, because the protein interacted with the lipids, shifting the melting temperature of lipids and sucrose towards lower temperatures and causing the melting onset of the lipids to decrease from 60°C (±4°C) in the absence of proteins to 53°C (±4°C) and 47°C (±2°C) in the presence of proteins A and B, and as a function of protein molecular weight (76 KDa and 134 KDa, respectively). The aggregation rate constants correlated well with the melting temperatures of liposomes obtained by DSC (Fig. 2b), showing that the lower the melting temperature of the liposomes, the greater the aggregation tendency. Liposomes containing proteins (L-A and L-B) did not aggregate significantly at 5°C and 25°C (Table 1) as expected (these storage temperatures were far below the melting temperatures of these liposomes). The aggregation rate for these samples were around 10 nm/month at both 5°C and 25°C, where as at 40°C, there was a sudden increase in aggregation rate for L-A and L-B to 40 nm/month and 110 nm/month, respectively.

Conclusion: This study shows that it is possible to predict aggregation tendency of liposomes at 5°C to 40°C temperatures from initial DSC studies at time zero (by measuring melting temperature of lipids). Additionally, in order to enhance thermal stability, it is better to entrap smaller proteins in the liposomes, or no protein at all.  These results suggest that solid state characterization of freeze-dried liposomes at time zero can predict the stability and optimum storage temperature of liposomes.


Figure 1. Average melting temperatures for lipids phase (a) and sucrose (b) in L, L-A, and L-B, as a function of Mw of proteins in the liposomes. All liposomes have the same lipid content and charge. Liposomes containing protein also have the same protein content. The solid line shows melting temp for pure crystalline sucrose.


Figure 2. Aggregation rate of liposomes at 40°C as a function of molecular weight of entrapped proteins (a) and liposome melting temperature in the solid-state at time zero (t0) as measured by DSC.


Table 1. Characterization data