Lipid nanoparticles (LNPs) have been widely utilized as carriers for nucleic acid drug delivery; however, their inherent heterogeneity impedes the accurate characterization of physicochemical and biological properties. Conventional analytical methods are inherently limited in resolving such heterogeneity. This study employed sucrose density gradient centrifugation (S-DGC) to separate LNP subpopulations of varying densities. It investigated the effects of lipid formulation parameters-including nitrogen-to-phosphorus (N/P) ratio, lipid composition, polyethylene glycol (PEG) concentration-and microfluidic preparation conditions (flow rate) on LNP heterogeneity and biological functionality. Formulation stability was assessed via freeze-thaw testing. The results demonstrated that S-DGC could effectively separate LNP subpopulations with divergent densities and physicochemical characteristics. Changes in the N/P ratio and lipid composition significantly modulate subphase distribution and properties. When cholesterol (Chol) and distearoylphosphatidylcholine (DSPC) are absent from the formulation, LNPs aggregate in the low-density layer (0-10% sucrose density layer). The concentration of PEGylated lipids serves as a critical regulatory factor. When the concentration increased from 0.5% to 2.5%, the LNP particle size decreased from approximately 202 nm to 118.7 nm. Furthermore, the S-DGC profile indicates that LNP transitions from an aggregated low-density distribution to a uniformly dense subpopulation concentrated within the 0-20% sucrose layer, where transfection efficiency is optimal. In freeze-thaw stability assessment, unprotected LNP exhibited a drastic decline in encapsulation efficiency to 5.3% after three freeze-thaw cycles at -80 °C. The S-DGC diagram revealed aggregation in the 20-30% high-density region. However, adding 5% sucrose maintained encapsulation efficiency above 96%. This study confirms that the S-DGC analytical platform serves as a potent tool for resolving LNP heterogeneity and correlating formulation structure with function. Based on these findings, this study contends that during the early prescription development of LNP-based nucleic acid therapeutics, formulation screening should not be confined to meeting overall particle size and encapsulation rate targets. Instead, S-DGC can be employed to proactively identify and minimize ineffective subpopulations (such as particles distributed in extremely high or low density zones), thereby enhancing product quality uniformity and predictability from the outset of R&D.
