Optimizing Microfluidic Channel Design with Tilted Rectangular Baffles for Enhanced mRNA-Lipid Nanoparticle Preparation

Abstract

RNA therapeutics represent a pivotal advancement in contemporary medicine, pioneering innovative treatments in oncology and vaccine production. The inherent instability of RNA and its delivery challenges necessitate the use of lipid-based nanoparticles as crucial transport vehicles. This research focuses on the design, simulation, and optimization of various microfluidic channel configurations for fabricating poly(dimethylsiloxane) (PDMS) microfluidic chips, aimed at producing lipid nanoparticles (LNPs) encapsulating green fluorescent protein mRNA (GFP mRNA). Aiming for high mixing efficiency and acceptable pressure drop suitable for scale-up, we designed and improved multiple microfluidic channels featuring flow focusing and diverse tilted rectangular baffle structures via computational fluid dynamics (CFD). Simulation results indicated that baffle angles ranging from 70 to 90° exhibited similar mixing efficiencies at different total flow rates, with pressure drops increasing alongside the baffle angle. Additionally, increasing the baffle length at a fixed angle of 70° not only improved mixing efficiency but also increased the pressure drop. To validate these findings, PDMS microfluidic chips were fabricated for all designs to prepare empty LNPs. The baffle structure with a 70° angle and 150 μm length was identified as the best configuration based on both simulation and experimental results. This optimal design was then used to prepare LNPs with varying GFP mRNA concentrations, demonstrating that an N/P ratio of 5.6 yielded the highest transfection efficiency from in vitro experiments. This work not only advances the production of lipid-based nanoparticles through microfluidics but also provides a scalable and reproducible method that can potentially enhance the clinical translation of RNA therapeutics.

Keywords: CFD simulations; lipid nanoparticles; microfluidic; transfection.

1

Baffle structure microfluidic design and mixing results: (a) Geometry of the microfluidic channels design. (b) Detailed design of the mixing channel unit. (c) PDMS microfluidic chip. (d) Experiment setup. (e) Velocity streamline plot simulation result of baffle structure with a 70° angle and 150 μm length at 1200 μL/min. (f) Mixing results using PDMS microfluidic chip with 70° angle and 150 μm length using rhodamine B at 1200 μL/min. (g) Concentration simulation results of baffle structure with a 70° angle and 150 μm length at 1200 μL/min.

2

Mixing index and pressure drop results for different structural designs: Mixing indexes for various baffle angles (a) at a total flow rate of 300 μL/min; (b) at a total flow rate of 600 μL/min; (c) at a total flow rate of 900 μL/min; and (d) at a total flow rate of 1200 μL/min. (e) Pressure drop at Inlet 1 for different baffle angles. (f) Pressure drop at Inlet 2 for different baffle angles. Mixing indexes for various baffle lengths with a fixed 70° angle (g) at a total flow rate of 300 μL/min; (h) at a total flow rate of 600 μL/min; (i) at a total flow rate of 900 μL/min; and (j) at a total flow rate of 1200 μL/min. (k) Pressure drop at Inlet 1 for different baffle lengths with a fixed 70° angle. (l) Pressure drop at Inlet 2 for different baffle lengths with a fixed 70° angle.

3

Lipid nanoparticle size and PDI results for different structural designs: Size and PDI results for various baffle angles (a) at a total flow rate of 300 μL/min; (b) at a total flow rate of 600 μL/min; (c) at a total flow rate of 900 μL/min; and (d) at a total flow rate of 1200 μL/min. Size and PDI results for various baffle lengths (e) at a total flow rate of 300 μL/min; (f) at a total flow rate of 600 μL/min; (g) at a total flow rate of 900 μL/min; and (h) at a total flow rate of 1200 μL/min.

4

LNPs with GFP mRNA and cell experiments: (a) Preparation process; (b) size and PDI results of LNPs with different GFP mRNA concentrations; (c) zeta potential results of LNPs with different GFP mRNA concentrations; (d) encapsulation efficiency results of LNPs with different GFP mRNA concentrations; (e) fluorescence microscopy images of GFP expression captured with a 4× objective. Scale bars represent 200 μm. (f) Integrated fluorescence intensity related to GFP expression, semiquantified using ImageJ software. (g) Cell viability of HEK cells tested using the alamarBlue assay.