Scalable mRNA and siRNA Lipid Nanoparticle Production Using a Parallelized Microfluidic Device

Purpose: Microfluidic mixing can produce precise LNPs but is limited in throughput, while bulk mixing is scalable but yields larger, heterogeneous particles with variable potency. A scalable microfluidic device is needed to produce potent RNA-LNPs across discovery and clinical scales.

Hypothesis: A parallelized microfluidic device (PMD) incorporating an array of staggered herringbone micromixers (SHMs) can scale LNP production >100-fold without compromising LNP size, homogeneity, encapsulation, or in vitro/in vivo potency compared with single-channel microfluidic devices.

Aims: Fabricate a PMD with 1×, 10×, and 128× SHM arrays using ladder geometry and flow resistors. - Validate uniform mixing and flow distribution across all channels. - Formulate siRNA- and mRNA-loaded LNPs using the PMD and compare with single-channel microfluidics and bulk mixing. - Evaluate in vitro gene silencing and in vivo siRNA knockdown and mRNA expression.

Delivery system: Platform: Parallelized microfluidic device (PMD) made of PDMS, with 128 SHMs arranged as 4 rows × 32 channels. - Key design: Ladder geometry, upstream flow resistors, in-line filters; operates at 50 psi; production rate up to 18.4 L/h. - LNP components: Ionizable lipid C12-200; phospholipid DOPE (siRNA) or DSPC (mRNA); cholesterol; PEG-lipid. - Payloads: Factor VII siRNA, luciferase siRNA, luciferase mRNA. - Targeting ligand: None. - Route: Intravenous tail-vein injection in mice.

Approach: Mixing validation: FITC-dextran and rhodamine-dextran flowed through single-channel, 10× single-row PMD, and 128× PMD; mixing quantified by channel length for 90% mixing. - In vitro: HeLa cells expressing firefly luciferase; luciferase siRNA-LNPs at 5 nM; IC50 determination. - In vivo: C57BL/6 mice; Factor VII siRNA-LNPs at 0.2 mg/kg; luciferase mRNA-LNPs at 0.2 mg/kg; n = 3–4. - Controls: Bulk-mixed LNPs, single-channel microfluidic LNPs, PBS.

Key methods: Fluorescence microscopy for mixing characterization. - Dynamic light scattering (DLS) for LNP size and size distribution. - Encapsulation efficiency measurements. - Luciferase expression assay in HeLa cells. - Factor VII activity assay in mouse plasma. - IVIS imaging for in vivo luciferase expression. - H&E staining and AST/ALT/ALP for toxicity.

Key results: PMD achieved 18.4 L/h production, >100-fold higher throughput than a single microfluidic channel. - Mixing performance was uniform across 10× and 128× devices; 90% mixing channel length correlated linearly with ln(Pe) across flow rates. - Microfluidic LNPs were small (<85 nm), while bulk-mixed LNPs were large (>120 nm). - In vitro luciferase siRNA knockdown was comparable across all 10 channels and equivalent to single-channel LNPs; IC50 was not significantly different among bulk, single-channel, and PMD formulations. - In vivo Factor VII siRNA: PMD and single-channel LNPs reduced Factor VII activity by >90%, versus only ~20% for bulk-mixed LNPs, corresponding to a 4-fold increase in hepatic gene silencing. - In vivo luciferase mRNA: microfluidic LNPs produced a 5-fold increase in luciferase expression compared with bulk-mixed LNPs. - No significant toxicity differences by AST, ALT, ALP, or liver histology.

Interpretation: The PMD enables scalable, reproducible LNP production at clinically relevant rates without sacrificing the desirable physical properties or potency of microfluidic-generated LNPs. This addresses a key manufacturing bottleneck for RNA therapeutics and vaccines.

Limitations: Only mice were used; no human or large-animal validation. - No long-term stability, repeated-dose, or disease efficacy studies. - PDMS prototyping may require translation to silicon/glass for GMP manufacturing. - Limited to two RNA cargos and specific lipid formulations. - No active targeting ligand. - Small group sizes (n = 3–4) for in vivo studies. - No direct comparison with all scalable manufacturing platforms.