Liposome production by microfluidics: potential and limiting factors

Abstract

This paper provides an analysis of microfluidic techniques for the production of nanoscale lipid-based vesicular systems. In particular we focus on the key issues associated with the microfluidic production of liposomes. These include, but are not limited to, the role of lipid formulation, lipid concentration, residual amount of solvent, production method (including microchannel architecture), and drug loading in determining liposome characteristics. Furthermore, we propose microfluidic architectures for the mass production of liposomes with a view to potential industrial translation of this technology.

Figure 1. Schematic representation of the process of liposome formation.

Panel A shows a schematic representation of a microfluidic device, namely a MHF microchip (#chip1-MHF), and the process of liposome (SUV) self-assembly. Panel B shows a schematic representation of the ethanol injection procedure. Panel C illustrates the geometrical characteristics of the chips employed for the microfluidic experiments. #chip1-MHF comprises three inlet microchannels with 30° intersection angle between each other, and a mixing microchannel having width, depth and length of 150 μm, 100 μm and 30 mm, respectively. #chip2-YJ comprises two inlet microchannels with 120° intersection angle between each other, and a mixing microchannel having width, depth and length of 320 μm, 320 μm and 66 mm, respectively. #chip3-MP comprises three inlet microchannels with 90° intersection angle between each other, and a mixing microchannel with three pillar mixing elements, having width, depth and length of 150 μm, 150 μm and 65 mm, respectively. Teflon^TM tubes were used to connect the microfluidic platform with syringes. The volumetric flow rate was controlled using syringe pumps.

Figure 2

Effect of the variation of lipid concentration (A), ethanol content (B) and simultaneous variation of both parameters (C) on the dimension of liposomes produced by ethanol injection. For comparison, data relative to liposomes prepared by MHF microfluidics are also reported in panel C (dashed line). Data correspond to the Z-average, determined by DLS. Liposomes were constituted of PC/cholesterol 4.0−0.4 mM, and data represent the mean of three independent samples, measured in triplicate ± S.D.

Figure 3

Effect of the variation of lipid composition on the size (upper part, panel A) and dispersity (upper part, panel B) of liposomes produced by MHF microfluidics. Liposomes were constituted of PC/DDAB 9.0−1.0 mM (filled circles, dashed line) or PC/cholesterol 9.0−1.0 mM (open circles, plain line). Data correspond to the Z-average, determined by DLS, and are reported as the mean of three independent samples, measured in triplicate ± S.D. In the lower part, cryo-TEM and macroscopic aspect (insets) of empty PC/cholesterol (A) and PC/DDAB (C) are reported. For comparison, images of the corresponding ivermectin loaded liposomes are also reported (B,D). Bar corresponds to 100 nm.

Figure 4. Comparative analysis of liposomes produced by microfluidic technique with different microchips.

Data refer to Z-average of liposomes produced by #chip1-MHF (circles), #chip2-YJ (hexagons) or #chip3-PM (triangles). Liposomes were constituted of PC/DDAB 9.0−1.0 mM, produced at the indicated FRR and TFR = 37.50 μl/min. Data are reported as the mean of three independent samples, measured in triplicate ± S.D.

Figure 5. Effect of the encapsulation of 0.1 mM ivermectin on the size of microfluidic produced liposomes (filled circles).

For comparison, the size of empty (water-filled) liposomes is also reported (open circles). Liposomes were produced by #chip1-MHF at TFR = 37.50 μl/min and FRR = 30. Data are reported as the mean of three independent samples, measured in triplicate ± S.D.

Figure 6

(A) Geometrical characteristics of easy-to-build chips for liposome production. (B) Schematic of #chip4-OFF3, and potential parallelized network for mass production of liposomes. (C) Schematic of #chip5-MHF-LC and related fabrication method developed in house (μMi-REM).

Figure 7. Size and size distribution of liposomes produced by high-throughput microfluidic architecture with scaled up channel dimension (#chip5-MHF-LC).

(A,B) Dependence of liposome mean size (A) and dispersity index (B) on the flow rate ratio (FRR, ranging from 5 to 100) at a fixed TFR of 6 ml/min. (C,D) Dependence of liposome mean size (A) and dispersity index (B) on the total flow rate (TFR, ranging from 3 to 18 ml/min) at a fixed FRR of 100.