The role of gel-phase domains in electroporation of vesicles

Abstract

Transient permeabilisation of the cell membrane is a critical step to introduce drugs or DNA into living cells, yet challenging for both biological research and therapeutic applications. To achieve this, electroporation (or electropermeabilisation) has become a widely used method due to its simplicity to deliver almost any biomolecule to any cell type. Although this method demonstrates promise in the field of drug/gene delivery, the underlying physical mechanisms of the response of the heterogeneous cell membrane to strong electric pulses is still unknown. In this study, we have investigated the role of gel-phase lipids in the electroporation of binary giant unilamellar vesicles (GUVs), composed from DPPC (gel-phase) and DPhPC (fluid-phase) lipids (molar ratio 8:2 and 2:8). We have observed that the exposure to electric pulses leads to expel of fluid-phase lipids and concomitant decrease in GUV size, whereas the gel-phase domains become buckled. Based on experiments on pure fluid-phase and gel-phase GUVs, we have found that fluid-phase lipids can be expelled by electrical forces and the highly viscous gel-phase lipids cannot. Moreover, our analyses suggest that pore formation occurs primarily in fluid-phase domains and that the pore size is similar in all GUVs containing fluid-phase lipids, irrespective of the gel-phase percentage.

Figure 1

The experimental setup and analysis for the electroporation of the GUVs. (A) A picture of the electroporation setup. (B) A schematic of the sucrose-filled GUV in the glucose environment, in between the electrodes with a distance ranging between 0.6 and 1 mm. The drawing is not to scale. (C) Bright field image of a sucrose-filled GUV in a glucose environment. (D) The analysis of the GUVs: the contour is tracked, after which the area, A, and the perimeter, P, are determined. From these, the normalised area and the form factor are calculated. The form factor is 1 for a circular GUV, and less than one for a non-circular GUV. The scale bar is 10 μm.

Figure 2

The response of the fluid-phase (in blue) and gel-phase (in pink) GUVs to the electric field. (A) and (B) show, respectively, the averaged normalised area and form factor plotted versus the electric field of the pulse. The fluid-phase data is averaged over 11 different GUVs, with radii between 11 and 28 μm. The gel-phase data is averaged over 12 different GUVs, with radii between 10 and 22 μm. The error bars in the plots represent the standard deviation. The dotted lines represent the least-squares fit with a sigmoid curve. (C) and (D) are snapshots of the fluid-phase and gel-phase GUVs showing the gradual contrast loss. (C) From left to right: before pulse application, more than 100 seconds after the tenth pulse at 52 V/mm and after the twentieth pulse at 100 V/mm. (D) From left to right: before pulse application, more than 100 seconds after the fourth pulse at 74 V/mm and after the twenty-fourth pulse at 372 V/mm. (E) and (F) are snapshots of the fluid-phase and gel-phase GUVs showing the complete contrast after one of the pulses. (E) From left to right: before pulse application, more than 100 seconds after the second pulse at 89 V/mm and after the third pulse at 97 V/mm. (F) From left to right: before pulse application, more than 100 seconds after the third pulse at 595 V/mm and after the fourth pulse at 744 V/mm. The scale bar in all images is 10 μm.

Figure 3

A DPhPC GUV (A) before pulse application and two representative examples of the shrinkage mechanism of the DPhPC GUVs (B) showing tubule formation (C) and vesicle formation. In these examples, the GUVs have been exposed to 2–8 pulses, duration 5 ms, amplitude 90 V/mm. The majority of the GUVs have exhibited vesicle formation. The scale bar in all images is 5 μm, the cathode is on the left side and the anode on the right side of the GUVs.

Figure 4

The fluorescence images of the binary GUVs before electroporation. (A) A schematic drawing of the three different focal planes (bottom, middle and top planes) of the GUVs, not drawn to scale. (B) Two examples of the 2:8 DPPC:DPhPC GUVs, showing different gel-phase domains. The GUVs are visualised by use of fluorescence microscopy, where the dark parts represent the gel-phase domains, and the bright parts the fluid-phase or mixed domains. (C) The two different categories of the 8:2 DPPC:DPhPC GUVs. Left: a homogeneous-GUV, showing no visible domains. Right: a domain-GUV, showing a uniform distribution of dark gel-phase domains on the surface of the GUV. The scale bar is 10 μm.

Figure 5

The normalised area and the form factor of the 2:8 DPPC:DPhPC GUVs. (A) The averaged normalised area and (B) the form factor of the 2:8 DPPC:DPhPC GUVs. The data is averaged over 13 different GUVs, with radii between 11 and 26 μm. The dotted lines represent the least-squares fit with a sigmoid curve. The error bars in the plots represent the standard deviation. (C) Images of a GUV before and more than 100 seconds after the second pulse at 445 V/mm pulse. (D) Fluorescence images of the same GUV. The fluorescence image after the pulse shows that the buckled patch is located at the gel-phase domain (the dark patch). (E) Bright field and fluorescence images of a GUV before the pulse, during the pulse when a macropore is observed, and after the third pulse at 29 V/mm. The fluorescence images are only captured before and after the pulse. The scale bar is 10 μm.

Figure 6

The response of the 8:2 DPPC:DPhPC GUVs to increasing electric pulses. (A) The averaged normalised area and (B) the form factor of the homogeneous- and domain-GUVs. The data is averaged over 13 different GUVs, with radii between 11 and 28 μm, where 7 GUVs contain domains (radius between 11 and 28 μm) and 6 GUVs have fully mixed phases (radius between 14 and 22 μm). The dotted lines represent the least-squares fit with a sigmoid curve. (C) The images of the domain- and the homogeneous-GUVs before and more than 100 seconds after the first pulse at 445 V/mm (domain-GUV) or the third pulse at 890 V/mm (homogeneous-GUV), respectively. (D) Bright field and fluorescence images of the homogeneous-GUV shown in (C) before the pulse, during the pulse when a macropore is observed, and after the pulse. The fluorescence images are only captured before and after the pulse. The scale bar is 10 μm.