Microbubbles-Assisted Ultrasound Triggers the Release of Extracellular Vesicles

Abstract

Microbubbles-assisted ultrasound (USMB) has shown promise in improving local drug delivery. The formation of transient membrane pores and endocytosis are reported to be enhanced by USMB, and they contribute to cellular drug uptake. Exocytosis also seems to be linked to endocytosis upon USMB treatment. Based on this rationale, we investigated whether USMB triggers exocytosis resulting in the release of extracellular vesicles (EVs). USMB was performed on a monolayer of head-and-neck cancer cells (FaDu) with clinically approved microbubbles and commonly used ultrasound parameters. At 2, 4, and 24 h, cells and EV-containing conditioned media from USMB and control conditions (untreated cells, cells treated with microbubbles and ultrasound only) were harvested. EVs were measured using flow cytometric immuno-magnetic bead capture assay, immunogold electron microscopy, and western blotting. After USMB, levels of CD9 exposing-EVs significantly increased at 2 and 4 h, whereas levels of CD63 exposing-EVs increased at 2 h. At 24 h, EV levels were comparable to control levels. EVs released after USMB displayed a heterogeneous size distribution profile (30-1200 nm). Typical EV markers CD9, CD63, and alix were enriched in EVs released from USMB-treated FaDu cells. In conclusion, USMB treatment triggers exocytosis leading to the release of EVs from FaDu cells.

Keywords: drug delivery; endocytosis; exocytosis; exosomes; microvesicles; sonoporation.

Figure 1

Comparison of the level of mean fluorescence intensities (MFI) of bead-captured extracellular vesicles (EVs) exposing CD9 and CD63 measured using flow cytometry. MFI CD9 and MFI CD63 were corrected from their IgG1 controls. Percentages of MFI CD9 (A) and CD63 (B) of EVs derived from conditioned media of FaDu cells collected at 2, 4 and 24 h after microbubbles-assisted ultrasound (USMB), ultrasound (US), and microbubbles (MB) treatments were calculated by comparing those with the untreated counterparts. Bars represent mean of three independent experiments + standard deviation. Statistical analysis was performed using two-way ANOVA followed by Tukey’s test. p value < 0.05 was considered significant (*).

Figure 2

Representative EM images showing immunogold labeling of EVs using anti-CD9 or -CD63 antibody and lactadherin. Both antibodies were coupled with 6-nm gold, whereas lactadherin was coupled with 10-nm gold. EVs were derived from conditioned media of FaDu cells collected at 4 h after microbubbles-assisted ultrasound (USMB) treatment. EVs were stained positive by anti-CD9 antibody (A), anti-CD63 antibody (B,H white arrows, and I), or lactadherin, a surface marker of the negatively charged phospholipid, phosphatidylserine (PS+; C,H black arrows, and J) only. Some were double stained by anti-CD9 antibody and lactadherin (D,G) or anti-CD63 antibody and lactadherin (E). EVs without gold label were also detected (F). Large EVs with an apparent diameter of 500–1200 nm were also detected (G,I,J).

Figure 3

Quantification of EVs detected using EM in conditioned media collected at 4 h from cells after microbubbles-assisted ultrasound (USMB), ultrasound (US), or microbubbles (MB) treatments, and from untreated cells. Ten images per grid recorded at 60,000 × magnification were used for analysis. Total number of EVs from USMB treatment was significantly higher than those from US or MB treatment or untreated (A). Subpopulations of single, double stained, or unstained EVs were quantified (B). EV number represents the mean of four independent experiments + standard deviation. Statistical analysis was performed using one-way ANOVA followed by Tukey’s test. p value < 0.05 is considered significant (*).

Figure 4

Size distribution of EVs detected in conditioned media collected at 4 h from cells after microbubbles-assisted ultrasound (USMB, A), ultrasound (US, B), or microbubbles (MB, C) treatments, and from untreated cells (D). Surface area of EVs was measured using Image J [20] to calculate the apparent diameter of EV (nm). Bin size is 50 nm. EVs were detected from 10 images per grid (60,000× magnification) derived from four independent experiments. Percentage of EVs represents the percentage of mean of EVs at certain size range + standard deviation.

Figure 5

Western blotting on total cell and EV lysates prepared from FaDu cells and their supernatant collected at 4 h after microbubbles-assisted ultrasound (USMB). For western blotting, equal protein amounts of cell and EV lysates were loaded (4.5 µg of total proteins). Detection of the proteins was by using antibodies alix, tsg101, CD9, CD63, calnexin, and actin. Presence of alix (95 kDa), tsg101 (49 kDa), and CD9 (25 kDa) are shown (A). CD63 is shown as a broad band between 30–65 kDa (B). Calnexin (95 kDa) and actin (42 kDa) are present in both cell and EV lysates (C).

Figure 6

Timeline of microbubbles-assisted ultrasound (USMB) experiment of FaDu cells. First, medium containing microbubbles (MB) were added into the FaDu cells grown overnight in the cell culture chamber, and then ultrasound (US) was applied. Medium was directly removed after USMB and cells were washed once. Cells received new medium and incubated for 2, 4, and 24 h. At these time points, cells and conditioned medium were collected.