*3.1. Physicochemical Characterization*

Figure 1A presents the number weighted size distributions of indirect DSPC-based microbubbles with and without cholesterol. For microbubbles without cholesterol (0 mol%; *N* = 5) and microbubbles with 12 mol% cholesterol (*N* = 6), the size distributions of batches for 4Pi microscopy and for acoustic experiments are both included, and the mean number (%) per diameter is shown with the standard error of the mean (SEM). For microbubbles with 7, 10, and 14 mol% cholesterol a representative curve is shown from 2 batches, as these types of microbubbles were produced for 4Pi microscopy only. The concentration of microbubbles ranged from 2.78 <sup>×</sup> <sup>10</sup><sup>8</sup> to 1.17 <sup>×</sup> <sup>10</sup><sup>9</sup> microbubbles per mL (Supplemental Table S1). The indirect DSPC-based microbubbles without cholesterol had more particles with diameter >3 µm than all types of microbubbles with cholesterol in the coating. Indirect DSPC-based microbubbles with 32 mol% cholesterol in the coating

were highly unstable, with a concentration too low for measurement of the size distribution. Therefore, indirect DSPC-based microbubbles with 32 mol% cholesterol were not investigated further. highly unstable, with a concentration too low for measurement of the size distribution. Therefore, indirect DSPC-based microbubbles with 32 mol% cholesterol were not investigated further.

number (%) per diameter is shown with the standard error of the mean (SEM). For microbubbles with 7, 10, and 14 mol% cholesterol a representative curve is shown from 2 batches, as these types of microbubbles were produced for 4Pi microscopy only. The concentration of microbubbles ranged from 2.78 × 108 to 1.17 × 109 microbubbles per mL (Supplemental Table S1). The indirect DSPC-based microbubbles without cholesterol had more particles with diameter >3 µm than all types of microbubbles with cholesterol in the coating. Indirect DSPC-based microbubbles with 32 mol% cholesterol in the coating were

*Pharmaceutics* **2021**, *13*, x FOR PEER REVIEW 7 of 22

**Figure 1.** (**A**) Number weighted size distribution, (**B**) number weighted mean diameter (µm), and (**C**) span value of indirect DSPC-based microbubbles with cholesterol in a range from 0 to 14 mol%. In B and C, each symbol represents one batch of microbubbles; jittering was applied to avoid overlapping. The overlaid black lines represent the median and interquartile range. Statistical significance is indicated with \* *p* < 0.05. **Figure 1.** (**A**) Number weighted size distribution, (**B**) number weighted mean diameter (µm), and (**C**) span value of indirect DSPC-based microbubbles with cholesterol in a range from 0 to 14 mol%. In B and C, each symbol represents one batch of microbubbles; jittering was applied to avoid overlapping. The overlaid black lines represent the median and interquartile range. Statistical significance is indicated with \* *p* < 0.05.

Figure 1B shows the mean diameter (µm) of indirect DSPC-based microbubbles without cholesterol and with 7, 10, 12, or 14 mol% cholesterol. Microbubbles with 12 mol% cholesterol had a smaller mean diameter than those without cholesterol (*p* = 0.045). Figure 1C shows the width of the size distributions represented as the span value. The size distributions of microbubbles with 12 mol% cholesterol were more polydisperse than those of microbubbles without cholesterol (*p* = 0.068). Figure 1B shows the mean diameter (µm) of indirect DSPC-based microbubbles without cholesterol and with 7, 10, 12, or 14 mol% cholesterol. Microbubbles with 12 mol% cholesterol had a smaller mean diameter than those without cholesterol (*p* = 0.045). Figure 1C shows the width of the size distributions represented as the span value. The size distributions of microbubbles with 12 mol% cholesterol were more polydisperse than those of microbubbles without cholesterol (*p* = 0.068).

The ligand and lipid phase distribution in the microbubble coating were imaged in indirect DSPC-based microbubbles without cholesterol (*N* = 58), with 7 mol% cholesterol (*N* = 34), with 10 mol% cholesterol (*N* = 40), with 12 mol% cholesterol (*N* = 61), and with 14 mol% cholesterol (*N* = 45). Images were recorded of at least two batches of microbubbles for all formulations, with microbubble diameters ranging from 2.2 µm to 8.7 µm. Typical examples of all formulations are presented in Figure 2. The ligand distribution is shown in the top row, the LE phase in the middle row, and a composite of both channels in the bottom row. Figure 3 shows a quantitative analysis of the 4Pi confocal microscopy images, with the calculated ligand distribution inhomogeneity in Figure 3A and the LC phase relative to the total surface area analyzed per microbubble in Figure 3B. Indirect DSPC-based microbubbles without cholesterol had a homogeneous ligand distribution (Figure 2A, Figure 3A). The inhomogeneity of the ligand distribution can be observed in Figure 2B,C,E, where the ligand is enriched in some areas of the microbubble surface. All The ligand and lipid phase distribution in the microbubble coating were imaged in indirect DSPC-based microbubbles without cholesterol (*N* = 58), with 7 mol% cholesterol (*N* = 34), with 10 mol% cholesterol (*N* = 40), with 12 mol% cholesterol (*N* = 61), and with 14 mol% cholesterol (*N* = 45). Images were recorded of at least two batches of microbubbles for all formulations, with microbubble diameters ranging from 2.2 µm to 8.7 µm. Typical examples of all formulations are presented in Figure 2. The ligand distribution is shownin the top row, the LE phase in the middle row, and a composite of both channels inthe bottom row. Figure <sup>3</sup> shows a quantitative analysis of the 4Pi confocal microscopy images, with the calculated ligand distribution inhomogeneity in Figure 3A and the LCphase relative to the total surface area analyzed per microbubble in Figure 3B. IndirectDSPC-based microbubbles without cholesterol had a homogeneous ligand distribution (Figure 2A, Figure 3A). The inhomogeneity of the ligand distribution can be observed in Figure 2B,C,E, where the ligand is enriched in some areas of the microbubble surface. All indirect DSPC-cholesterol microbubbles had a significantly more heterogeneous ligand distribution compared to those without cholesterol (Figure 2B–E, Figure 3A). Microbubbles with 12 mol% cholesterol had a more homogeneous ligand distribution than those with 7 mol% cholesterol (*p* = 0.070), 10 mol% cholesterol (*p* = 0.040), and 14 mol% cholesterol (*p* < 0.001).

*Pharmaceutics* **2021**, *13*, x FOR PEER REVIEW 8 of 22

terol (*p* < 0.001).

terol (*p* < 0.001).

indirect DSPC-cholesterol microbubbles had a significantly more heterogeneous ligand distribution compared to those without cholesterol (Figure 2B–E, Figure 3A). Microbubbles with 12 mol% cholesterol had a more homogeneous ligand distribution than those with 7 mol% cholesterol (*p* = 0.070), 10 mol% cholesterol (*p* = 0.040), and 14 mol% choles-

indirect DSPC-cholesterol microbubbles had a significantly more heterogeneous ligand distribution compared to those without cholesterol (Figure 2B–E, Figure 3A). Microbubbles with 12 mol% cholesterol had a more homogeneous ligand distribution than those with 7 mol% cholesterol (*p* = 0.070), 10 mol% cholesterol (*p* = 0.040), and 14 mol% choles-

**Figure 2.** Selected views of 4Pi confocal microscopy *y*-stacks of indirect 1,2-distearoyl-*sn*-glycero-3-phosphocholine (DSPC)-based microbubbles without cholesterol (**A**,**F**,**K**, diameter (*d*) = 6.4 µm, liquid condensed (LC) phase area 35%), with 7 mol% cholesterol (**B**,**G**,**L**, *d* = 5.6 µm, LC phase area 22%), with 10 mol% cholesterol (**C**,**H**,**M**, *d* = 6.1 µm, LC phase area 22%), with 12 mol% cholesterol (**D**,**I**,**N**, *d* = 3.6 µm, LC phase area 7%), and with 14 mol% cholesterol (**E**,**J**,**O**, *d* = 5.8 µm, LC phase area 22%) in the phospholipid coating. Images show the ligand distribution (**A**–**E**; Oregon Green 488), liquid expanded (LE) phase (**F**–**J**; rhodamine-DHPE), and composite view (**K**–**O**). Scale bars are 1 µm. **Figure 2.** Selected views of 4Pi confocal microscopy *y*-stacks of indirect 1,2-distearoyl-*sn*-glycero-3-phosphocholine (DSPC) based microbubbles without cholesterol (**A**,**F**,**K**, diameter (*d*) = 6.4 µm, liquid condensed (LC) phase area 35%), with 7 mol% cholesterol (**B**,**G**,**L**, *d* = 5.6 µm, LC phase area 22%), with 10 mol% cholesterol (**C**,**H**,**M**, *d* = 6.1 µm, LC phase area 22%), with 12 mol% cholesterol (**D**,**I**,**N**, *d* = 3.6 µm, LC phase area 7%), and with 14 mol% cholesterol (**E**,**J**,**O**, *d* = 5.8 µm, LC phase area 22%) in the phospholipid coating. Images show the ligand distribution (**A**–**E**; Oregon Green 488), liquid expanded (LE) phase (**F**–**J**; rhodamine-DHPE), and composite view (**K**–**O**). Scale bars are 1 µm. **Figure 2.** Selected views of 4Pi confocal microscopy *y*-stacks of indirect 1,2-distearoyl-*sn*-glycero-3-phosphocholine (DSPC)-based microbubbles without cholesterol (**A**,**F**,**K**, diameter (*d*) = 6.4 µm, liquid condensed (LC) phase area 35%), with 7 mol% cholesterol (**B**,**G**,**L**, *d* = 5.6 µm, LC phase area 22%), with 10 mol% cholesterol (**C**,**H**,**M**, *d* = 6.1 µm, LC phase area 22%), with 12 mol% cholesterol (**D**,**I**,**N**, *d* = 3.6 µm, LC phase area 7%), and with 14 mol% cholesterol (**E**,**J**,**O**, *d* = 5.8 µm, LC phase area 22%) in the phospholipid coating. Images show the ligand distribution (**A**–**E**; Oregon Green 488), liquid expanded (LE) phase (**F**–**J**; rhodamine-DHPE), and composite view (**K**–**O**). Scale bars are 1 µm.

**Figure 3.** (**A**) Parts classified as inhomogeneity (%) in the ligand distribution, and (**B**) size of the LC area (% of total surface area) of indirect DSPC microbubbles without cholesterol (*N* = 58), with 7 mol% (*N* = 34), 10 mol% (*N* = 40), 12 mol% (*N* = 61), and with 14 mol% (*N* = 45) cholesterol in the coating. Boxplots show the median and interquartile range with whiskers from minimum to maximum. Statistical significance is indicated with \* *p* < 0.05, \*\* *p* < 0.01, or \*\*\**p* < 0.001. **Figure 3.** (**A**) Parts classified as inhomogeneity (%) in the ligand distribution, and (**B**) size of the LC area (% of total surface area) of indirect DSPC microbubbles without cholesterol (*N* = 58), with 7 mol% (*N* = 34), 10 mol% (*N* = 40), 12 mol% (*N* = 61), and with 14 mol% (*N* = 45) cholesterol in the coating. Boxplots show the median and interquartile range with whiskers from minimum to maximum. Statistical significance is indicated with \* *p* < 0.05, \*\* *p* < 0.01, or \*\*\**p* < 0.001. **Figure 3.** (**A**) Parts classified as inhomogeneity (%) in the ligand distribution, and (**B**) size of the LC area (% of total surface area) of indirect DSPC microbubbles without cholesterol (*N* = 58), with 7 mol% (*N* = 34), 10 mol% (*N* = 40), 12 mol% (*N* = 61), and with 14 mol% (*N* = 45) cholesterol in the coating. Boxplots show the median and interquartile range with whiskers from minimum to maximum. Statistical significance is indicated with \* *p* < 0.05, \*\* *p* < 0.01, or \*\*\* *p* < 0.001.

The lipids were phase-separated in indirect DSPC-based microbubbles without cholesterol, as shown in Figure 2F and quantified in Figure 3B. The fluorescent dye rhodamine-DHPE was enriched in bright interdomain regions (i.e., LE phase) and absent in LC domains. In indirect DSPC-cholesterol microbubbles, the LC domains were less pro-The lipids were phase-separated in indirect DSPC-based microbubbles without cholesterol, as shown in Figure 2F and quantified in Figure 3B. The fluorescent dye rhodamine-DHPE was enriched in bright interdomain regions (i.e., LE phase) and absent in LC domains. In indirect DSPC-cholesterol microbubbles, the LC domains were less pro-The lipids were phase-separated in indirect DSPC-based microbubbles without cholesterol, as shown in Figure 2F and quantified in Figure 3B. The fluorescent dye rhodamine-DHPE was enriched in bright interdomain regions (i.e., LE phase) and absent in LC domains. In indirect DSPC-cholesterol microbubbles, the LC domains were less pronounced compared to those without cholesterol (Figure 2G–J). With increasing concentrations of cholesterol up to 12 mol%, the lipid phase distribution was increasingly affected, as reflected by quantification of the LC phase area (Figure 3B). Microbubbles without cholesterol had a significantly larger surface area in the LC phase than those with cholesterol in their coating. Microbubbles with 7 mol% cholesterol displayed LE phase areas with an enriched fluorescent dye (Figure 2G) and had a significantly larger surface area in the LC phase

than those with more cholesterol in their coating. Microbubbles with 10 mol% cholesterol displayed LE phase areas as well (Figure 2H). Microbubbles with 12 mol% cholesterol had a homogeneous distribution of the fluorescent dye rhodamine-DHPE (Figure 2I), with the smallest LC phase area per microbubble of all formulations (Figure 3B). In microbubbles with 14 mol% cholesterol, rhodamine-DHPE was not only distributed homogeneously in the coating but also present in buckles on the outside of the coating (Figure 2J). The LC phase area in microbubbles with 14 mol% cholesterol was comparable to the LC phase area in microbubbles with 10 mol% cholesterol (Figure 3B). lesterol displayed LE phase areas as well (Figure 2H). Microbubbles with 12 mol% cholesterol had a homogeneous distribution of the fluorescent dye rhodamine-DHPE (Figure 2I), with the smallest LC phase area per microbubble of all formulations (Figure 3B). In microbubbles with 14 mol% cholesterol, rhodamine-DHPE was not only distributed homogeneously in the coating but also present in buckles on the outside of the coating (Figure 2J). The LC phase area in microbubbles with 14 mol% cholesterol was comparable to the LC phase area in microbubbles with 10 mol% cholesterol (Figure 3B). Figure 4 shows the percentage of indirect DSPC-based microbubbles with buckles

nounced compared to those without cholesterol (Figure 2G–J). With increasing concentrations of cholesterol up to 12 mol%, the lipid phase distribution was increasingly affected, as reflected by quantification of the LC phase area (Figure 3B). Microbubbles without cholesterol had a significantly larger surface area in the LC phase than those with cholesterol in their coating. Microbubbles with 7 mol% cholesterol displayed LE phase areas with an enriched fluorescent dye (Figure 2G) and had a significantly larger surface area in the LC phase than those with more cholesterol in their coating. Microbubbles with 10 mol% cho-

Figure 4 shows the percentage of indirect DSPC-based microbubbles with buckles per batch. An example of a microbubble with buckles is shown in Figure 2J,O. Microbubbles without cholesterol in the coating had the lowest incidence of buckles. Microbubbles with 12 mol% cholesterol in the coating had a higher incidence of buckles (*p* = 0.050) than those without cholesterol. Furthermore, the variability between batches increased with higher concentrations of cholesterol. per batch. An example of a microbubble with buckles is shown in Figure 2J,O. Microbubbles without cholesterol in the coating had the lowest incidence of buckles. Microbubbles with 12 mol% cholesterol in the coating had a higher incidence of buckles (*p* = 0.050) than those without cholesterol. Furthermore, the variability between batches increased with higher concentrations of cholesterol.

*Pharmaceutics* **2021**, *13*, x FOR PEER REVIEW 9 of 22

**Figure 4.** Percentage of microbubbles (MBs) with buckles per batch of indirect DSPC-based microbubbles without cholesterol and with 7, 10, 12, or 14 mol% cholesterol. Each symbol represents one batch of microbubbles. Overlaid black lines represent the median and interquartile range. Statistical significance is indicated with \* *p* < 0.05. **Figure 4.** Percentage of microbubbles (MBs) with buckles per batch of indirect DSPC-based microbubbles without cholesterol and with 7, 10, 12, or 14 mol% cholesterol. Each symbol represents one batch of microbubbles. Overlaid black lines represent the median and interquartile range. Statistical significance is indicated with \* *p* < 0.05.
