*3.2. Properties of the HA-Coated Liposomes*

The physicochemical properties of the four different liposome formulations produced for the experiments are displayed in Table 1. The lipid composition of all liposomes is based on an optimized formulation for light-activated ICG liposomes reported earlier [18,20]. Depending on the liposome coating, the ICG is either located within the lipid bilayer or bound by the surface polymers [21]. HA-coated liposomes were produced at different sizes (~70 nm and ~100 nm), demonstrating their robust size control. Polydispersity indexes were in the range of 0.05–0.08. The HA-coated liposomes showed lower zeta potential than the uncoated and PEG-coated liposomes. The PEG- and HA-coated liposomes had comparable ICG absorbances, suggesting that the polymer coating did not affect ICG binding on the liposomes, but the uncoated liposomes bound less ICG (Table 1).


*Pharmaceutics* **2020**, *12*, x FOR PEER REVIEW 8 of 25

**Table 1.** Physico-chemical properties of the liposome formulations produced for the experiments (mean of triplicate measurements with SD). ICG—indocyanine green; PEG—polyethylene glycol; HA—hyaluronic acid; n.a.—not available.

The coating had no significant effect on the lipid phase transition in the liposomes (Figure 1A). The transition to the liquid-disordered phase began at ~42 ◦C and reached its peak at ~44 ◦C. On the other hand, temperature-induced calcein release from small HA-coated liposomes took place at 39–41 ◦C, while the 100 nm liposomes released their contents at 38–40 ◦C (Figure 1B). Light activation in buffer for 15 s was sufficient to induce complete content release from all liposomes, while, at 5 s exposure, the small HA-coated liposomes showed lower content release than the 100 nm liposomes (Figure 1B). Differences in content release between light-activated and normal liposomes were significant (*p* < 0.01) (Figure 1C). **Table 1.** Physico-chemical properties of the liposome formulations produced for the experiments (mean of triplicate measurements with SD). ICG—indocyanine green; PEG—polyethylene glycol; HA—hyaluronic acid; n.a.—not available. **Size (nm) Zeta Potential (mV) ICG Absorbance (a.u.)** Uncoated 102 ± 24 −0.44 ± 1.32 0.72 ± 0.08 PEG 119 ± 30 −2.94 ± 0.69 0.90 ± 0.13 HA 104 ± 25 −11.27 ± 0.21 0.88 ± 0.12 HA (30 nm extrusion) 68 ± 16 n.a. 0.86 ± 0.03

**Figure 1.** (**A**) Calorimetric analysis of liposomes. Heat flow as a function of temperature (*T*) is shown. (**B**) Calcein release from liposomes at different temperatures. (**C**) Light-activated content release from liposomes after 5 s or 15 s exposures (800 nm laser at 3.2 W/cm<sup>2</sup> ) was significant compared to controls (ctrl) absent the light exposure (\*\* *p* < 0.01). **Figure 1.** (**A**) Calorimetric analysis of liposomes. Heat flow as a function of temperature (*T*) is shown. (**B**) Calcein release from liposomes at different temperatures. (**C**) Light-activated content release from liposomes after 5 s or 15 s exposures (800 nm laser at 3.2 W/cm<sup>2</sup> ) was significant compared to controls (ctrl) absent the light exposure (\*\* *p* < 0.01).

The coating had no significant effect on the lipid phase transition in the liposomes (Figure 1A). The transition to the liquid-disordered phase began at ~42 °C and reached its peak at ~44 °C. On the other hand, temperature-induced calcein release from small HA-coated liposomes took place at 39– 41 °C, while the 100 nm liposomes released their contents at 38–40 °C (Figure 1B). Light activation in buffer for 15 s was sufficient to induce complete content release from all liposomes, while, at 5 s exposure, the small HA-coated liposomes showed lower content release than the 100 nm liposomes (Figure 1B). Differences in content release between light-activated and normal liposomes were significant (*p* < 0.01) (Figure 1C). Liposomes with a doubled amount of HA (molar ratio = 2) did not show significant differences compared to the liposomes with lower HA content (Figure S2, Supplementary Materials). In order to Liposomes with a doubled amount of HA (molar ratio = 2) did not show significant differences compared to the liposomes with lower HA content (Figure S2, Supplementary Materials). In order to determine the ICG binding capacity of the HA coating, we successfully doubled the amount of ICG (Figure S3, Supplementary Materials). The ICG was bound to the liposomes at higher quantities (ICG absorbance of 1.61 ± 0.04 a.u., as compared to 0.88 ± 0.12 a.u. for the regular HA-coated liposomes). However, if the ICG content was doubled at normal HA levels (molar ratio = 1), significant calcein leakage was observed below 37 ◦C, which rendered these liposomes unsuitable for in vivo use. Due to the suboptimal stability and pharmacokinetics of the uncoated liposomes, they were excluded from the stability studies.

#### determine the ICG binding capacity of the HA coating, we successfully doubled the amount of ICG (Figure S3, Supplementary Materials). The ICG was bound to the liposomes at higher quantities (ICG *3.3. Stability Studies*

#### absorbance of 1.61 ± 0.04 a.u., as compared to 0.88 ± 0.12 a.u. for the regular HA-coated liposomes). 3.3.1. ICG Stability and Passive Content Leakage

3.3.1. ICG Stability and Passive Content Leakage

vitreous or plasma samples.

leakage was observed below 37 °C, which rendered these liposomes unsuitable for in vivo use. Due to the suboptimal stability and pharmacokinetics of the uncoated liposomes, they were excluded from the stability studies. *3.3. Stability Studies*  Light-activated liposomes should stay intact and avoid passive content leakage for sufficient duration. The stability of the PEG- and HA-coated liposomes was assessed in porcine vitreous and human plasma samples at 35 ◦C and 37 ◦C, respectively (Figure 2). The ICG absorbance and calcein leakage were measured for one week. Significant changes in ICG absorbance were not seen in vitreous or plasma samples.

However, if the ICG content was doubled at normal HA levels (molar ratio = 1), significant calcein

Light-activated liposomes should stay intact and avoid passive content leakage for sufficient

*Pharmaceutics* **2020**, *12*, x FOR PEER REVIEW 9 of 25

**Figure 2.** Stability of the HA- and PEG-coated liposomes at physiological temperatures monitored for 168 h (one week) based on ICG absorbance (**A**,**C**) and calcein fluorescence (**B**,**D**) in porcine vitreous (**A**,**B**) and human plasma (**B**,**D**) samples. Liposome solution (25%) was mixed with the biological fluids (75%). The snapshot (**E**) shows the time-points during the first 12 h of calcein release (\* *p* < 0.05). **Figure 2.** Stability of the HA- and PEG-coated liposomes at physiological temperatures monitored for 168 h (one week) based on ICG absorbance (**A**,**C**) and calcein fluorescence (**B**,**D**) in porcine vitreous (**A**,**B**) and human plasma (**B**,**D**) samples. Liposome solution (25%) was mixed with the biological fluids (75%). The snapshot (**E**) shows the time-points during the first 12 h of calcein release (\* *p* < 0.05). **Figure 2.** Stability of the HA- and PEG-coated liposomes at physiological temperatures monitored for 168 h (one week) based on ICG absorbance (**A**,**C**) and calcein fluorescence (**B**,**D**) in porcine vitreous (**A**,**B**) and human plasma (**B**,**D**) samples. Liposome solution (25%) was mixed with the biological

fluids (75%). The snapshot (**E**) shows the time-points during the first 12 h of calcein release (\* *p* < 0.05).

Both PEG- and HA-coated liposomes showed significant calcein leakage in 2–4 days in the vitreous sample (Figure 2B). Leakage of the liposomal contents was faster in plasma than in vitreous samples, as complete leakage was observed within 24 h (Figure 2E). However, the content release in plasma from the HA-coated liposomes was significantly slower than from the PEG-coated liposomes. Both PEG- and HA-coated liposomes showed significant calcein leakage in 2–4 days in the vitreous sample (Figure 2B). Leakage of the liposomal contents was faster in plasma than in vitreous samples, as complete leakage was observed within 24 h (Figure 2E). However, the content release in plasma from the HA-coated liposomes was significantly slower than from the PEG-coated liposomes. Both PEG- and HA-coated liposomes showed significant calcein leakage in 2–4 days in the vitreous sample (Figure 2B). Leakage of the liposomal contents was faster in plasma than in vitreous samples, as complete leakage was observed within 24 h (Figure 2E). However, the content release in plasma from the HA-coated liposomes was significantly slower than from the PEG-coated liposomes.

3.3.2. Light-Activated Content Release in Vitreous and Plasma Samples 3.3.2. Light-Activated Content Release in Vitreous and Plasma Samples 3.3.2. Light-Activated Content Release in Vitreous and Plasma Samples

The light-triggered content release mechanism should be functional in biological media. Lightactivated calcein release from HA-coated liposomes was studied at 3 h and 24 h in the vitreous (35 °C) and at 3 h in plasma (37 °C) samples (Figure 3). The longer light exposure (15 s vs. 5 s) significantly increased the calcein release in the vitreous sample at 3 h and 24 h (\* *p* < 0.05). Even more pronounced triggered release was observed in plasma (Figure 3). The light-triggered content release mechanism should be functional in biological media. Light-activated calcein release from HA-coated liposomes was studied at 3 h and 24 h in the vitreous (35 ◦C) and at 3 h in plasma (37 ◦C) samples (Figure 3). The longer light exposure (15 s vs. 5 s) significantly increased the calcein release in the vitreous sample at 3 h and 24 h (\* *p* < 0.05). Even more pronounced triggered release was observed in plasma (Figure 3). The light-triggered content release mechanism should be functional in biological media. Lightactivated calcein release from HA-coated liposomes was studied at 3 h and 24 h in the vitreous (35 °C) and at 3 h in plasma (37 °C) samples (Figure 3). The longer light exposure (15 s vs. 5 s) significantly increased the calcein release in the vitreous sample at 3 h and 24 h (\* *p* < 0.05). Even more pronounced triggered release was observed in plasma (Figure 3).

**Figure 3.** Light-activated release from HA-coated liposomes with 5 s and 15 s light exposures (800 nm laser at 3.2 W/cm<sup>2</sup> ) and passive content release (leakage) following a 3 h or 24 h incubation of the liposomes in 75% porcine vitreous (35 °C) or human plasma (37 °C) samples. Significant differences in calcein release were observed between different exposure times (\**p* < 0.05). **Figure 3.** Light-activated release from HA-coated liposomes with 5 s and 15 s light exposures (800 nm laser at 3.2 W/cm<sup>2</sup> ) and passive content release (leakage) following a 3 h or 24 h incubation of the liposomes in 75% porcine vitreous (35 °C) or human plasma (37 °C) samples. Significant differences in calcein release were observed between different exposure times (\**p* < 0.05). **Figure 3.** Light-activated release from HA-coated liposomes with 5 s and 15 s light exposures (800 nm laser at 3.2 W/cm<sup>2</sup> ) and passive content release (leakage) following a 3 h or 24 h incubation of the liposomes in 75% porcine vitreous (35 ◦C) or human plasma (37 ◦C) samples. Significant differences in calcein release were observed between different exposure times (\* *p* < 0.05).
