Repetitive drug releases from light-activatable micron-sized liposomes

Abstract

In this work, a novel light activatable micron-sized liposomal drug carrier that has a unique capability to release drug repetitively in proportion to the cycle number of short irradiation (5 s) of near-infrared (NIR) pulsed laser is reported. We synthesized methotrexate (MTX)-loaded liposomes based on a modified reverse-phase evaporation method. Gold nanorods (AuNR) were attached to the liposomal surfaces, enabling the liposomes to release drug under short NIR irradiation via the photothermal effect. The concentrations of methotrexate (MTX) released from the liposomes were 10.6, 29.8, 43.7 and 65.9 μg/mL after one, two, three or four NIR laser cycles (1.1 W at 1064 nm, 5 s per cycle), respectively. The current finding will provide possible solution to the previously reported inconsistency in drug release from light activatable liposomal drug carriers at each activation cycle. The repeatability of drug release described in this work is believed to be due to reversible nature of the liposomes. The liposomes release drug via lipid bilayer melting when irradiated by laser due to gold nanorods’ plasmonic heat on the lipid bilayer surface and quickly regain their original structure once the laser source is removed. We provided evidence of the reversible liposomal structures by monitoring the change of number densities of liposomes using a microelectrode sensor with different laser irradiation durations and powers. We also assessed the micron-sized liposome with respect to long-term stability, drug encapsulation efficiency, and drug-releasing efficiency, demonstrating the possibility of utilizing these liposomes as long-term drug delivery vehicles for various drugs.

1. Introduction

Since the discovery in early 1960 s to use dry lipids as the starting materials to form liposomes with a hydrophilic or hydrophobic core to encapsulate drug by simple hydration, liposomes had been used in various drug delivery applications for targeted drug release [1–5]. When liposomes combined with light-triggering materials such as gold nanorods, external light source including near infrared laser can trigger release drug from liposomes, enabling tunable dosage control by laser parameters, i.e. energy, duration and beam area [1,6,7]. The light-triggered drug release mechanism is considered as photothermal process, where heat is generated via surface plasmonic resonance of gold nanorod surface when the gold nanorod is irradiated and the heat alters the lipid bilayer structure on-demand basis [8–11]. Liposomes with photothermal agent that can release a payload stepwise by multiple on/off laser cycles have been reported [12–21].

Although the surface plasmonic resonance photothermal effect is very local and transient at the nanoscale [22], the most studies relies on increasing bulk temperature of the whole liposome dispersion beyond the lipid bilayer’s phase-transition temperature. by heating up the liposome dispersion via long laser irradiation duration at least for a few minutes. We had previously shown that a light-activatable gold nanorod (AuNR) coated liposomal system (liposome diameter: ~100 nm) released drug repetitively by 5 s irradiation using a pulsed laser [13].

In this study, we have created light-activatable drug-encapsulated micron-sized liposomes with an average liposome size of ~1.5 μm, in order to increase drug capacity and to utilize for a long-term drug delivery system combined with a porous capsule. Although our previous study showed repetitive drug releases with relatively short irradiation durations with liposomes with ~100 nm, low drug encapsulation capacity and efficiency (~ 0.4%) due to the small liposome size limited the use for long term drug delivery systems.

The micron-sized liposomes were synthesized via the reverse-phase evaporation method. This method was introduced by Szoka et al. in 1978 [23] as an improved method to overcome a limit of low encapsulation volume of liposome cores synthesized via lipid film hydration [24]. Liposomes synthesized with the reverse-phase evaporation method have been exploited as drug carriers because of the large drug encapsulation capacity and high encapsulation efficiency [25,26]. However, long-term multiple drug releases using micron-sized liposomes has not been reported to the best of our knowledge.

Our group has shown a potential long-term drug delivery system, that comprises a nanoporous polylactide-co-glycolic acid (PLGA) capsule and micron-sized AuNR-coated liposomes [27,28]. We demonstrated that this composite system did not cause inflammation or foreign body responses during six months in a rabbit eye, indicating no significant leakage occurred on liposomes [27]. In this study, we will further verify the drug release mechanism from the light-activatable micron--sized liposomes using microelectrodes. Understanding drug release mechanism, i.e. whether or not the liposome structure is reversible after irradiation under various laser parameters, is important because it will determine strategy of laser irradiation for optimal, controlled and consistent drug delivery, especially for long-term usage.

We hypothesize in this study that drug release from the micron-sized AuNR-coated liposomes synthesized by the reverse-phase evaporation method will be repetitively responsive to light with reversible liposomal structure (Fig. 1). According to Paasonen et al., liposome bilayers near the gold nanoparticles undergo a phase transition from gel-phase (L_β′) to a metastable rippled phase (P_β′) then to a liquid-phase (L_α) when exposed to UV light (365 nm), which is mainly responsible for releasing the payload [25,29]. They also found that the process from P_β′ to L_α was reversible when on/off cycles of UV was applied. The liposomes eventually return to the gel phase L_β′ after several hours of cooling time. This study proved the possibility of using the laser as an on-off switch to repetitively control the drug release.

Fig. 1.

Schematic diagram of the reversible drug-releasing process from AuNR-MTX liposomes activated by laser irradiation, using a near-infrared (NIR) laser.

Here, we have shown drug releases proportional to laser irradiation cycles from micron-sized liposomes repetitively, demonstrating the possibility of utilizing these liposomes as long-term drug delivery vehicles for drug or macromolecular therapeutic substances with encapsulation efficiency 20 times higher than our previous study on small liposomes [13]. We provided evidence of the reversible liposomal structures by monitoring the change of number densities of liposomes with different laser irradiation durations and powers. We also assessed the micron-sized liposomes with respect to long-term stability, drug encapsulation efficiency, and drug-releasing efficiency. The novelty of this study is twofold: for the first time, 1. we have shown repetitive drug releases from micron-sized liposomes using pulsed laser with short laser duration (5 s), and 2. we have demonstrated the drug release mechanism by measuring the number of liposomes before and after laser irradiation using a microelectrode sensor. The study will suggest potential of the light-activatable micron-sized liposomes as a long-term drug delivery carrier.

2. Materials and methods

2.1. Materials

Potassium hexacyanoferrate (II) trihydrate (K₄Fe(CN)₆.3H₂O), methanol (analytical grade) and chloroform (analytical grade) were obtained from Fisher Scientific (Waltham, MA). All the other materials used in this study can be found in He et al. [27].

2.2. Synthesis of gold nanorod-coated methotrexate-encapsulated liposomes (AuNR-MTX liposomes)

The AuNR-MTX liposomes were prepared following a reverse-phase evaporation method [23,30] with several modifications to organic solvents and lipid compositions, as described in our previous study [27]. Briefly, 50 μmol of mixed lipids (DSPC: cholesterol: stearylamine: DSPE-PEG 5000: = 50: 35: 10: 5, molar ratio) was dissolved in 1 mL mixture of methanol: chloroform (1:2, v/v). 1 mL MTX solution (200 mg/mL in 0.49% sodium chloride) was added to the lipid phase dropwise and then stirred. After the organic solvents were removed by the rotary evaporator, liposomes were synthesized [27]. Then they were passed through a PD-10 Column for purification. The AuNRs are citrate-coated and have a diameter of 10 nm and a length of 59 nm, exhibiting a surface plasmon resonance (SPR) peak at ~980 nm [31]. AuNRs were attached to the liposome surfaces via attractive electrostatic forces. The MTX-encapsulated bare liposomes with no gold coating (MTX liposomes) were prepared in the same way without adding AuNR suspension.

2.3. Laser Irradiation

The laser irradiation tests were performed by a pulsed laser (pulse width: 700 picosecond, Wedge-HF, RPMC Lasers Inc., O’Fallon, MO) (1064 nm, pulse repetition frequency at 10 kHz) with a setup previously described in Das et al. [13]. Briefly, a glass capillary loaded with AuNR-MTX liposome sample (~30 μL) was inserted in the 3D-printed holder vertically (Fig. 2). The entire sample in the capillary was covered by the laser beam during each laser-irradiation test.

Fig. 2.

Photograph (left) and schematic (right) of the laser irradiation setup.

2.4. Characterization of the AuNR-MTX Liposomes

2.4.1. Transmission Electron Microscopy (TEM)

The TEM images were obtained by following the procedures described previously [13]. Briefly, a drop of the AuNR-MTX liposomes were added onto a formvar-coated grid. Then it was treated with negative staining using 2% uranyl acetate solution. The sample was freeze-dried at −20 °C and kept frozen until imaged by a JEM-1230 microscope (JEOL, Tokyo, Japan).

2.4.2. Confocal fluorescence microscopy

The confocal fluorescence image was obtained by a Leica SP8 DIVE MP microscope (Leica, Wetzlar, Germany) [27]. To obtain fluorescently labeled liposomes, TopFluor cholesterol was added in the lipid composition (5 mol%) and Sulfo-cyanine 5 carboxyl acid (Cy5) was co-encapsulated inside liposomes with MTX [27].

2.4.3. Dynamic Light Scattering (DLS)

DLS was used to determine the hydrodynamic sizes of AuNR-MTX liposomes as described in He et al. [27] The AuNR-MTX liposomes were diluted 200 times in PBS. Then it was measured by a NanoBrook Omni particle analyzer (Brookhaven Instruments Co., Holtsville, NY) at a scattering angle of 90°. Three measurements were conducted and averaged for each sample.

2.4.4. Encapsulation efficiency

After the non-encapsulated MTX was removed by PD-10 desalting column, the liposome suspension was collected for analyzing the MTX content. First, a calibration curve of standard MTX solutions in PBS was generated (Fig. S1) using a UV–vis spectrophotometer (SpectraMax, Molecular Devices, LLC., San Jose, CA). Then liposomal MTX was diluted 50× and measured for the optical density by the UV-Vis spectrophotometer, and the concentration of MTX was obtained through the calibration curve. The encapsulation efficiency (EE%) was obtained by

$$EE\%=\frac{Mas{s}_{MTXencapsulatedinsideliposomes}}{Mas{s}_{toralMTX}}\times 100.$$ (1)

The determination of EE% were repeated with three different samples (n = 3).

2.5. Stability of the liposomes against aggregation and passive leakage

The stability against aggregation was determined by measuring the size changes over time utilizing DLS. AuNR-MTX liposomes were concentrated to a viscous suspension by centrifugation (6000 rpm for 30 min). The suspension was protected from light and stored at room temperature for monitoring over time. 10 μL of the sample was taken and diluted 200× with PBS and measured by DLS (n = 3) every month for up to 5 months.

For evaluating stability against passive leakage, 100 μL of the AuNR-MTX liposome was carefully loaded into a micro dialyzer with a 5 kDa MWCO membrane (Harvard Apparatus, Holliston, MA). The microdialyzer was put in a sealed glass vial with 5 mL PBS as dialysis buffer, and a stir bar was used for constant agitation. The glass vial with the dialysis unit was covered with aluminum foil and placed at room temperature. To determine the amount of MTX passively released from the liposomes, 100 μL of the dialysis buffer was measured at 370 nm with the UV-Vis spectrophotometer and the MTX concentration was subsequently calculated from the calibration curve (Figure S1). The amount of released MTX was evaluated daily for the first month, then weekly for the second and the third month, and biweekly for subsequent three months to reduce the likelihood of contamination from frequent operations.

2.6. Electrochemical measurement of collision frequency

Electrochemical measurements to determine the collision frequency were performed at room performed on a CHI 630E (CH Instruments, Austin, TX) Electrochemical Workstation. The detail description of the experimental setup and equipment parameters were described in our pervious study [13]. Briefly, A two-electrode set up consists of a working electrode made of Pt with 25 μm diameter (fabricated in-house) and a quasi-reference counter electrode (QRCE) made of silver chloride-coated silver wire (Ag/AgCl) was used. Around 30 μL of liposomes solution was irradiated in capillaries for a desired laser irradiation duration (0 s, 20 s, 40 s and 60 s) and an aliquot of 1.25 μL from the irradiated stock solution was diluted 400 times with K₄Fe(CN)₆ solution (200 mM). The working electrode was held at +0.4 V versus the Ag/AgCl QRCE, and current data due to oxidation of ferrocyanide to ferricyanide were recorded every 50 ms. Each condition was tested in triplicate. The number of current steps (above a threshold of 0.3 nA) was counted using MATLAB 2019b (MathWorks) and the first current-time transient data of 300 s was used to calculate the collision frequency. This current threshold was used above the noise level, meaning that no steps (of 0.3 nA) were recorded in the blank sample. The size of particles was not analyzed electrochemically, and it is known that the location on the electrode surface that the particle collides can result in a different magnitude of current step decrease, for the same size of particle.

2.7. Amount of laser-triggered drug release

The concentration of released MTX was determined by UV–vis spectroscopy [13]. Briefly, 20 μL of irradiated liposome dispersion was diluted 20 times in PBS and stored at room temperature for 18 h. Then the diluted sample was centrifuged (10,000 g for 10 min) and 100 μL of the supernatant was taken for determining the concentration of MTX (C_irradiated) by UV-Vis spectroscopy. The MTX concentration obtained from non-irradiated liposome suspensions (C_{non-irradiated}) was also determined and it represented the background (e.g., MTX not induced by laser). The total concentration of MTX released by laser treatment was calculated by the equation below:

$$C_{\text{MTX}\text{released}\text{by}\text{laser}}=C_{\text{irradiated}}-C_{\text{non}-\text{irradiated}}$$ (2)

Five measurements were conducted and averaged for each condition.

3. Results and discussion

3.1. Characterization of AuNR-MTX liposomes before and after irradiation

The fluorescence confocal image (Fig. 3A) showed the presence of particles with two main populations of different sizes, one at ~5 μm and the other < 1 μm, which is consistent with the DLS results (Fig. 3B). The blue fluorescence signals in the liposome core resulting from the Cy5 dye indicate that the hydrophilic molecules including the dye and MTX were co-encapsulated in the aqueous cores. The lipid membranes were also visualized by the fluorescence-labeled cholesterol (yellow in Fig. 3 A). Cholesterol molecules were distributed throughout the membranes. The average hydrodynamic diameter was 1.524 ± 0.104 μm with a polydispersity index (PDI) of 0.186 ± 0.052, according to the DLS data. The TEM image in Fig. 3C confirmed the presence of two AuNRs on the liposome surface (white arrow). The creased and wrinkled surface of the liposome is due to dehydration during the sample preparation for TEM imaging, i.e., freeze-drying process.

Fig. 3.

(A) Fluorescence confocal image of MTX/Cy5 co-encapsulated AuNR-coated liposomes. Blue: the aqueous core labeled by Cy5, yellow: the lipid membrane labeled by TopFluor-cholesterol. (B) Size distribution and (C) TEM image of a AuNR-MTX liposome with no laser irradiation. White arrow points to the location of two adjacent AuNRs. (D) Size distribution and (E) TEM image of AuNR-MTX liposomes after 20 s laser irradiation (5 s × 4 cycles) at 1.1 W. Run #1, #2 and #3 are triplicate DLS measurements (n= 3). White arrow points to a AuNR.

After 20 s laser irradiation (5 s × 4 cycles, 1.1 W), the size distribution of irradiated liposomes was similar to the samples before laser irradiation (Fig. 3D). The TEM image in Fig. 3E also showed no apparent changes in the shape or structure. The shape of AuNRs after irradiation remained unaltered (white arrow). The crystals scattered around the liposome could be precipitates formed by uranyl acetate precipitates (negative staining solution), which were observed throughout the sample. The AuNR-MTX liposomes showed an average diameter of 1.692 ± 0.246 μm with a PDI of 0.239 ± 0.112 after 20 s irradiation. Compared to the average diameter of 1.524 ± 0.104 μm before laser irradiation, this result indicated that the size did not change significantly after laser irradiation.

3.2. Long-term stability of AuNR-MTX liposomes

Liposome-embedded composite drug delivery systems (DDS) have been recently studied as a solution to overcome limitations, such as burst release, instability, and short release time. [32] The drug release profiles were significantly prolonged by incorporating liposomes into the composite system, such as hydrogel [33–35] and polymeric scaffold [36]. However, the most studies reported the release period from several days to weeks. It is necessary to improve to further prolong the efficacy duration to treat life-long chronic diseases to minimize frequency of surgery or injection. In order for the micron-sized liposomes to be utilized for long-term drug delivery for 6 months and beyond, the stability of the liposomes needs to be determined.

Stability of AuNR-MTX liposomes against aggregation was tested over 150 days (5 months) when stored at room temperature. Overall, the average effective diameter of liposomes remained stable over the period (Fig. 4A), indicative of the colloidal stability of the liposomes against aggregation. Fig. 4B shows that the passive leakage of MTX from the liposomes in PBS was less than 1.5% over 180 days. Low leakage is important for not only preventing the adverse effect associated with drug leakage over prolonged therapeutic period and but also extending the life-time for trigger-release. Our other studies have demonstrated the feasibility of using the micron-sized liposomes in rabbit eyes up to 6 months, when the liposomes were encapsulated in a polymeric capsule, showing the stability of the liposomes against passive leakage.

Fig. 4.

(A) Average size of AuNR-MTX liposomes over 180 days measured by DLS. (B) MTX leakage from the AuNR-MTX liposomes measured by UV–vis spectrometer over 180 days.

3.3. Effect of laser irradiation on reversibility of AuNR-MTX liposomes

Fig. 5 shows the electrochemical measurement of collision frequencies of AuNR-MTX liposomes in K₄Fe(CN)₆ solution at 1.1 Watts and 1.8 Watts laser power for four laser irradiation times, 0 s, 20 s, 40 s and 60 s Fig. 5A shows the schematic of this experiment where K₄Fe (CN)₆ was used as redox agent that produces current at Pt microelectrode surface. Every time a current step is produced when a single liposome collides with the Pt electrode and attaching either on the electrode surface or on other liposomes on the electrode surface, or on sites close to the electrode perimeter. All of these conditions are responsible for depleting the flux of K₄Fe(CN)₆ to the Pt electrode and therefore reducing the current measured on the electrode surface. In the absence of liposomes, K₄Fe(CN)₆ solution (Blank) reached a steady current of ~650 nA within 20 s at a laser power of 1.1 W or 1.8 W (Fig. 5B and D, respectively). The situation changed when AuNR liposomes were introduced in the K₄Fe(CN)₆ solution. The liposomes were responsible to decrease the measured current at the electrode surface in a stepwise manner and the collision frequency was calculated by the number of current steps. The collision frequency is expected to have proportional relationship with the concentration of the liposomes. Fig. 5C and 5E show the collision frequency comparision at 1.1 W and 1.8 W respectively for all four laser duration times. When the laser was initially turned off (laser irradiation time= 0 s), the collision frequency of AuNR-MTX liposomes was around 0.14 ± 0.03 Hz. At 1.1 W, for 20 s, 40 s and 60 s laser durations, the collision frequencies were 0.16 ± 0.02 Hz, 0.14 ± 0.02 Hz and 0.14 ± 0.03 Hz respectively with no statistical difference among the values indicated by p-values > 0.05 for all combinations from the t-test (Table S1). Similar situation was also observed in the case of 1.8 W laser power, where the collision frequencies were 0.15 ± 0.03 Hz, 0.15 ± 0.02 Hz and 0.16 ± 0.03 Hz for 20 s, 40 s and 60 s laser irradiation. No significant difference was observed among the values (Table S2). The unaltered behavior of collision frequencies for AuNR-MTX liposomes with and without laser exposures, indicating that the liposomes did not undergo structural change that can be detected by the microelectrode. Note that in our previous study of nanodroplets, the collision frequencies reduced significantly due to the nanodroplets undergoing a phase transition to bubbles after laser irradiation at the same laser parameters. [13] This was because the nanodroplets became microbubbles and burst, where the structure is irreversible after phase-transition.

Fig. 5.

(A) Schematic of electrochemical measurement of collision frequency. A current step corresponds to a blocking collision of a liposome with the electrode. (B) The stepwise decrease of the current over 300 s for the AuNR-MTX liposomes after 0 s, 20 s, 40 s and 60 s-irradiation at a laser power of 1.1 W. The insets display the current steps in a zoom-in region; (C) The corresponding collision frequencies under 0 s, 20 s, 40 s and 60 s-irradiation with a laser power of 1.1 W. (D) The step wise decrease at 1.8 W and (E) The collision frequencies at 1.8 W.

To calculate the total liposome concentration (C) from collision frequency (F_collision) the following Eq. (3) had been used,

$$C=\frac{F_{\text{collision}}}{4D{r}_{\text{elec}}{N}_A}$$ (3)

where r_elec is the radius of the electrode (12.5 μm), N_A is Avogadro’s number (6.023 × 10²³), and D is the diffusion coefficient of the liposomes

$$\text{D}=\frac{k_{B}T}{6\pi \mu r_{\text{lipo}}}$$ (4)

where k_B is Boltzmann’s constant (1.38 × 10⁻²³ m²·kg·s⁻²·K⁻¹), T is the temperature (298 K), μ is the viscosity of solution at 25 °C, and r_lipo is the average radius of the liposomes, which was determined by DLS (0.762 μm). Using the average collision frequency (F_collision, 0.14 Hz), the number concentration (C) of liposomes was determined to be 1.45 × 10⁻¹¹ mol/L or 8.70 × 10¹² /L. This result was consistent with the value which was calculated alternatively based the number of lipids per liposome using

$$\text{C}=\frac{M_{\text{lipid}}{N}_A}{N_{\text{lipid}}}$$ (5)

where M_lipid is 50 μmol and N_A is Avogadro’s number. N_lipid is the number of lipid molecules per liposome which is obtained using

$$\text{Nlipid}=\frac{4\pi \left\{r_{\text{lipo}}{}^2+{\left(r_{\text{lipo}}-t_{\text{bilayer}}\right)}^2\right\}}{A_{\text{h}}}$$ (6)

where r_lipo is the average hydrodynamic radius, t_bilayer is the thickness of lipid bilayer (5 nm) and A_h is the cross sectional area per lipid (0.71 nm²). According to the number-based size distribution determined by DLS (Figure S2), there were two populations which we have assigned the names “small”, for the population with average radius r_{lipo, small} = 113 nm and “large” for the second population with average radius r_{lipo, large} = 530 nm. Thus, the concentration of the liposomes was calculated precisely via

$$N_{\text{lipid,large}}\cdot n_{\text{lipo,large}}+N_{\text{lipid,small}}\cdot n_{\text{lipo,small}}=M_{\text{lipid}}\cdot N_A$$ (7)

The lipid number per large liposome (N_{lipid, large}) and small liposome (N_{lipid, small}) were calculated as 9.85 × 10⁶ and 4.30 × 10⁵, respectively, based on the respective sizes. The ratio between the number of large liposomes to the small liposomes was n_{lipo, large}: n_{lipo, small} = 1: 3.91, based on the peak areas of the DLS results. As a result, the concentration (C) of the liposomes was 6.52 × 10¹²/L after considering dilution factor (400×) for the collision frequency experiment.

3.4. Drug release from AuNR-MTX liposomes at multiple laser irradiation cycles

MTX release from the AuNR-MTX liposomes (number density 7.57 × 10¹¹/mL) by multiple cycles of 5 s irradiation was demonstrated in Fig. 6. Total MTX released from the AuNR-MTX liposomes was found to be proportionate to the number of laser irradiation cycles. The concentrations of MTX released were 10.6, 29.8, 43.7 and 65.9 μg/mL after one, two, three or four irradiation cycles, respectively. The analysis of variance (Table S3) indicated that these results were statistically significant (except for the 2nd and 3rd cycle). On the other hand, unmodified MTX liposomes (no AuNR attachment) showed minimal releases with no statistical differences among the different cycles (Table S4 and S5), implying that the releases of MTX were attributed to AuNR generating photothermal effect upon laser irradiations. In the drug release study, irradiation duration per cycle was 5 s because shape changes of AuNRs due to melting were observed after continuous 60-s irradiation at 1.1 W or 20-s irradiation at 1.8 W [13]. The shape change would affect light adsorption of AuNR and drug release efficiency. It is important to keep the shape of AuNRs in order to obtain consistent drug release per treatment for long term repetitive use.

Fig. 6.

Laser-triggered release measurements for AuNR-MTX liposomes and MTX liposomes under different laser irradiation cycles (5 s per cycle) at 1.1 W (n = 5).

According to the release tests and the collision frequency measurements, we confirmed that the structures of AuNR-MTX liposomes did not rupture after releasing MTX. Also, the fact that accumulative drug releases increased proportionally with irradiation cycles suggests that the amount of drug release can be controlled by laser durations. Additionally, DLS results confirmed that the average size of AuNR-MTX liposomes did not change after 20 s irradiation at 1.1 W (Fig. 3).

In Table 1, we compared the performances of the micron-sized AuNR-MTX liposomes and small AuNR-MTX fabricated by tip-sonication method in our previous publication [13]. While the two types of liposomes both used 200 mg/mL MTX solution in the fabrication, the micron-sized AuNR-MTX liposomes showed significantly higher encapsulation efficiency (EE%) than the small ones. It should be attributed to the large aqueous cores and therefore higher capacity of the micron-sized liposomes. Nevertheless, the EE% of both liposomes were low (<10%). This is considered due to high ionic strength of the MTX solution (200 mg/mL MTX disodium salt in 0.49% NaCl). The EE% is proven to have an inverse correlation with the ionic strength of drug solution for reverse phase evaporation method [23] or thin film hydration method [37]. Even though lowing the MTX concentration can increase the EE%, it would cause a reduced MTX concentration of the liposomes. Thus, the 200 mg/mL MTX solution was eventually used for the purpose of maximizing the MTX dosage. Novel liposome fabrication techniques can be explored in future to improve the EE%.

Table 1.

Comparison between the large AuNR-MTX liposomes fabricated by reverse-phase evaporation and small AuNR-MTX liposomes prepared by sonication method, on different properties and performances.

Liposomes made via different method	Avg. diameter (nm)	Total lipids (μmol)	EE%	Number density* (/mL)	Released MTX after 4-cylcle laser§ (μg/mL)	Released MTX after 4-cycle laser per liposome (μg)	Released MTX after 4-cycle laser per AuNR (μg)
Reverse-phase evaporation	1524 ± 104	50	8.00±0.39	7.57×1011	65.9	8.71×10−11	4.51×10−10
Sonication [13]	99.80 ± 2.80	10	0.40±0.01	1.89×1013	53.3	1.31×10−12	2.27×10−12

^*It refers to the number density of the liposome sample tested for laser-triggered drug-releasing.

^§5 s per cycle at 1.1 W, with 3-min cooling time between each cycle.

With the number density of the liposomes and AuNRs, we converted the released MTX amounts to releasing efficiency based on liposome number or AuNR number. It was demonstrated that each individual micron-sized MTX-AuNR liposome can release a greater amount of MTX than each small AuNR-MTX liposome (8.71 × 10⁻¹¹ μg vs. 1.31 × 10⁻¹² μg). Meanwhile, the micron-sized liposomes used in this study were able to release more MTX (4.51 × 10⁻¹⁰ μg) per AuNR than the small liposomes used in our previous study (2.27 × 10⁻¹² μg). In other words, the drug-releasing from micron-sized liposome was more efficient based on AuNR consumption.

The investigations involved with the phase-transition of lipid bilayers for light-activatable liposomes have been reported previously [29, 38–43]. A gel-to-liquid phase transition of the lipid bilayer upon light activation has been confirmed by small angle X-ray scattering and other techniques. However, the exact mechanism responsible for drug-releasing remains unclear. In addition, consistent drug releases per irradiation have rarely been demonstrated. An et al. [18] reported that gold nanoparticles-incorporated liposomes release entrapped berberine when triggered by UV light (250 nm wavelength). The drug releases were induced by four cycles and the released percentage decreased with the triggering times (2.5 min UV irradiation per cycle, with 5 min visible light between the cycles). The integrity of gold-liposomes after UV exposure was only investigated by indirect measurements (DLS). On the other hand, this study shows (1) consistent drug doses released from liposomes under multiple cycles of laser irradiation with relatively short irradiation 5 s and (2) the reversibility of individual liposome structure after exposure to laser irradiation confirmed by electrochemical experiment of particle-collision. These important features are desirable to utilize the AuNR-MTX liposomes as long-term drug delivery vehicles for various drugs.

As depicted in Fig. 1, we hypothesized that the temperature of lipid bilayers near AuNR exceeded the chain-melting transition temperature of the main lipid DSPC (approx. 55.5 °C [44,45]) by laser irradiation and undergo gel-to-liquid phase transition, during which MTX is released through the “melted” region of lipid bilayers. After cooling down, the lipid bilayers return to gel state with partial drug retained in the core. Our previous study indirectly measured the lipid bilayer temperature during the laser irradiation using a colorimetric method. [13] When the polymerized 10, 12-pentacosadiynoic acid (PCDA) lipid was incorporated in the lipid layer, AuNR-coated liposomes showed a color shift as opposed to bare liposomes, indicating the main mechanism of drug release was reversible phase-transition of lipid bilayer. We assume the local temperature of the lipid bilayer is over 75 °C as previously shown because the liposome composition is the same.

Overall, we have demonstrated micron-sized drug-encapsulated liposomes were light-activated to release drug, which suggests that these liposomes can be used to deliver drug in a repetitive and on-demand manner with the pulse laser as an on/off switch. By tuning the laser duration, the released amounts of drug from liposomes would be controlled in a precise way. Further studies are needed to examine the post-irradiation properties, e.g., photostability and post-irradiation stability of AuNR-MTX liposomes, for long term applications. For the safety of using laser, one should carefully consider its applications. Depending on the body part and applications, the laser parameters including power, duration, pulse repetition frequency, and pulse duration can vary. Our calculation for pupil and retina in the human eye is in our previous publication [28].

4. Conclusions

We proposed a formulation for micron-size liposomes which were coated with AuNRs and encapsulated with MTX, exhibiting long-term stability (up to at least 6 months). These AuNR-MTX liposomes can perform repetitive drug releases with multiple laser-irradiation cycles (5 s per cycle), and the released amount was proportional to laser cycles. We demonstrated that the drug releases resulted from a reversible phase-transition process induced by the plasmonic heating of AuNRs. In addition, the study has provided an enhanced understanding in the mechanism of laser-triggered drug release. The study is significant and fundamentally advances the field because the results showed feasibility of precisely controlled drug release utilizing the liposomes as long-term repetitive therapeutics and suggested strategy how to determine laser parameters for optimal drug release for each application.