Ultrasound-triggered drug release and cytotoxicity of microbubbles with diverse drug attributes

Abstract

Ultrasound (US)-triggered cavitation of drug-loaded microbubbles (MBs) represents a promising approach for targeted drug delivery, with substantial benefits attainable through precise control over drug release dosage and form. This study investigates Camptothecin-loaded MBs (CPT-MBs) and Doxorubicin-loaded MBs (DOX-MBs), focusing on how properties such as hydrophilicity, hydrophobicity, and charged functional groups affect their interaction with the lipid surfaces of MBs, thereby influencing the fundamental characteristics and acoustic properties of the drug-loaded MBs. In comparison to DOX-MBs, CPT-MBs showed larger MB size (2.2 ± 0.3 and 1.4 ± 0.1 μm, respectively), a 2-fold increase in drug loading, and an 18 % reduction in leakage after 2 h at 37℃. Under 1 MHz US with a 100 ms pulse repetition interval (PRI), 1000 cycles, 5-minute duration, and 550 kPa acoustic pressure, CPT-MBs undergo inertial cavitation, while DOX-MBs undergo stable cavitation. Drug particles released from these MBs under US-induced cavitation were analyzed using dynamic light scattering, NanoSight, cryo-electron microscopy, and density gradient ultracentrifugation. Results showed that CPT-MBs mainly release free CPT, while DOX-MBs release multilayered DOX-lipid aggregates. The cytotoxicity to C6 cells induced by US-triggered cavitation of these two types of MBs also differed. DOX-lipid aggregates delayed initial uptake, leading to less pronounced short-term (2 h) effects compared to the rapid release of free CPT from CPT-MBs. These findings underscore the need to optimize drug delivery strategies by fine-tuning MB composition and US parameters to control drug release kinetics and achieve the best tumoricidal outcomes.

1. Introduction

Traditional systemic chemotherapy involves administering high doses of chemotherapeutic agents periodically to suppress tumor growth. However, this approach faces several challenges, including rapid drug clearance [1], a complex tumor microenvironment [2], [3], and heterogeneous tumor structures characterized by high interstitial fluid pressure [4]. These factors often lead to inadequate drug concentrations at the target site and suboptimal therapeutic outcomes [5], [6], [7]. Moreover, the significant toxicities associated with these drugs not only limit treatment efficacy but also narrow safety margins, highlighting the urgent need for precise and targeted drug delivery strategies.

To mitigate off-target side effects, numerous biocompatible drug delivery systems have been developed. Nanoparticles, in particular, have garnered significant attention due to their small size, prolonged circulation time [8], [9], and passive targeting capabilities through the enhanced permeability and retention effect [10]. These characteristics are especially advantageous for tumor tissue therapy.

However, the application of microbubbles (MBs) in ultrasound (US)-mediated drug delivery demonstrates significant potential for controlling drug release kinetics and can be tailored to individual patient conditions, thereby significantly enhancing treatment outcomes. This approach has emerged as a promising non-invasive strategy for targeted drug delivery in cancer therapy [11], [12], [13]. Specific acoustic pressures can induce stable cavitation, interacting with microstreaming around the MBs [14], [15]. This interaction can reversibly alter cell membrane permeability, facilitating drug entry into tissues [16], [17]. Furthermore, increasing US power may lead to inertial cavitation and localized bursts of drug release, as the rapid collapse of MBs generates considerable energy, producing liquid jets and shock waves [18], [19]. These mechanical effects can enhance drug transport across microvascular barriers, tumor stroma, and cellular membranes [20], [21]. Additionally, the turbulence, elevated temperatures, and reactive oxygen species (ROS) generated by US may further influence drug release and absorption [22], [23]. Furthermore, US holds the potential to activate drugs through the modification of their chemical properties, thereby enhancing their anticancer efficacy. This is exemplified by sonosensitizers like protoporphyrin IX, which are activated by US, leading to the production of reactive oxygen species (ROS) [24], [25]. However, achieving precise control over drug dosage and release kinetics from these MBs remains a significant challenge. Researchers are actively exploring strategies to optimize US parameters—including frequency, power density, and exposure duration—to effectively regulate drug release profiles [26].

The physical properties of MBs, including size, gas core type, shell composition, Typically, these MBs encapsulate a gas core within a stable shell made of phospholipids [27], [28], [29], polymers [30], or proteins [31], commonly using stable gases like SF₆ [32] or C₃F₈ [28]. Additionally, characteristics of the loaded drugs, and their distribution on the surface, significantly influence their acoustic behavior [27], [28]. Ensuring the stability of drug-loaded MB structures to maximize drug loading and minimize off-target effects is essential for effective in vivo drug delivery.

Research on encapsulating hydrophobic drugs in MBs is limited due to structural and size-related factors that impede their integration into the lipid membrane. High lipophilicity can lead to aggregation within the lipid matrix, reducing encapsulation efficiency [33]. While polymers are often used for their higher loading capacities, their rigid structures may negatively impact US responsiveness and targeted drug release compared to lipid-based MBs [34], [35]. Moreover, oil-coated MBs can help retain drugs but face issues like polydispersity, low stability, and impractical size distributions [36], [37].

Camptothecin (CPT) is a potent topoisomerase I inhibitor that encounters clinical challenges due to its hydrophobic nature and suboptimal pharmacokinetics [38], [39], complicating its incorporation into MBs. Literature on loading CPT and its related derivatives into MBs remains relatively scarce [40], [41], [42], [43], [44]. CPT can embed within the lipid acyl chain region [45], [46], [47]; therefore, this study utilizes long-chain (18-carbon) saturated lipids, such as DSPC, combined with DSPE-PEG2000 to enhance encapsulation efficiency and MB stability. The hydrophilic PEG chain forms a protective shell that increases the solubility of hydrophobic drugs, minimizes drug-water interactions, prevents aggregation, and prolongs circulation time, thereby improving bioavailability. Additionally, the incorporation of cholesterol enhances MB rigidity, reducing rupture risk and extending lifespan [48].

Doxorubicin (DOX), a widely used chemotherapeutic agent, operates by intercalating into DNA, inhibiting topoisomerase II, and generating free radicals [49], [50]. DOX is a hydrophobic compound with low solubility in its free form. By presenting it as a hydrochloride salt, known as doxorubicin hydrochloride (DOX-HCl), its solubility in water is significantly increased [51]. The work of Tinkov et al. (2023) has modified Dox molecules to carry a positive charge and exhibit amphiphilic properties, enabling them to form stable non-covalent complexes with anionic phospholipids on MB shells. These interactions are primarily stabilized through electrostatic attraction and hydrophobic forces [52]. There are several studies that describe the possible binding methods of DOX between the lipids of MB. The positively charged sugar moiety of DOX interacts with the phospholipid headgroup, which results in the embedding of its aglycone portion or the formation of stacks on the MB surface [52], [53].

Under US irradiation, drug-loaded MBs undergo various forms of disruption in their phospholipid monolayer membranes due to resonance with sound waves, followed by resealing. This process leads to the gradual shedding of lipid aggregates, potentially resulting in the formation of drug-loaded liposomes or micelles while also releasing free drug molecules [54], [55]. Liposomes, composed of phospholipid bilayers, provide higher drug loading capacities and the capability to carry both hydrophilic and hydrophobic agents [56], [57]. In contrast, while micelles serve as effective drug carriers with favorable biocompatibility, they exhibit lower stability and are rapidly cleared from the body [58], [59]. Due to their smaller size, free drug molecules can diffuse directly into cells [60]. In the tumor microenvironment, abnormal vascular structures enable nanoparticles (380–780 nm) to accumulate within tumor tissues by traversing the tight junctions between endothelial cells [61], [62]. Conversely, slightly larger drug molecules or other encapsulated agents enter cells via various uptake pathways, including clathrin-mediated endocytosis, caveolae-mediated endocytosis, macropinocytosis, phagocytosis, and non-canonical endocytosis [63], [64], [65]. Additionally, interactions between US and released liposomes or micelles [66], as well as the dynamics of free or encapsulated drugs in the surrounding tissue fluid, enhance drug dispersion within tissues, thereby facilitating drug uptake by tumor cells through endocytic or pinocytic mechanisms [29], [67].

Currently, research on the use of lipid-based MBs for encapsulating hydrophobic drugs is relatively sparse, particularly regarding the effects of both hydrophilic and hydrophobic drugs on the acoustic behavior and drug release profiles of MBs. This study aims to investigate the fundamental properties, leakage, and acoustic characteristics of MBs loaded with hydrophobic camptothecin (CPT) and hydrophilic doxorubicin hydrochloride (DOX-HCl), as well as the resulting released drug forms, including liposomes, micelles, and free drugs. Additionally, this study will pioneer the examination of whether the particle form of released chemotherapeutic agents influences their cytotoxic effects on C6 tumor cells. A visual summary of the key findings is provided in Fig. 1.

Fig. 1.

Summary diagram of the study.

2. Materials and methods

2.1. Preparation of drug-loaded MBs

CPT-MBs and DOX-MBs are both fabricated and optimized using the thin film hydration method, as previously published in the literature [27], [40]. To prepare the films for producing CPT-MBs and DOX-MBs, the materials used are as follows: for CPT-MBs, Camptothecin (Thermo Fisher Scientific, Waltham, MA, USA), 1,2- distearoyl-sn-glycero-3-phosphocholine (DSPC, Avanti Polar Lipids, AL, USA), 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(poly(ethyleneglycol))-2000] (DSPE-PEG2000, Avanti Polar Lipids, AL, USA), and cholesterol, mixed in a weight ratio of 9:6:7:1 and dissolved in a 4:1 mixture of chloroform and methanol; and for DOX-MBs, DSPC, 18:0 PG 1,2-Distearoyl-sn-Glycero-3-[Phospho-rac-(1-glycerol)] (DSPG, Avanti Polar Lipids, AL, USA), and DSPE-PEG2000, mixed in a weight ratio of 6:9:1 and dissolved in chloroform. Each sample is aliquoted into separate vials, and the solvent is removed through heating and evaporation to form a thin film. Then, around 800 µL phosphate-buffered saline containing 0.5 % glycerol is added to the vials with the thin films, and the samples are dispersed uniformly using an ultrasonic processor (Model 2510, Branson, NY, USA) at 60 °C. For DOX-MBs, an additional 0.5 mg of DOX-HCl (Biosynth, Staad, Switzerland) is added to each vial containing the thin film and mixed thoroughly. The resulting solution is degassed using a motor, filled with perfluoropropane (C₃F₈), and then shaken at 4550 rpm for 45 s to form CPT-MBs or DOX-MBs. To remove unloaded drugs and excess lipids, the CPT-MBs solution is centrifuged at 500 g for 1 min. The lower layer is removed and replaced with an equal volume of 0.5 % glycerol-PBS, and this process is repeated three times. For DOX-MBs, a similar procedure is followed, except the solution is centrifuged at 500 g for 2 min, and the process is repeated four times.

2.2. Characterizations of Drug-Loaded MBs

2.2.1. Size Distribution, Concentration, drug Loading, and drug leakage kinetics of MBs

The diameter and concentration of the fabricated CPT-MBs and DOX-MBs were measured using a Coulter counter (Multisizer 3, Beckman Coulter, FL, USA). The drug payload of CPT-MBs and DOX-MBs was quantified by disrupting the MB structure using a sonicator (Branson 2510, Branson Ultrasonics Corp., Danbury, CT, USA) until the solution became clear. The solution was then centrifuged at 2000 g for 5 min to remove any remaining intact MBs. The supernatant was appropriately diluted and mixed with methanol in a 1:1 ratio. Following this preparation, the amount of CPT was measured by its emission at Fluorescence 430 nm using a spectrophotometer (Infinite® 200PRO series, Tecan, AG, Switzerland), while the amount of DOX was determined by measuring its emission at Fluorescence 562 nm. The encapsulation rate was estimated by calculating the percentage of the drug payload encapsulated within the drug-loaded MBs relative to the initial amount of DOX or CPT added. Measure the amount of drug naturally leaked from CPT-MBs and DOX-MBs every 10 min at 37 °C, and plot drug release curves to analyze the pharmacokinetic characteristics of leakage.

2.2.2. Acoustic stability of MBs

To evaluate the stability of CPT-MBs and DOX-MBs, a clinical 7.5 MHz US imaging system with a 128-element linear array probe (12L5A, Terason, MA, USA) was used to measure the echo response of MBs at 37 °C. MBs were diluted in PBS to a concentration of 5 × 10⁷ MB/ml, which was identified as the optimal visualization concentration after testing, and introduced into a three-hole model made with PBS and 2 % (w/v) Ultra-Pure Agarose (Invitrogen, CA, USA). Echoic analysis of drug-loaded MBs was performed at 0, 10, 20, 30, 40, 50, and 60 min after the addition of the MBs to monitor changes in stability. Contrast-to-noise ratio (CNR) was defined by the formula where the echo signal of the MBs (AMB(t)) minus the background signal of the water sample (ABK(t)) is divided by the background signal (ABK(t)): Contrast-to-noise ratio (dB) = 20 × log₁₀[(AMB(t) − ABK(t)) / ABK(t)].

2.2.3. Stable and inertial cavitation thresholds of MBs

This experiment employed a passive cavitation detection method to capture the acoustic emissions generated by MBs. To evaluate the stable cavitation of CPT-MBs and Dox-MBs, a MB solution with a concentration of 5 × 10⁸ MBs/ml was infused at a flow rate of 10 ml per hour into a 200 mm diameter cellulose tubing. This corresponds to a calculated flow velocity of 8.87 cm/second, which is consistent with the typical range for animal blood flow, and was controlled by a syringe pump (KDS120, KD Scientific, New Hope, PA, USA). A waveform generator (AWG 2040, Tektronix, Beaverton, OR, USA) combined with an RF power amplifier (2100L, E&I, Rochester, NY, USA) was used to operate a 1 MHz focused US transducer (V302, Olympus Panametrics-NDT, USA), which was set to 500 cycles with a pulse repetition interval of 10 ms and an acoustic pressure range of 0 to 550 kPa to sonicate the MBs. The acoustic emissions generated by the MBs were collected by a 0.5 MHz focused US transducer (V301, Olympus Panametrics-NDT, USA) amplified by an amplifier (5072PR, Olympus Panametrics-NDT), and recorded by an oscilloscope (LT322, Teledyne Technologies, Thousand Oaks, CA, USA). The recorded signals were processed using MATLAB software for fast Fourier transform, and the dose of stable cavitation was quantified by comparing the peak intensities of the fundamental frequency (F₀, 1.0 MHz) and subharmonic signals (1/2F₀, 0.5 MHz) [68]. The specifications and parameters of US transducers adopted in the experiments are summarized in Table 2. To analyze the inertial cavitation activity, the receiving transducer was replaced with a 5 MHz focused US transducer (V308, Olympus Panametrics-NDT, USA). The presence of inertial cavitation was indicated by broadband signals in the spectrum. The dose of inertial cavitation was determined by calculating the spectral area within the bandwidth of the receiving transducer (approximately 4.3–4.8 MHz), excluding frequencies associated with the harmonics and ultraharmonic signals of the transducer [69]. The experimental setup for measuring ICD and SCD is shown in Fig. 3-C.

Table 2.

List of ultrasound transducers used in this experiment.

Parameters Applications	Model	Type	Central Frequency	−6 dB Bandwidth	Focal Depth
MB Stability Measurements	12L5A	Linear	7 MHz	60 %	N/A
ICD and SCD Measurements	V302	Focused	1 MHz	70.0 %	50.75 mm
	V308	Focused	5 MHz	57.1 %	51.41 mm
	V301	Focused	0.5 MHz	80.8 %	34.16 mm
MBs destruction/ Cell experiments	Homemade	Planar	1 MHz	N/A	N/A

Fig. 3.

Comparison of the stability and evaluation of the acoustic properties of CPT-MBs and DOX-MBs. (A) Conversion of an acoustic impedance image into a B-mode image. (B) Quantitative analysis of MB retention percentage based on B-mode images. (C) Schematic of ICD and SCD measurement system. (D) ICD and SCD of CPT-MBs. (E) ICD and SCD of DOX-MBs. The ICD and SCD data presented in the figure have been adjusted by subtracting the PBS background values.

2.3. Size distribution and concentration of particles released from MB Post-US treatment

2.3.1. Process of US-Induced MB destruction and drug release

To assess the drug release properties of CPT-MBs and DOX-MBs under US conditions, first, adjust the concentration of DOX-MBs and CPT-MBs to 5 × 10⁹ MBs/ml and add them to a 96-well plate. Seal the 96-well plate with a transparent film and invert it during the stimulation process. Apply US coupling gel to eliminate any air gaps between the 1 MHz US transducer and the film, ensuring that the primary energy passes through the transparent film to minimize US energy attenuation. Additionally, to further reduce the impact of the sound field, fill the gaps between the well plates with culture medium. Perform US irradiation with a sound pressure range of 0 to 550 kPa, 1000 cycles, a PRI of 100 ms, and a total duration of 5 min. The selection of these parameters is based on the optimization of MB activation and drug release efficiency as reported in the literatures [70], [71].

2.3.2. Detection of size distribution and concentration of particles released from MBs

After the US treatment, the samples were collected from the wells and centrifuge at 2000g for 5 min. Take the lower layer of the samples, dilute the solution with PBS to achieve a concentration range of 10⁶ to 10⁹ particles/ml, and then use DLS and NanoSight to measure the size distribution and concentration of MB fragments and drug particles post-US treatment.

2.3.3. Analysis of the primary structure forms of US-Induced MB drug release

To investigate the structures of the particles released from CPT-MBs and DOX-MBs, cryo-electron microscopy imaging of the destruction fragments from CPT-MBs or DOX-MBs was conducted using a FEI Tecnai G2 F20 TWIN TEM (FEI, Hillsboro, OR, USA). A 4 μl sample was then applied to the 200-mesh copper grid with a holey carbon support film (HC200-Cu, Electron Microscopy Sciences, Hatfield, PA, USA) in a 100 % humidity chamber at 4 °C and blotted dry within 3 s. Rapid freezing was performed using liquid ethane cooled by liquid nitrogen and the Vitrobot sample plunger system (FEI, Hillsboro, OR, USA). Imaging was carried out using a cryo-electron microscope in bright field mode at an operating voltage of 200 kV. Images were captured at a magnification of 50,000 x using a charge-coupled device camera (Gatan, Pleasanton, CA, USA). To investigate the drug release profiles of CPT-MBs and DOX-MBs, ultracentrifugation was performed. First, sucrose solutions at concentrations of 20 % and 40 % were prepared. The sucrose gradient was then carefully layered in an ultracentrifuge tube, starting from the highest concentration of sucrose to the lowest, with each sucrose concentration being 1 ml in volume. Subsequently, 2 ml of the sample was gently layered on top of the 20 % sucrose layer. The samples were centrifuged at speeds ranging from 160,000x g for 3 h at 4 °C. Gradient fractions were collected starting from the top (lowest density) and moving downwards (highest density). The collected fractions were analyzed for the drug contents (CPT or DOX) using spectrophotometry described above.

2.4. Us-driven MBs for enhanced chemotherapy drug cytotoxicity

2.4.1. C6 cell culture and detection of IC50 of CPT and DOX on C6 cells

The C6 cell line has been shown to exhibit significant cytotoxic effects in response to the chemotherapy drugs CPT and Dox, making it a suitable choice for this experimental study [72], [73]. C6 cancer cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10 % fetal bovine serum (FBS) and 1 % penicillin–streptomycin and maintained under standard conditions (37 °C, 5 % CO₂). For the half maximal inhibitory concentration (IC50) assay of CPT and DOX, 100 μl of a cell suspension containing 3 × 10⁵ C6 cells was dispensed into each well of a 96-well plate and allowed to adhere and grow for 24 h. Stock solutions of CPT and DOX were prepared at 10 mg/ml in dimethyl sulfoxide (DMSO) and diluted with cell culture medium to achieve final concentrations ranging from 0.02 to 20 μg/ml. The cells were treated with these drug solutions, ensuring each well contained a total volume of 200 μl. Wells containing only cell culture medium served as negative controls to account for baseline cell viability. Cell viability was assessed 24 h post-treatment using the Cell Counting Kit-8 (CCK-8) assay: 20 μl of CCK-8 solution (diluted 1:9) was added to each well, avoiding bubble formation to prevent interference with optical density (O.D.) readings. The plate was incubated for 2 h under standard conditions, and absorbance was measured at 450 nm using a microplate reader. The absorbance readings were used to calculate the percentage of viable cells relative to the untreated controls, and the IC50 values, representing the concentration of drug required to inhibit cell growth by 50 %, were determined from dose-response curves using curve fitting software (GraphPad prism).

2.4.2. Us-driven MBs for enhanced chemotherapy drug cytotoxicity

Dispense 100 μl of a cell suspension containing 5 × 10³ C6 cells into each well of a 96-well plate one day prior to the experiment. The drug-loaded MBs were diluted with the culture medium to match the IC50 concentration for C6 cells, based on the drug loading capacity of the MBs. Using the IC50 of each drug as the MB treatment baseline will facilitate precise assessment of cellular responses and toxic effects, providing a basis for comparison across different drug treatments. Add the MB-containing medium solutions to the wells, except for the control group. Wells containing only cell culture medium served as negative controls to account for baseline cell viability. A blank control (without seeding any cells into the well, just loading CCK-8 medium solution) was also included. Ensure that each well is filled to approximately 450 μl to avoid bubbles, and then seal the plate with a transparent film. US stimulation was then performed under the following conditions: 1 MHz, 200 and 550 kPa, 1000 cycles, PRI of 100 ms, duty cycle of 1 %, for a duration of 5 min. After US stimulation, remove 250 μl of the MB-medium from each well. Subsequently, culture the cells in a medium containing the drug continuously for 2 h, wash once with PBS, and replace with a drug-free medium for an additional 22 h of culture. After this period, remove the medium, wash with PBS once, and then add 100 μl of CCK-8 medium solution (CCK-8 and medium ratio, 1:9) to each well, taking care to avoid introducing bubbles that could interfere with optical density (O.D.) readings. Incubate the plate for 2 h at 37 °C in the cell culture incubator, then measure the absorbance at 450 nm using a microplate reader.

This study investigates the potential of US-driven MBs to release free drugs or drug-loaded particles, with the possibility that certain drug forms may further enhance chemotherapy drug cytotoxicity. However, in vivo experiments were not conducted. The primary goal was to explore the fundamental mechanisms of US-mediated drug release and evaluate its efficacy in a controlled in vitro environment. In vitro models, particularly single-cell setups, were used to precisely assess drug release dynamics, cellular uptake, and cytotoxicity of drug-loaded MBs. These models minimize the confounding factors typically found in in vivo systems [74], offering clearer insights into the interactions between the drugs released from the MBs and their effects on tumor cells.

2.5. Statistics

All results are presented as mean values with standard deviations calculated from at least three independent experiments. Statistical significance was determined using either a two-tailed Student’s t-test or one-way ANOVA. Differences were considered significant when the p-value was below 0.05 (p < 0.05).

3. Results

3.1. Characteristics of Dox-MBs and CPT-MBs

First, we examine whether the properties of the encapsulated drugs influence the fundamental characteristics of the two types of MBs, including their size, drug loading capacity, and drug loading stability. The results of the measurements reveal that the average sizes of CPT-MBs and DOX-MBs are 2.2 ± 0.3 μm and 1.3 ± 0.1 μm, respectively, with CPT-MBs having a larger and more broadly distributed size compared to DOX-MBs (Fig. 2-A). The concentrations of CPT-MBs and DOX-MBs are 8.6 ± 5.7 × 10⁹ MBs/ml and 30.3 ± 3.8 × 10⁹ MBs/ml, respectively, and their respective payload amounts are 6.5 ± 0.3 × 10^-8 μg/MB and 3.2 ± 0.2 × 10^-8 μg/MB (Table 1, Fig. 2-B). For drug loading stability, CPT-MBs had an initial drug leakage rate of 18.8 ± 0.1 %, compared to 30.8 ± 5.2 % for DOX-MBs. After 2 h at 37 °C, CPT-MBs leaked 26.8 ± 1.0 %, while DOX-MBs leaked 44.6 ± 4.1 %. This shows a significant difference in drug release rates between the two types of MBs at the same temperature (p < 0.05) (Table 2, Fig. 2-C).

Fig. 2.

Comparative analysis of MB properties. (A) Differences in size distribution between CPT-MBs and DOX-MBs. (B) Comparison of the drug payloads between CPT-MBs and DOX-MBs (***, p < 0.001). (C) Cumulative drug leakage between CPT-MBs and DOX-MBs. The percentage of cumulative drug leakage shows a significant difference between CPT-MBs and DOX-MBs at 120 min (***, p < 0.001). (D) Molecular structural formula of CPT and DOX. (E) Diagram of CPT or DOX molecular linkage with lipids. (F) Structural diagrams of CPT-MBs and DOX-MBs.

Table 1.

The diameters, concentrations, and payloads of CPT-MBs and DOX-MBs.

	Diameters (μm)	Concentrations (MBs/ml)	Payloads (μg/MB)
CPT-MBs	2.2 ± 0.3	8.6 ± 5.7 × 109	6.5 ± 0.2 × 10-8
DOX-MBs	1.4 ± 0.1	30.3 ± 3.8 × 109	3.2 ± 0.1 × 10-8

The binding modes of CPT or DOX with lipids (Fig. 2-E) are inferred based on the structures of CPT and DOX drug molecules (Fig. 2-D), their hydrophilic or hydrophobic characteristics, the nature of their charged functional groups, and previous literature on drug linkage with lipids [45], [46], [47], [52], [53]. Additionally, Fig. 2-F presents diagrams of CPT-MBs and DOX-MBs, reflecting the experimental results regarding their composition, size, drug-lipid covalent conjugation, and drug-loading amounts.

To assess the stability of MBs, we used brightness-mode US imaging, which relies on the strong reflection of US by gas-filled MBs. Intact MBs appeared as bright spots against a dark background. If the structure of MB was compromised, such as by gas leakage or rupture, these bright spots would either diminish or disappear [75]. B-mode imaging data revealed that the stability of CPT-MBs and DOX-MBs remained above 80 % of their initial concentration within 60 min (CPT-MBs decreased from 100 ± 0.1 % to 90.7 ± 2.3 %; DOX-MBs decreased from 100 ± 0.1 % to 85.4 ± 11.5 %) at 37°C (Fig. 3-A and B). These results suggest that most of the prepared CPT-MBs and DOX-MBs are expected to maintain sufficient structural stability for at least one hour at body temperature post-injection.

We developed a system (see Fig. 3-C) to study the stable and inertial cavitation doses of two types of MBs. The behavior of MBs under US is critical for drug delivery applications. At lower US intensities, MBs undergo steady expansion and contraction, known as stable cavitation, which allows for gradual drug release with minimal tissue damage. At higher intensities, MBs experience inertial cavitation, causing rapid drug release and violent collapse, which produces strong shock waves and microjets [70], [76]. With a 1 MHz US probe and stimulation parameters set at 100 ms PRI and 1000 cycles, we determined the stable and inertial cavitation thresholds for CPT-MBs to be 200 kPa and 400 kPa, respectively (Fig. 3-D). For DOX-MBs, the stable cavitation threshold was also 200 kPa, but inertial cavitation was not clearly observed under US conditions ranging from 0 to 550 kPa (Fig. 3-E). These results highlight that the interaction between various drugs, whether hydrophilic or hydrophobic, and the lipid shells of MBs can influence both the stability of the MBs and their acoustic properties.

3.2. Drug release of CPT-MBs and DOX-MBs under US irradiation

Following an in-depth analysis of the acoustic characteristics of two distinct MB types, we proceed to investigate the divergent patterns of drug release elicited by US. Utilizing stimulation parameters with a fixed acoustic pressure of 550 kPa, we investigate the variations in drug release profiles between the two MB variants through DLS, NanoSight, and cryo-EM. The sizes of particles measured by DLS ranged from 50-255 nm for CPT-MBs and 295–396 nm for DOX-MBs respectively (Fig. 4-A). NanoSight measurements indicated that particle sizes of CPT-MBs and DOX-MBs primarily ranged from 85-135 nm and 165–465 nm, respectively (Fig. 4-B). Moreover, cryo-EM observations revealed that CPT-MBs formed 110–280 nm bilayer liposomes upon US treatment, whereas DOX-MBs formed 80–400 nm irregular, multilayered lipid-drug complexes (Fig. 4-C).

Fig. 4.

Size distribution and structural forms of liposomes or lipid-drug complexes released from CPT-MBs and DOX-MBs after treatment with a 1 MHz US with the following parameters: PRI of 100 ms, 1000 cycles, 5-minute duration, and acoustic pressure of 550 kPa. (A) Size distribution of liposomes or lipid-drug complexes detected by DLS. (B) Concentration distribution of liposomes or lipid-drug complexes at various particle sizes detected by NanoSight. (C) Structural forms of liposomes or lipid-drug complexes captured by cryo-EM.

Nonetheless, the limitations inherent to particle size detection with DLS and NanoSight—necessitating particles to exceed 1 nm and 10 nm respectively—unable to assess the proportion of released free drug. To further illuminate the distribution of drug concentrations in the diverse release modalities from CPT-MBs and DOX-MBs, we employed density gradient ultracentrifugation using sucrose solutions of variable concentrations, thereby enabling stratification based on density disparities. Following the addition of samples collected from US-induced destruction of drug-loaded MBs, ultracentrifugation was conducted at 160,000 × g for 3 h at 4 °C using a density gradient (Fig. 5-A). Samples were collected from different depths of the centrifuge tube and sequentially divided into four layers: top, middle, bottom, and pellet, following ultracentrifugation (see Fig. 5-C). Fig. 5-A illustrates the separation status of released drug particles of CPT-MBs and DOX-MBs under acoustic pressure at 0 kPa (control group) and 550 kPa treatment after ultracentrifugation. Compared to the control group (without US stimulation), CPT-MBs treated at 550 kPa showed an increase in liposomes suspended in 20 % sucrose, although significant pellet formation was also observed (showed in Fig. 5-A). By measuring the CPT concentration in each layer using a spectrophotometer, we found that the drug was primarily present in the pellet layer, accounting for about 80.7 ± 10.0 % (Fig. 5-B). Fee CPT, like other hydrophobic drugs, tends to form pellets after ultracentrifugation. The release of the drug from DOX-MBs is due to the formation of a complex between the drug and lipid with a relatively lower density, causing the DOX-lipid complex to predominantly remain in the 20 % sucrose and PBS layers after ultracentrifugation. By measuring the DOX concentration in each collected layer, we found that the drug was primarily present in the middle to top layers, accounting for approximately 79.2 ± 4.0 % in total, indicating that the drug released from US-induced DOX-MBs existed almost entirely in the form of DOX-lipid complexes (Fig. 5 A-C).

Fig. 5.

Separate the structural forms of drug particles released from CPT-MBs and DOX-MBs induced by 1 MHz US with the following parameters: PRI of 100 ms, 1000 cycles, 5-minute duration, and acoustic pressure of 550 kPa, using density gradient ultracentrifugation. (A) The image shows the results after using density gradient ultracentrifugation with 40 % and 20 % sucrose to separate fragments and drug particles released from MBs. Black and red arrowheads indicate regions of dense liposome aggregation, while the yellow arrow points to the pellet where hydrophobic drug precipitates. (B) Percentage-based normalization of CPT or DOX content in various layers relative to the initial total payload. (C) Diagram of the solution volumes of each layer after gradient centrifugation (upper panel) and the boundary of each layer for drug concentration analysis after ultracentrifugation (lower panel). The figure also includes a simplified illustration of the drug release forms based on results from ultracentrifugation and cryo-EM analysis.

3.3. Us-driven MBs for enhanced chemotherapy drug cytotoxicity

Under US stimulation, if MBs release free drugs, these drugs can rapidly diffuse into cells. Conversely, if US induces the release of larger drug-loaded particles from the MBs, these particles need more time to enter cells via endocytosis or pinocytosis, leading to a delay in the onset of cytotoxic effects [63], [64], [65]. Therefore, we hypothesize that when C6 cells are exposed to IC50 concentrations of CPT-MBs and DOX-MBs under ultrasonic stimulation for two hours, the observed cell death primarily results from drug diffusion or direct MB-mediated delivery, rather than endocytosis, as endocytosis typically requires more than six hours [1]. These results may be directly linked to the differences in the particle sizes released by CPT-MBs and DOX-MBs.

Initially, IC50 concentrations of CPT and DOX for C6 cells were determined, as shown in Fig. 6-A. The study found that, under observation at different concentrations, C6 cells are more sensitive to the toxicity of DOX compared to CPT. Subsequently, C6 cells were treated with MBs containing IC50 doses of DOX or CPT and exposed to varying US pressures to assess immediate effects on cell viability. The data show that compared to CPT-MBs-2 hr (treated with CPT-MB for 2 h, without US stimulation) which has a 73.0 ± 8.5 % survival rate, both 200 kPa and 550 kPa US stimulation led to lower cell survival rates of 58.8 ± 3.3 % and 50.4 ± 2.0 %, respectively, showing a negative correlation with pressure, as shown in Fig. 6-B. This suggests that increased US enhances CPT cytotoxicity, possibly due to inertial cavitation of CPT-MBs enhancing intracellular drug delivery. The data confirm a predominant release of free drugs from CPT-MBs, resulting in relatively higher toxicity to C6 cells within 2 h post-treatment with drugs and US. However, after co-culturing CPT-MBs with C6 cells for 24 h, the survival rate of C6 cells decreased to 45.9 ± 1.9 %, which is approximately 4.5 % lower compared to the survival rate observed after 2 h of 550 kPa US stimulation with CPT-MBs (Fig. 6-B).

Fig. 6.

Comparison of the impact on C6 cell viability under different acoustic US pressures following treatment with CPT-MB or DOX-MB. (A) IC50 detection of CPT or DOX in C6 cells. The detection process is shown the upper panel. (B) The experimental timeline is shown above. The C6 cells not exposed to chemotherapeutic drugs serve as the control group. Experimental groups include C6 cells treated with CPT-MBs or DOX-MBs at concentrations corresponding to the IC50 values of these drugs, followed by 1 MHz US stimulation with the following parameters: PRI of 100 ms, 1000 cycles, 5-minute duration, and acoustic pressure of 0–550 kPa. After US-induced drug release, cells are co-cultured for an additional 2 h to assess cell viability. Additionally, there are groups treated with free CPT or free DOX for 2 h, as well as groups treated with CPT-MBs or DOX-MBs for 24 h for comparative analysis. Statistical significance is indicated as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001.Conflict of Interest.

Conversely, for US-induced DOX-MBs drug release, the primary release is DOX-lipid complexes, which are not readily absorbed by cells via diffusion, resulting in lower short-term cytotoxicity. Consequently, cell survival rates remained as high as 93.4 ± 4.5 % within two hours after treatment with DOX-containing MBs, even following 200 kPa or 550 kPa US stimulation, with cell survival rates at 95.4 ± 1.0 % and 92.1 ± 4.1 %, respectively. This result suggests that US-induced stable cavitation of MB may not immediately augment drug delivery due to the large size of drug-releasing particles, resulting in weaker short-term cytotoxic effects compared to free-form DOX treatment over 2 h (cell survival rate of 83.1 ± 2.3 %). Notably, after 24 h of co-culturing with DOX-MBs, the survival rate decreased to 76.6 ± 1.9 %, representing a 15.5 % reduction compared to the cell survival observed after 2 h of 550 kPa US stimulation with DOX-MBs (Fig. 6-B).

4. Discussion

The overall diagram of this study is presented in Fig. 1 to visually summarize the key results. These findings are consistent with previous studies that emphasize the importance of aligning drug properties with physical and acoustic characteristics of MBs [70], [77]. CPT-MBs are larger than DOX-MBs possibly due to hydrophobic drugs are less soluble and tend to gather at the gas–liquid interface of the MBs. This uneven distribution of the hydrophobic drug can destabilize the interface, resulting in larger MBs [78], [79]. Notably, CPT-MBs demonstrated a lower initial and sustained leakage rate at body temperature compared to DOX-MBs. This suggests differences in structural integrity and drug retention capabilities under physiological conditions, aligning with literature findings that hydrophobic drugs tend to have less leakage and reduced off-target effects [80]. Notably, while the payload of a single CPT-MB is about twice that of a DOX-MB, the payloads per unit area for CPT-MBs and DOX-MBs are 4.3 × 10^-9 μg/μm² and 5.2 × 10^-9 μg/μm², respectively, indicating only a small difference.

Our investigation into drug release dynamics under US stimulation revealed distinct mechanisms for CPT-MBs and DOX-MBs. Using DLS [81], NanoSight [82], and cryo-EM [83], we confirmed differences in the particle sizes of drugs released from CPT-MBs and DOX-MBs. To address measurement limitations, we employed density gradient ultracentrifugation [84] to separate MBs based on density. Hydrophobic drugs sedimented, while hydrophilic drugs dissolved in a 40 % sucrose solution. Drug-lipid complexes, including micelles and liposomes, were suspended in the 20 % sucrose layer. This analysis showed that CPT, being hydrophobic, was mainly found in the pellet layers, whereas DOX remained in the lower-density layers as DOX-lipid complexes [85]. Further, with US stimulation at 550 kPa, we observed that CPT-MBs undergo inertial cavitation, while DOX-MBs experience stable cavitation. This suggests that under these conditions, CPT-MBs primarily release free CPT with only a minor amount forming liposomes, whereas DOX-MBs predominantly release drug-lipid complexes with average diameters of approximately 200–400 nm.

In addition to the close relationship between the size of these drug particles and cellular drug uptake mechanisms, which affects cytotoxicity [84], [86], [87], the impact of inertial cavitation—including associated thermal effects, microstreaming, and shear forces—plays a crucial role [20], [21]. These effects lead to increased cell membrane permeability and improved drug uptake. The swift release of free drugs from CPT-MBs into cells causes immediate and potent short-term cytotoxic effects. This process is likely driven by inertial cavitation from 550 kPa US, which enhances drug penetration and exerts significant mechanical forces contributing to cell death [88]. On the other hand, the release of drug-lipid complexes from DOX-MBs under US is primarily driven by stable cavitation. This process involves weaker mechanical forces and relatively larger drug release, resulting in slower cytotoxic effects [89], [90]. After 2 h of treatment with DOX-MBs, cell survival remains above 90 % across all US pressures (0, 200, or 550 kPa), compared to an 83.1 ± 2.3 % survival rate with free DOX. Nonetheless, after 24 h of co-culturing with C6 cells, DOX-MBs result in a lower cell survival rate of 76.6 ± 1.9 %, suggesting that extended absorption through endocytosis is needed to enhance drug effectiveness. However, if the released drug is not promptly absorbed, there will be increased potential for off-target interactions, thereby amplifying the risk of adverse effects [91], [92].

The present study investigates the drug release profiles of MBs with varying drug-loading properties, activated by US-induced cavitation, and their subsequent effects on tumor cells. When drugs are released as free small molecules, their rapid cellular uptake leads to swift therapeutic responses. However, this rapid release is typically followed by quick systemic clearance, thereby constraining the therapeutic window and limiting long-term efficacy. In contrast, encapsulating drugs within carriers such as liposomes or micelles modifies their pharmacokinetics, including extended circulation time and reduced clearance via the reticuloendothelial system (RES) [93], [94]. Furthermore, surface modifications like PEGylation enhance targeting efficiency, promoting selective accumulation at tumor sites and enabling sustained drug release [95].

A critical challenge in cancer therapy remains the targeting of drug-resistant tumor cells. Tumor heterogeneity facilitates the rapid emergence of resistance mechanisms, driven by genetic mutations, drug efflux pumps, and selective pressures within the tumor microenvironment [96]. To address this, the study employed a single-cell model to evaluate the drug release dynamics, cellular uptake, and cytotoxicity of drug-loaded MBs. This high-throughput approach allowed for the generation of preliminary efficacy data while minimizing confounding factors associated with in vivo models [74]. Although in vitro models cannot fully replicate the complexity of in vivo environments, they provide valuable insights into fundamental drug delivery mechanisms and serve as effective platforms for screening and selecting promising candidates for further investigation [97].

The translation of in vitro findings into successful in vivo applications remains a significant challenge, particularly with regard to overcoming immune surveillance and metabolic clearance, which can significantly limit the therapeutic efficacy of drug-loaded MBs. Future research efforts should focus on optimizing drug release profiles, developing strategies to evade immune clearance, and enhancing the retention of drug-loaded MBs in tumor tissues. Addressing these hurdles will be crucial for improving the precision, effectiveness, and overall impact of cancer therapies utilizing advanced drug delivery systems.

5. Conclusion

This study provides a detailed analysis of MBs loaded with two drugs of differing hydrophobicities—CPT and DOX—under varying acoustic pressure conditions, revealing significant differences in drug release kinetics and cytotoxic effects. At an US pressure of 550 kPa, the MBs exhibit distinct cavitation behaviors: CPT-MBs undergo inertial cavitation, while DOX-MBs undergo stable cavitation. Additionally, DOX, aside from binding with lipids, also tends to self-aggregate in solution, further impacting its release dynamics. CPT-MBs primarily release free CPT, leading to rapid cellular uptake and strong short-term cytotoxic effects. In contrast, DOX-MBs release DOX-lipid complex, which, due to their structure, delay drug uptake and reduce immediate cytotoxicity. This study highlights the importance of optimizing MB composition and US parameters to effectively influence the cavitation behavior of MBs. Furthermore, it investigates how MBs with different drug-loading properties, when activated by US, release distinct drug profiles that induce significant secondary effects on tumor cells, ultimately enhancing therapeutic efficacy.

CRediT authorship contribution statement

Chi-Fen Chuang: Project administration, Visualization, Writing – review & editing. Chia-Wei Lin: Investigation, Writing – review & editing, Project administration. Chih-Kuang Yeh: Conceptualization, Funding acquisition, Methodology, Resources, Supervision, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.