Introduction

Neurodegenerative diseases and brain cancers are among the leading causes of mortality, disability, and hospitalization worldwide. Despite progress in therapeutic development, effective treatment for Central Nervous System (CNS) disorders remains limited, largely due to the challenges imposed by the blood–brain barrier (BBB)1. The BBB is a dynamic interface that regulates molecular transport into the brain, preventing the entry of harmful substances while simultaneously restricting many therapeutic agents2. Pathological alterations, such as those accompanying primary brain tumors, can further modify its structure and permeability. According to the World Health Organization (WHO), glioblastoma (GBM) is the most prevalent and aggressive primary brain tumor, accounting for about 15% of all intracranial neoplasms and up to 75% of astrocytic tumors3,4. GBM typically arises from astrocytic glial cells, exhibiting uncontrolled proliferation, high invasiveness, and resistance to therapy. Understanding the physiology of the BBB has been fundamental for designing innovative drug delivery systems aimed at improving CNS drug accessibility5.

Numerous approaches have been developed to overcome BBB-related barriers, including chemical modification of drugs and modulation of BBB permeability6. An ideal delivery system should be efficient, safe, non-invasive, and capable of achieving localized drug release to minimize systemic toxicity. Drug delivery strategies can broadly be classified as invasive or non-invasive. Invasive methods, which involve direct delivery into brain tissue, often bypass the neurovascular unit (NVU) and are associated with considerable procedural risks and low targeting efficiency7,8. Non-invasive methods, on the other hand, utilize endogenous transport mechanisms or external physical stimuli to facilitate drug penetration. These include receptor-mediated transcytosis, neurotropic virus-based systems, cell-penetrating peptides (CPPs)9, nanoparticulate carriers such as liposomes and exosomes10, intranasal administration, and ultrasound (US)-mediated delivery11.

Among CPPs, the gH625 peptide, derived from glycoprotein H of Herpes simplex virus type I, has attracted interest as a versatile carrier for enhancing cellular uptake and intracellular delivery. The peptide interacts electrostatically with negatively charged phospholipid membranes and can internalize through both direct membrane translocation and energy-dependent endocytosis12. Although evidence of the exact internalization mechanism is scarce and elusive, it has been commonly accepted that more than one mechanism may be involved13. During direct penetration, gH625 adopts an α-helical conformation that transiently perturbs the lipid bilayer14, while its hydrophobic residues (glycine, leucine, alanine, tryptophan, tyrosine) stabilize membrane binding. Arginine and histidine residues facilitate lipid interaction and oligomerization, enhancing the peptide’s fusogenic activity, an effect magnified by histidine addition at the N-terminus15. Previous studies have demonstrated that gH625 can facilitate the transport of conjugated molecules across the BBB via transient transcytosis without compromising barrier integrity16,17. Although gH625 has been reported to assist in BBB transport, the present study does not investigate BBB permeability directly. Instead, this work focuses on the in vitro evaluation of gH625-mediated enhancement of cellular uptake and anticancer activity in glioblastoma cells.

Recent advances in “smart drug delivery” have emphasized external stimulus–responsive systems capable of controlled, localized drug release with high biocompatibility18. Ultrasound (US)-triggered delivery represents one such strategy, enabling spatiotemporal control via acoustic cavitation19. When US is applied to a targeted area, oscillating gas-filled agents such as microbubbles (ultrasound contrast agents, UCAs) generate transient membrane pores, a process known as sonoporation, which facilitates drug and gene entry into cells20,21,22,23. The extent of drug delivery is influenced by parameters such as acoustic intensity, exposure duration24,25, microbubble size, shell composition, and proximity to target cells26,]27. Microbubbles, composed of an outer shell and gas core28,29 are clinically used UCAs that enhance diagnostic imaging at medical frequencies (1–14 MHz)30,31. Upon ultrasound exposure, they undergo acoustic radiation and cavitation, which can improve vascular permeability and trigger local drug release32. However, traditional microbubbles have limited circulation times (3–5 min for SonoVue™) and short stability under acoustic stress33. To overcome these limitations, phase-convertible nanodroplets have been developed as precursors that transform into microbubbles under ultrasound stimulation, offering improved backscatter and prolonged activity34.

As a therapeutic model compound, amygdalin (C₂₀H₂₇NO₁₁; MW 457.42) is a naturally occurring cyanogenic glycoside found abundantly in the seeds of rosaceous plants35. It can hydrolyze under the action of β-glucosidase enzymes to produce benzaldehyde and hydrocyanic acid (HCN), the latter contributing to its cytotoxic and anticancer properties36. Amygdalin exerts its anticancer effects through multiple mechanisms, including inhibition of the Akt-mTOR pathway, modulation of integrin and catenin expression, induction of oxidative stress, and activation of apoptosis via caspase-3, Bax, and PARP cleavage37,38. It can also arrest the cell cycle at the G0/G1 phase by downregulating cyclin A and CDK2, thereby suppressing tumor cell proliferation39. Although several adverse effects of amygdalin have been claimed, due to reports of systemic cyanide toxicity and inconsistent clinical outcomes. However, despite these paradoxes, studies reported its selective cytotoxicity against several tumor cell lines, including liver, cervical, prostate, MCF-7 breast cancer, and SK-BR3. This supports its potential as a biologically active compound worth re-evaluating under controlled delivery conditions40,41,42.

Therefore, in this study, we synthesized gH625-tagged pre-PGS phase-convertible nanodroplets to enhance the delivery and therapeutic efficacy of amygdalin (a targeted nanocarrier system) in glioblastoma cells. Using the U87 glioblastoma cell line as an in vitro model, we assessed the effect of peptide tagging on cellular uptake and cytotoxic potential (Fig. 1). To the best of our knowledge, this combination has not been reported elsewhere.

Fig. 1
figure 1

Schematic representation of US-mediated drug delivery using gH625-tagged Pre-PGS nanodroplets.

Full size image

Materials and methodology

Materials

Ethanol (99.9% purity), glycerol (99% purity), sebacic acid, Tween 80, Span 60 (99% purity), NHS (N-hydroxysuccinimide), and EDC EDC (1-ethyl-3-(-3-dimethylaminopropyl) carbodiimide hydrochloride) were purchased from Sigma-Aldrich. Perfluoropentane was purchased from Shanghai Tianfu, and PBS tablets were from Oxoid UK. Synthetic Peptide gh625 (H2N-HGLASTLTRWAHYNALIRAF-CONH2) was purchased from AB-clonal, Penicon, while the ultrasound Gel pad (Aquaflex) was purchased from Parker Laboratories. The cell lines (U87) were generously provided by Cancer Cell Culture and Precision Oncomedicine Lab (C3POL). All the experiments on the U87 cell line were performed in the cell culture lab (C3POL) of Khyber Medical University.

Synthesis of pre-PGS nanodroplets

Pre-PGS nanodroplets comprised of two phases: an organic phase (100%w/v) of yellowish sticky pre-PGS in ethanol and an aqueous phase of PFP (perfluoropentane). Pre-PGS was synthesied via a melt condensation reaction using an equimolar concentration (1:1) of sebacic acid and glycerol in a three-neck flask, which was heated for 15 min to form a homogeneous mixture. The reaction was then proceeded for 3 h under continuous nitrogen supply at 180 °C. A yellowish sticky pre-polymer was obtained upon completion of the reaction, which was then stored at -20°C until further use. The aqueous phase, 5% PFP (500 µl) of the nanodroplets, was prepared by sonicating (probe sonication at 60% amplitude) PBS (10 mL) with 20 µl of Tween 80 as a surfactant at 1 1-minute cycle of sonication with a 1-minute break in an ice bath. The process was continued for 10 min. It was then extruded through a 0.22 μm filter paper. Using a solvent displacement procedure, pre-PGS phase convertible nanodroplets were synthesized43. To the 100% w/v organic phase, prepared from 2 g of pre-PGS dissolved in 2 mL of ethanol (organic phase), 180 µl of Span 20 was added and mixed. After extruding the emulsion, it was added dropwise to the organic phase, followed by solvent evaporation under continuous magnetic stirring at a moderate speed (≈ 500 rpm) for 3 h at room temperature. The resulting nanodroplets were then washed three times with distilled water by centrifugation at 1000 rpm. For drug-loaded microbubbles, synthesis was achieved by co-dissolving 10 mg/mL of amygdalin and the organic phase (pre-PGS) at 1 g/mL in ethanol, followed by the solvent displacement method44. The resulting phase convertible nanodroplets were then washed three times with distilled water by centrifugation at 1000 rpm for 10 min.

Peptide tagging

Synthetic peptide gh625, having sequence (FITC-HGLASTLTRWAHYNALIRAF-GGG-CONH2)15, was labeled with fluorescent dye FITC. The peptide was tagged to pre-PGS nanodroplets using cross-linkers, NHS-EDC (5 mg/mL each). These were added sequentially to the 4mL of pre-PGS organic phase at the mentioned concentrations under stirring, followed by the addition of the peptide at a concentration of 50 µM (final concentration of peptide in the reaction mixture). The reaction was conducted under continuous stirring for 2 h at 37 °C and at a pH of 5.5. Stirring was performed to ensure the solvent evaporated. The pellet was then washed three times at 1000 rpm for 10 min with deionized water to remove any unloaded peptide. These synthesised peptide-tagged pre-PGS were then mixed with the aqueous phase via the solvent displacement method to form nanodroplets.

Drug encapsulation efficiency

To determine the amount of drug encapsulated in the shell, the phase convertible nanodroplets are first disrupted to release the encapsulated drug. For this, a methanol-water45 mixture was added to the pellet and vortexed. The chilled methanol-water-containing pellet was centrifuged at 5000 rpm for 25 min to release the drug. The supernatant was collected, and UV-spectrophotometric analysis was performed at 255 nm.

For reference, a standard solution was prepared from a chilled solvent containing methanol and water in a concentration of 15:85% was prepared at 4 °C. Five different concentration solutions were prepared by dissolving 25 mg, 20 mg, 15 mg, 10 mg, and 5 mg of amygdalin in 4 mL of a methanol–water solution. It was stirred for a few minutes to achieve homogeneous solutions with different concentrations. UV spectrophotometric analysis was performed on them to develop the standard curve. The amount of drug encapsulated in the shell of pre-PGS phase convertible nanodroplets was calculated using the formula:

$${\text{Encapsulation efficiency}}\% = {\text{Concentration of amygdalin in medium }} \div {\text{Concentration of amygdalin added initially}} \times 100\% .$$

Fourier-transformed infrared spectrometry

The potassium bromide disk technique was employed for the FTIR analysis of the material.). The sample was added to the KBr disc, which was then placed in the FTIR sample holder. Analysis was carried out at a resolution of “4 cm− 1” and a wavenumber of “400–4000 cm− 1”. The spectrum was analyzed using Spectrum-100 software (Perkin Elmer, USA).

Water contact angle measurement

Water contact angle measurement was performed using the Sessile drop method, in which 3 µL of pure water was poured onto a slide covered in pre-PGS nanodroplets. A Fibro DAT 1100 (Sweden) was used to obtain all measurements46. Using the ADVANCE software installed in the shape analyser, the angle was measured between the interfaces. We measure the Contact angle by repeating the same procedure to assess changes in hydrophobicity and hydrophilicity after introducing amygdalin into pre-PGS microbubbles.

Drug release study of amygdalin from pre-PGS MBs

A dialyzed release study of Amygdalin was performed in the absence and presence of ultrasound (0.9 MI and 3.2 MHz) by suspending 1 ml of the drug-loaded pre-PGS nanodroplets pellet in 400 ml of PBS at pH 7.4 and 37 °C. At time intervals (0, 0.5, 1, 1.5, 2, 2.5, 4, 6, 8, 16, 24, 36, 48, and 72 h) the aliquots were taken, and UV-visible spectrophotometer analysis was performed at λ max 255 nm using phosphate buffer as blank.

Morphological characterization

Zeta sizing and zeta potential

Using a zeta size analyzer, phase convertible nanodroplets were analysed to determine the zeta potential, average size distribution, and polydispersity index (Malvern Zeta Sizer ver 7.12, UK, Serial No MAL1168467). Samples were pipetted into a plastic cuvette after being diluted in PBS, and readings were obtained using a zeta analyzer at ambient temperature. The Malvern zeta capillary cuvette was utilised for sample analysis for zeta potential measurements.

Scanning electron microscopy (SEM)

SEM-JEOL model no. JSM-6490LA was used for scanning electron microscopy (SEM) analysis, which is required for morphological characterization. Samples were diluted in PBS and then sonicated in a bath at 37 °C to prepare them for SEM. It was applied in a drop form to the slide piece, allowed to dry, and then gold sputtered onto it. Images were captured when a 10 kV accelerating voltage was applied and at various magnification levels.

Confocal microscopy

For confocal microscopy, the oil immersion method was used, in which 10 µl of nanodroplet suspension was placed onto a slide sealed with a coverslip. The slide was observed under an inverted microscope and was subjected to excitation wavelength and the result was obtained at an emission wavelength of 490–520 nm.

In-silico methodology

Retrieval of target and ligand

Target proteins Bcl-2 and MDM-2 were retrieved from the Protein Data Bank (PDB code: Bcl-2 4LVT and MDM-2 1T4E). The target molecules were saved in pdb format from rcsb.org. Now save the ligand (Amygdalin) molecule from PubChem (PubChem CID-656516) in sdf format.

Protein structure purification

Prepare the protein using the AutoDock Tool. Delete water molecules as they interfere with the docking while present in the pocket region. Add hydrogen atoms that are polar only and Kollman charges, and then save the file in the pdbqt format.

Ligand preparation

The ligand was downloaded from PubChem in SDF format and subsequently converted to PDB format using PyMOL software. The resulting PDB file was then converted to PDBQT format using AutoDock Tools. The ligand was thus prepared for docking.

Molecular docking

Molecular docking studies were performed using AutoDock Vina 1.5.7 to predict the preferred binding conformations and interactions between the amygdalin and target proteins. Prior to docking, both ligand and protein structures were prepared by removing water molecules, adding polar hydrogen atoms, and assigning Kollman charges. Amygdalin structure was prepared and converted to PDBQT format in AutoDock Tools, which automatically optimizes ligand geometry during charge assignment and torsional setup. A grid box was defined around the active site to restrict the search space for ligand binding. For 4LVT, the grid box dimensions were 54 × 72 × 56 points with center coordinates of x = 6.584, y = − 1.499, z = − 4.115, while for 1T4E, the grid box dimensions were 44 × 50 × 70 points with center coordinates of x = 43.790, y = 10.990, z = 29.820.

Docking simulations were carried out with an exhaustiveness value of 8, and binding affinities were estimated using Vina’s empirical scoring function, which reports the predicted binding free energy (kcal·mol⁻¹). The resulting ligand–protein complexes were visualized and analyzed using PyMOL to evaluate binding affinities, hydrogen bonding, and hydrophobic interactions. The most favorable binding poses were selected based on the lowest binding energy and stability of interactions within the active site.

Apparatus and setting of ultrasound machine

A GE LOGIQ Book XP portable ultrasound machine was used in the experiments with a phased array probe having a frequency range of 1-5 MHz. Ultrasound Gel replaced the air between the probe and the 96-well plate. The gel pad was placed above the 96-well plate, followed by the addition of gel to ensure there would be no entrapped air. A probe with a frequency set at 3.2 MHz and 0.9 MI47 was applied to the shell, entrapping the drug in nanodroplets.

In-vitro cytotoxicity (diffusion-based)

Cytotoxicity analysis of different groups of phase convertible nanodroplets was performed using the crystal violet drug assay method. U-87 cells were seeded in a 96-well plate (10,000 cells per well) and incubated for 24 h. Variable concentrations such as 1*106, 1*105, and 1*104 MBs per ml of different groups of phase convertible nanodroplets (pre-PGS microbubbles-MB, amygdalin-loaded microbubbles-MD, and peptide-tagged drugs loaded microbubbles-MDP) were prepared. 200 µL of these concentrations were consecutively added to the 96-well plate and incubated for 72 h (Amygdalin, MB, MD, MDP). Images of the plate were taken after 24,48, and 72 h and then rinsed with 200 µL of PBS before fixation with 4% formaldehyde solution and incubated for 10 min at room temperature. Crystal violet dye was added to the wells and incubated for 10 min at room temperature. The dye was removed and washed three times with PBS, 200 µL. After 24 h, add 20 µL of acetic acid to the wells and then read the plate using the ELISA GENEX reader at 480 nm, and 630 nm.

In-vitro cytotoxicity (ultrasound mediated)

Cells were seeded in a 96-well plate (10,000 cells per well) and incubated for 24 h. The U87 cell line was exposed to an MI of 0.9 with a probe frequency of 3.2 MHZ for 30 s after treatment with Amyg-loaded and gH625-tagged Amyg-loaded MBs. In this experimental setup, the cells were divided into five different treatment groups, such as US only, Amyg only, US + MB, US + MD, and US-MDP, in three different concentrations, such as 1*106, 1*105, and 1*104 MBs per ml of different groups of phase convertible nanodroplets (200 µl of each concentration were added to the wells). An ultrasound gel pad was placed between the probe and the 96-well plate, as it acts as a flexible spacer and provides stable contact between the probe and the plate surface, thereby enhancing the overall B-mode ultrasound diagnostic effect48. In addition to the gel pad, ultrasound gel was also used as a coupling medium to replace the amount of entrapped air between the two surfaces, which would otherwise disrupt the travelling of ultrasound waves. The overall depth of the probe was set to match the depth of the cranial submandibular ultrasound window. Cells were then incubated for 72 h at 37 °C in a CO2 incubator (amygdalin-encapsulated MB) to assess the drug’s cytotoxic effect.

Cells were washed after 72 h and fixed with 4% formalin (200 µl). Crystal violet dye was then added, and the mixture was incubated at room temperature for 10 min. After staining, cells were washed three times with 200 µL PBS and placed at room temperature for 24 h. 200 µL of acetic acid was then added, and the sample was read by the ELISA GENEX reader at 630 nm, 450 nm, and 480 nm.

In-vitro fluorescence analysis

To investigate the uptake of gH625-tagged phase-convertible nanodroplets by U87 cells, coverslips were seeded with 10,000 cells each and incubated overnight. The cells were washed with 300 µL PBS (per coverslip capacity) and examined under a microscope. Upon reaching confluency, the coverslips were treated with gH625-labeled nanodroplets at a concentration of 1 × 10^6 bubbles/mL and incubated overnight. The cells were then washed with 300 µL PBS, fixed with 4% formalin for 10 min, and washed three additional times with PBS.

$${\text{Efficiency }}\% {\text{ }} = \frac{{Drug~loaded~in~medium}}{{Drug~conc.~initially~used}} \times100$$

Nuclei were counterstained with 200 µL DAPI (4′,6-diamidino-2-phenylindole) for 10 min, followed by two PBS washes. The coverslips were mounted upside-down on glass slides using mounting medium, and the edges were sealed with nail varnish. Fluorescence imaging was performed using a microscope equipped with a 40× objective lens, using the FITC channel (excitation: 495 nm, emission: 519 nm) and the DAPI channel (excitation: 350 nm, emission: 461 nm).

Statistical analysis

GraphPad 5.01 and OriginPro 9.0 were used for statistical analysis. All the experiments were performed in triplicate. Two-way ANOVA was conducted to assess the variance between the control and experimental groups, as well as within the experimental groups, and was followed by the Bonferroni post hoc test. The results are presented as mean ± SEM.

Results and discussion

Amygdalin encapsulation efficiency

For calculating the encapsulation efficiency of Amygdalin (drug), and a standard curve was constructed by measuring UV absorbance at 255 nm. A plot was constructed with absorbance of the Y-axis and concentration of amygdalin (mg/ml) on the X-axis (Fig.S1 of supplementary Data). The free drug was calculated using the given equation: y = mx + b.

$${\text{x}} = 16.02\;{\text{mg/ml}}$$

The concentration of amygdalin released from MBs was calculated using the above equation.

The initial concentration of amygdalin added during the synthesis of pre-PGS phase convertible nanodroplets was 20 mg/mL. The encapsulation efficiency in the MBs was calculated using the above formula, which was 80.10%. Amygdalin, an anti-tumor drug, was encapsulated in the shell of pre-PGS phase convertible nanodroplets using solvent displacement44 followed by peptide tagging. The drug was reported to cause anticancer action by inhibiting the intrinsic apoptotic pathway49.

FTIR of amygdalin encapsulated gh625 tagged nanodroplets

FTIR analysis of pre-PGS phase convertible nanodroplets was performed to understand different types of bonding in their formation. Phase convertible nanodroplets were classified into different groups for chemical characterization. These groups are pre-PGS phase convertible nanodroplets MBs, Amygdalin (for reference), Amygdalin-loaded phase convertible nanodroplets MD, and gh625-tagged amygdalin-loaded microbubbles. The characteristic peaks associated with bond stretching and vibration involve C-F bond vibrations of PFP in the range of 1100 cm− 1 to 1460 cm− 1. The carbonyl stretching (C = O) of the ester bond was observed as a peak at 1731 cm− 1, shifting slightly in MDP to 1551 cm− 1 due to the peptide C = O stretch (amide I) or the amide C = O stretch. The small band appears at 2249 cm− 1 in MD and 2258 cm− 1 in MDP, indicating the presence of a CN/nitrile group, characteristic of amygdalin50. This represents that the amygdalin has been loaded into the microbubbles.

Furthermore, the broader band at 3398 cm− 1 corresponds to the stretching of the hydroxyl groups in the alcohol moieties of pre-PGS. In MD, the bands at 3454 cm− 1 refer to the stretching of primary and secondary OH groups in amygdalin. Aliphatic C-H stretching was observed in the form of two peaks in the region of 2926 cm− 1 and 2858 cm− 1. Peaks of MB, MD, MP, and Amyg alone are shown in the Fig. 2a below.

Fig. 2
figure 2

FTIR analysis of Nanodroplets-MBs (Green), Amygdalin-Amyg (Red), Amygdalin-loaded nanodroplets-MD (Blue), and Amygdalin-loaded gh625-tagged nanodroplets-MDP (Pink) Fig. 2b-Water contact angles of (a) Amygdalin, (b) pre-PGS MBs-Nanodroplets, (c) Amygdalin-loaded MBs, (d) Peptide-tagged and drug-loaded nanodroplets-MDP, Triplicate n = 3 Fig. 2c-Cumulative drug release percent from pre-PGS nanodroplets (With and without US), performed in triplicate n = 3.

Full size image

Water contact angle measurement

The water contact angle was measured for different groups of phase convertible nanodroplets and Amygdalin to assess their hydrophobic/hydrophilic properties. Among them, MB has a water contact angle of 48.5° due to the hydroxyl group of the glycerol, whereas for amygdalin, it is 20.8°, which is attributed to the presence of a cyanide group. Upon encapsulation of amygdalin, the contact angle of MD switched to 35.7°, slightly hydrophilic, and for MDP (loading of amygdalin and gH625 peptide tagging), the water contact angle increased to 41.1° as the peptide is hydrophobic having 17% hydrophilicity (Fig. 2b). Biological membranes are lipophilic, which selectively favours the uptake of smaller molecules rather than larger one. The constituents of the peptide are considered important for its interaction and for destabilizing the target membrane51. As previously reported, the hydrophobic domain initiates the process of membrane perturbation by interaction with the cell membrane phospholipids. It adopts a helical conformation with hydrophobic constituents on one side and charged residues on the other side52. The hydrophobic nature of our peptide-tagged drug-loaded nanodroplets supports the transport of the carrier across the cellular membrane.

Drug release study

The cumulative release percentage of amygdalin-loaded phase convertible nanodroplets was calculated after 72 h, and 90% of the drug was released from the construct, as given in Fig. 2c. In contrast, the diffusion-based release of amygdalin occurred at a slower rate, resulting in approximately 54.7% cumulative release over the same duration. This ultrasound-triggered enhancement in drug release is consistent with previous studies on phase-changeable nanodroplets53. Our nanodroplets exhibit an initial release of almost 3% of the drug in the absence of ultrasound after 1 h, indicating good encapsulation stability of amygdalin in the pre-PGS nanodroplets. Upon insonation (3.2 MHz, 30 s), the release rate increased markedly, likely due to acoustic droplet vaporisation (ADV) of the perfluoropentane core, which induces a liquid-to-gas phase transition. The release kinetics were maintained for 72 h, demonstrating a slow and controlled release profile following insonation, which supports the suitability of these nanodroplets for ultrasound-responsive drug delivery applications.

Morphology and surface charge of pre-PGS nanodroplets

Figure 3 shows scanning electron microscopy images of different groups of phase convertible nanodroplets, including pre-PGS MBs, Amygdalin (shell)- loaded nanodroplets, and peptide-tagged shell-encapsulated pre-PGS nanodroplets, all performed at SCME, NUST. The resolution power was 30,000X and 50,000X. Phase convertible nanodroplets exhibited spherical shapes with average sizes of MBs 32 nm (3a), MD 43 nm (3b), and MDP 55 nm (3c). Morphologically, pre-PGS phase convertible nanodroplets are core and shell nanobubbles, confirmed by performing confocal microscopy as depicted in a clear image (Fig. 3c). The shell was labeled with FITC-tagged peptide, and it gave fluorescence upon excitation, which further confirms the peptide adherence to the shell of phase convertible nanodroplets as the peptide. Hence, the Z-stack images of the phase convertible nanodroplets ensure the core-shell nature of phase convertible nanodroplets (Fig. 3b and c) with a dark core and a fluorescent shell in comparison with the microscopic image (Fig. 3a).

Fig. 3
figure 3

SEM images at 30,000X and 50,000X of (a) pre-PGS MBs (b) MD-Amygdalin loaded phase convertible nanodroplets (c) MDP-peptide tagged amygdalin loaded phase convertible nanodroplets (d) Optical image (100X) (e) Confocal Image-Core-Shell (f) Z-stack image of peptide-tagged phase convertible nanodroplets (Scalebar: 50 pixels).

Full size image

The colloidal stability of the phase convertible nanodroplets was measured using zeta potential54, while, Zeta sizing measured the hydrodynamic diameter of the molecules, unlike SEM, where dehydration takes place under a vacuum. The zeta size of the phase convertible nanodroplets measured was below 1 μm, smaller than SonoVue® and Definity® (which have sizes larger than 2 μm); hence, it is suitable for crossing biological barriers, such as the BBB. Figure 3b shows some bubbles clustering due to improper drying of the slides before exposure to the SEM and insufficient sonication during sample preparation. Furthermore, the microbubble size increases by encapsulating amygdalin and tagging peptides (Table 1). The zeta potential was calculated to be -14.96 mV for the pre-PGS microbubbles.

The SEM images revealed dehydrated nanodroplets with diameters ranging from 32 to 55 nm. However, the hydrodynamic size measured by the zeta sizer was significantly larger (< 2 μm), which is consistent with previous reports indicating volumetric expansion of perfluorocarbon-based nanodroplets during phase transition43,55. The perfluoropentane (PFP) has a low boiling point (~ 29 °C), and partial vaporization near this temperature or under ultrasound exposure can result in transient size enlargement and bubble formation, validating its phase-convertible nature.

Table 1 Zeta sizing and zeta potential of phase convertible nanodroplets MBs, amygdalin loaded phase convertible nanodroplets MD, and gH625 tagged amygdalin encapsulated phase convertible nanodroplets MDP. PDI* polydispersity Index.
Full size table

Docking results

in silico studies were performed to assess the impact of amygdalin on Bcl-2 (Fig. 4a) and MDM-2 (Fig. 4b) proteins.

Amygdalin contains the active functional moieties responsible for its therapeutic potential, whereas the other components of the nanodroplets primarily serve as carrier matrices to ensure stability and targeted delivery. When docked with Bcl-2, amygdalin exhibited a binding energy of − 7.5 kcal/mol, forming interactions with key active site residues Arg143, Gly142, Met112 and Val153. Since Bcl-2 is commonly overexpressed in glioblastoma (GBM), where it functions to inhibit mitochondrial-mediated apoptosis, its inhibition can restore apoptotic signaling and sensitize tumor cells to programmed cell death. Previous studies have shown that pharmacological Bcl-2 inhibitors, including venetoclax and obatoclax, effectively induce apoptosis in various malignancies, including GBM, by neutralizing the anti-apoptotic activity56. Therefore, the observed binding affinity of amygdalin with Bcl-2 suggests that it may mimic or augment the activity of these inhibitors, indicating a potential role in overcoming apoptotic resistance mechanisms in GBM.

Similarly, docking analysis of amygdalin with MDM-2 revealed a binding energy of − 7.0 kcal/mol, involving four key residues: Val93, Ile99, Ile61, Leu57, Leu54, and Gly58, through hydrogen bonding and van der Waals interactions. The selection of MDM-2 as a target was based on its established role as a negative regulator of the tumor suppressor p53, a pathway frequently dysregulated in GBM. Overexpression or amplification of MDM-2 promotes ubiquitination and proteasomal degradation of p53, thereby facilitating tumor progression and therapeutic resistance in p53-wild-type GBM57. Hence, the inhibitory potential of amygdalin against these two molecular targets, Bcl-2 and MDM-2, provides preliminary insight into its possible anti-glioblastoma activity and supports further exploration through experimental validation.

Fig. 4
figure 4

(a)-Protein and ligand interaction showing the binding position of Amygdalin with Bcl-2 -(b)-Amygdalin and MDM-2 interaction indicating the binding pose.

Full size image

It should be noted that the docking results presented in this study are computational predictions that provide preliminary insights into the potential interactions of amygdalin with apoptotic regulatory proteins Bcl-2 and MDM-2. These findings do not confirm actual target inhibition or modulation, which would require experimental validation through biochemical or cellular assays such as Western blotting or apoptosis analysis. Nevertheless, the observed binding affinities and interaction profiles are consistent with previously reported mechanisms37 of action for amygdalin and provide a rationale for subsequent in vitro and in vivo investigations.

Cytotoxicity and cell viability

Amygdalin-loaded phase convertible nanodroplets (diffusion-based)

A crystal violet assay has been performed to analyze the cytotoxic effect of pre-PGS nanodroplets, drug-loaded, and peptide-tagged amygdalin-loaded nanodroplets. Pre-PGS MBs were found to have excellent cell viability at 72 h without imposing any cytotoxic effect on the U87 cell line. As previously reported, the carrier molecule should be chosen in such a way that it does not impose any toxic effect and has a diameter below 8 μm58. Amygdalin-loaded phase convertible nanodroplets exhibit lower cell viability as compared to unloaded microbubbles. This is because it prevents cell growth by targeting apoptotic pathways49​. The viability percentage further decreases upon treatment with peptide-tagged amygdalin-loaded phase convertible nanodroplets (Fig. 5a). gH625 is a class of naturally occurring short peptide sequences that enhance the cellular uptake of carrier molecules and are involved in membrane interaction studies as reported in the literature59. Images were taken at 24 h,48 h and 72 h of cells incubation after treatment, as shown in the Tables S1(a-d) as supplementary data.

An enhanced cytotoxic effect was observed between the Control group and amygdalin-loaded nanodroplets-MD (p-value ≤ 0.1) at the lowest concentration, indicating a mild but not statistically robust difference. In contrast, gH625-tagged amygdalin-loaded phase-convertible nanodroplets (MDP) exhibited a highly significant reduction in cell viability across all tested concentrations (p ≤ 0.001). No cytotoxic effect/ interaction was observed between Control and Pre-PGS microbubbles-MBs with a p-value > 0.05. These findings suggest that while the amygdalin-loaded nanodroplets produced a weak cytotoxic trend, the addition of gH625 peptide and ultrasound activation substantially enhanced the biological effect, supporting the role of targeted delivery and triggered release in improving therapeutic response.

Ultrasound-triggered in-vitro cytotoxicity (Amyg-MBs)

It was previously reported that ultrasound enhances the effect of antitumor drugs when administered in combination60. Different groups of phase convertible nanodroplets were designed to determine the effect of sonoporation on cells at three different concentrations as shown in Fig. 5b.

Combining drug delivery with ultrasound insonation​ yields a more effective result when combined with microbubble-based ultrasound contrast agents, as they are destroyed in the ultrasound field and generate a Sonoporation effect. This process is known as ultrasound-triggered microbubble destruction27. Another experimental group was treated with ultrasound-triggered gH625-tagged microbubbles. High-frequency ultrasound waves (0.9 MI and 3.2 MHz) are used for the triggered release of the drug (Fig. 5b), as previously reported that high frequency and low intensity are required to ensure the tumor location61​. The cell survival rate in the ultrasound-exposed or experimental group was calculated relative to the viability in the control group. Supplementary data, having Tables S2(a-d) representing images of U87 cells after treatment.

Comparison between the cytotoxic effect of bare amygdalin, diffusion, and ultrasound-triggered drug delivery on U87 cells

Ultrasound combined with phase-convertible nanodroplets has been reported as an effective strategy for transiently opening biological barriers, thereby offering a promising approach for the targeted treatment of brain disorders. Previous studies have demonstrated that ultrasound-mediated phase-changeable nanodroplets can significantly enhance therapeutic efficacy by facilitating localized drug delivery and improving cellular uptake. For example, Cao et al.55reported a time-dependent decrease in the viability of MDA-MB-231 cells treated with drug-loaded lipid–polymer nanodroplets, where ultrasound exposure promoted drug release and enhanced cytotoxicity.

Consistent with these findings, our results showed that ultrasound had a substantial impact on cell viability. To further evaluate this, we compared ultrasound-triggered drug release with diffusion-based release from the carrier nanodroplets in U-87 cells at three different concentrations, using the Crystal Violet assay after 72 h. A significant reduction in cell viability was observed following treatment with ultrasound-activated, drug-loaded phase-convertible nanodroplets (Fig. 5c). When compared with bare amygdalin, the nanodroplets delivered amygdalin both with and without tagging gH625, resulting in a significant reduction in cell viability %. However, in the later case, this reduction is more than the former one.

Moreover, comparison of ultrasound-triggered and un-triggered (diffusion-based) treatments with free amygdalin revealed that the cytotoxic effect of the drug was markedly enhanced when combined with the cell-penetrating peptide and ultrasound activation. These findings support previous reports62 and confirm that ultrasound stimulation, in synergy with peptide-functionalized nanodroplets, enhances cellular uptake and therapeutic efficacy.

Fig. 5
figure 5

(a) Effect of diffusion-based drug release on U87 cells (b) Effect of ultrasound-triggered drug release on cell viability % (c) Comparison of Diffusion and ultrasound-triggered drug assay, *p ≤ 0.1, **p ≤ 0.01, ***p ≤ 0.001 and ns is non-significant (p-value is > 0.05). Bars represent mean ± SEM.

Full size image

It is essential to note that the present investigation employed an in vitro glioma cell model to evaluate the ultrasound-triggered cytotoxicity and cellular uptake of gH625-functionalized, amygdalin-loaded nanodroplets. The ultrasound exposure used in this study (MI = 0.9) falls within the clinically approved diagnostic range (FDA limit: MI ≤ 1.9) and has previously been reported to induce reversible blood–brain barrier (BBB) opening without eliciting cellular damage when used in conjunction with microbubbles61,63. Accordingly, the cytotoxic effects observed in our in vitro model are most likely attributable to enhanced drug internalization rather than ultrasound-induced cellular injury. Nonetheless, as this study did not incorporate a BBB or glial co-culture system, further investigations employing BBB-mimicking co-cultures or in vivo models are warranted to validate the safety, biocompatibility, and translational potential of this ultrasound-mediated drug delivery approach.

Fluorescence analysis in-vitro

Fluorescence microscopy was performed to examine the cellular internalization of gH625-labeled phase-convertible nanodroplets. U87 glioblastoma cells exhibited distinct green fluorescence distributed within the cytosol, confirming the efficient uptake of gH625-functionalized nanodroplets after 24 h of incubation (Fig. 6a–c). This enhanced internalization can be attributed to the membrane-penetrating properties of the gH625 peptide, which facilitate translocation across the lipid bilayer.

Moreover, treatment with ultrasound-triggered, amygdalin-loaded nanodroplets led to a significant reduction in cell viability compared to untreated and non-irradiated controls, indicating that increased cellular uptake contributes to improved therapeutic efficacy. The combined effect of gH625-mediated targeting and ultrasound activation likely enhanced intracellular accumulation and triggered localized drug release, thereby potentiating cytotoxicity. Similar trends have been reported in previous studies, where ultrasound exposure promoted membrane permeability and enhanced the drug delivery efficiency of phase-changeable nanodroplets64.

Overall, these results suggest that the synergistic interaction between the gH625 peptide and ultrasound facilitates both efficient internalization and controlled drug release, making this approach a promising strategy for targeted glioblastoma therapy.

Fig. 6
figure 6

In-vitro uptake of FITC-gH625-tagged phase convertible nanodroplets by U87 cells (Green) (a) DAPI stain U87 cells (blue) (b) gH625-tagged amygdalin loaded phase convertible nanodroplets (c) Merged.

Full size image

Conclusions

In conclusion, the utilization of pre-PGS gH625-tagged phase convertible nanodroplets as versatile and efficient delivery carriers, complemented by the non-invasive and cost-effective ultrasound technique, presents a promising avenue for advancing targeted therapy in central nervous system disorders. The observed reduction in cell viability and improved therapeutic response in U-87 glioblastoma cells indicate the potential of this platform to enhance intracellular drug delivery efficiency. The synergistic effect of the gH625 peptide and ultrasound in promoting cellular internalization, as demonstrated by fluorescence microscopy, supports their potential application in tumor-targeted therapy. With their combined diagnostic and therapeutic capabilities and inherent targeting ability, these phase-convertible nanodroplets offer a promising platform for addressing various central nervous system (CNS) disorders. However, the study was conducted on a single cell line, providing baseline proof of concept that needs to be further elaborated by in vivo and clinical studies. These early results laid the groundwork for more precise and targeted therapies concerning neurological health.