Abstract
Purpose
Doxorubicin is an effective chemotherapeutic agent for numerous cancers; however, its clinical utility remains severely limited by cumulative dose-dependent cardiotoxicity, which is a critical challenge in cardioprotective strategies of cardio-oncology. Therefore, real-time dynamic and accurate diagnosis of DOX-induced cardiotoxicity (DIC) is essential to improve patient prognosis. We aimed to integrate the high spatial resolution and deep tissue penetration of ultrasound with the strong optical contrast of photoacoustic imaging (PAI) and the high sensitivity of fluorescence imaging (FLI) to construct an ICG-loaded targeted nanobubble platform for visualized monitoring of DIC.
Materials and Methods
A novel nanosystem for multimodal (ultrasonic, fluorescence, and photoacoustic) imaging of DIC was developed. The nanobubbles (NBs) were functionalized with indocyanine green (ICG@ NBs), and then modified with vascular cellular adhesion molecule-1 (VCAM-1) targeting peptide, enabling targeting capability (ICG@VCAM-1 NBs). We described their basic characteristics and targeting capability and then evaluated VCAM-1 expression in mouse DIC models. Multimodal imaging both in vitro and in vivo were conducted.
Results
ICG@VCAM-1 NBs exhibited excellent biocompatibility and superior stability. Western blotting results showed the significant upregulation of the VCAM-1 expression in DIC mice, and the overexpression persisted. ICG@VCAM-1 NBs possess ultrasound, fluorescence, and photoacoustic multimodal imaging capabilities both in vitro or in vivo, and can be precisely located by targeting the myocardial injury, thus realizing real-time dynamic monitoring in DIC mice.
Conclusion
We used VCAM-1 targeting peptide and ICG to establish a noninvasive and diagnostic NB platform (ICG@VCAM-1 NBs) for real-time imaging. This investigation exhibits the potential for visualized monitoring of DIC.
Keywords: indocyanine green, VCAM-1, multimodal imaging, targeted nanocarriers, cardio-oncology
Introduction
Cancer is the leading cause of premature mortality worldwide.1 Advances in cancer therapy have improved survival but have simultaneously increased the risk of cancer treatment-related cardiovascular toxicity (CTR-CVT).2,3 Doxorubicin (DOX) is an effective chemotherapeutic agent for multiple malignancies;4 however, its clinical utility is limited by cumulative, dose-, and time-dependent cardiotoxicity, posing a major challenge in cardio-oncology.5,6 Real-time dynamic and accurate detection of DOX-induced cardiotoxicity (DIC) is essential for optimizing patient outcomes.
Current clinical assessment incorporates transthoracic echocardiography (TTE),7,8 including three dimensional (3D) left ventricular ejection fraction (3D-LVEF) and global longitudinal strain (GLS), together with cardiac biomarkers9 such as cardiac troponins (cTn) and natriuretic peptides. However, LVEF has limited sensitivity for detecting subclinical myocardial injury because of its low signal intensity and high variability measurement variability.10,11 Cardio-oncology guidelines recommend GLS and cardiac troponins for early diagnosis of subclinical myocardial injury.8 Clinically, GLS has clinical significance in the early subclinical functional detection.10,12 Nevertheless, GLS is constrained by angle-dependence.11 The existing imaging methods are unable to detect cardiotoxicity at prefunctional stage of chemotherapy-related injury. Therefore, there is a critical need for a noninvasive and real-time technique for prefunctional detection of DIC.
Vascular cellular adhesion molecule-1 (VCAM-1) is a cell-surface protein involved in leukocyte adhesion and transendothelial migration during inflammation. Pro-inflammatory stimuli drive the upregulation of VCAM-1. Myocarditis has emerged as the most relevant cardiovascular toxicity in cancer treatment and is considered resulting from overactivation of T lymphocytes with inflammatory response.11 Oxidative stress and chronic low-grade vascular inflammation play key roles in the pathogenesis of chemotherapy-induced cardiotoxicity.5 Overexpression of VCAM-1 has been observed in DOX-induced cardiotoxicity.13 VCAM-1 is also associated with several cardiovascular diseases, including heart failure.14 Accordingly, VCAM-1 may serve as an inflammation-related biomarker induced by anticancer therapy.
Ultrasound molecular imaging is an emerging modality that captures pathological changes with advantages including nonionizing energy, noninvasiveness, high resolution, and low cost. Conjugating specific ligands to contrast agents such as microbubbles, nanobubbles, or nanodroplets enables molecular-level detection and earlier disease diagnosis.15 Recently, PAI has gained prominence as a hybrid technique that combines optical contrast with ultrasound-level spatial resolution for noninvasive cardiovascular imaging.16 Meanwhile, significant progress in the field of fluorescence imaging probes has been witnessed, demonstrating the great potential for high sensitivity visualizing of cardiovascular diseases.17
Indocyanine green (ICG), the first Food and Drug Administration (FDA)-approved optical contrast agent, provides both photoacoustic and fluorescence imaging capabilities.18 However, when used alone, it has the disadvantages of fast elimination and poor stability in vivo. Recent studies have shown that encapsulating ICG in liposome carriers can enhance its stability, prolong its circulation time, and thereby increase its fluorescence and photoacoustic signal intensity.19
Building on previous work, we aimed to integrate the high spatial resolution and deep tissue penetration of ultrasound with the strong optical contrast of PAI and the high sensitivity of fluorescence imaging to construct an ICG-loaded targeted nanobubble platform. ICG@VCAM-1 NBs bind overexpressed VCAM-1 and enhance multimodal imaging of DIC, enabling real-time dynamic and precise detection (Scheme 1).
Scheme 1.
Synthesis of ICG@VCAM-1 NBs and the ICG@VCAM-1 NBs-based multimodal imaging. Based on mechanical vibration, all lipid components and ICG can self-assemble to form nanobubbles with a perfluoropropane gas core. The VCAM-1 targeting peptide was attached to the surface of nanobubbles through biotin–avidin system. ICG@VCAM-1 NBs can precisely locate damaged myocardium by targeting the overexpressed VCAM-1 protein, thus realizing multimodal imaging in DOX-induced mice. Blue and Yellow dashed areas indicate outline of left ventricular wall; Blue circles indicate ROIs of the same size.
Materials and Methods
Materials
1,2-Dipalmitoyl-glycero-3 phosphate (DPPA), 1,2-dipalmitoyl-sn-glycero-3-phosphoglycerol (DPPG), 1,2-dipalmitoyl-sn-glycero-3-phosphatidic ethanolamine (DPPE), and 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) were purchased from Corden Pharma, Liestal, Switzerland. Biotinylated 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-PEG2000 (DSPE-PEG2000-Biotin) was purchased from NANOCS, Boston, MA. Streptavidin was purchased from Solarbio, Beijing, China. ICG was purchased from MedChemExpresss (MCE), NJ. VCAM-1-binding peptides (sequence: Biotin-VHPKQHRGGSKGC (FITC)) were synthesized, purified, and labeled by TGpeptide Biotech, Nanjing, China.
Preparation and Synthesis of ICG@ NBs and ICG@VCAM-1 NBs
Targeted nanobubbles were prepared basing on the mechanical vibration method and Biotin–Avidin System. A total of 11 mg of DPPC, DPPE, DPPG, DPPA, DSPE-PEG2000-Biotin, and 0.5 mg of ICG lipid mixture at a ratio of 3:3:3:1:1 were dissolved in 1 mL phosphate-buffered saline (PBS)/glycerol solution (PBS: glycerol = 9:1) under ultrasonication. After the agents were completely dissolved, the solution was transferred into an airtight vial. The air therein was then replaced by perfluoropropane (C3F8, Tianjin Institute of Physical and Chemical Engineering of Nuclear Industry, China), and the vial was mechanically shaken for 90s at 3,600 rpm in an HL–AH series amalgamator (Hangzhou Zhong run Medical Instrument Co., Ltd., China). After oscillation, the suspension was maintained at 4°C for >2 h or overnight, and transferred into a 1.5-mL EP tube. After separation by dispersion and differential centrifugation, ICG@ NBs (ICG-NBs) were obtained. Streptavidin was added to the ICG@ NBs at a ratio of 3 μg/1 × 107 ICG@ NBs, incubated at 4°C for 1 h.
Afterwards, biotin-labeled VCAM-1 binding peptide was added at a ratio of 0.32 μg/1×107 ICG@ NBs and incubated at 4°C for 1 h. Then, we obtained targeted NBs carrying VCAM-1 binding peptide and ICG, which were referred to as ICG@VCAM-1 NBs. This entire procedure was performed in the dark.
Characterization of ICG@ NBs and ICG@VCAM-1 NBs
The particle size, distribution range, Zeta potential, and polydispersity index of ICG@VCAM-1 NBs and ICG@ NBs were characterized using a dynamic light scattering (DLS) instrument (Zetasizer Nano ZS90, Malvern Instruments, UK). After dilution 100-fold with PBS, the concentration was counted using a hemocytometer. The structure and morphology of the NBs were observed using an optical microscope and a JEM-1400 transmission electron microscope (JEOL, Tokyo, Japan).
The absorption spectra of free ICG, PBS, and ICG@VCAM-1 NBs were recorded over 300–900 nm using a UV–Vis spectrophotometer (Thermo Fisher), and the absorption curves was plotted. By plotting a standard curve of ICG and recording of absorbance of nonentrapped ICG in the supernatant, the drug loading efficiency (LE) and encapsulation efficiency (EE) of ICG were calculated as follows:
EE = (Entrapped ICG/Total added ICG ×100%)
LE = (Entrapped ICG/Total amount of nanobubbles)×100%.
The ICG@VCAM-1 NBs were sealed and stored at 4°C. The stability of ICG@VCAM-1 NBs was assessed by recording its particle size and polydispersity index continuously on 0, 3, 5, and 7 d using DLS.
The lipid shells of ICG@VCAM-1 NBs were labeled with DiI (red fluorescence), and the VCAM-1 targeting peptide was labeled with FITC (green fluorescence). ICG@VCAM-1 NBs were prepared using the aforementioned method, and the coupling of VCAM-1 binding peptides to NBs was verified under a confocal laser scanning microscopy (CLSM).
All experiments were repeated at least three times.
DIC Model
Animals and Treatment
Male C57BL/6 mice were obtained from Beijing HFK (Bioscience Co, Ltd, Beijing, China). All animal experiments were performed in compliance with the relevant regulations and approved by the Laboratory Animal Welfare and Ethics Committee of the Army Medical University (NO. AMUWEC20252089). Mice were housed at 25°C and a 12-h light/dark cycle and provided with sterilized food and water ad libitum. After 1 week acclimation period, the mice were injected intraperitoneally with DOX (3 mg/kg, once a week for 6 weeks) to induce cardiotoxicity, with saline serving as a control. Multimodal imaging was conducted in two weeks after the sixth DOX injection in the mice. Animals were anesthetized with isoflurane and sacrificed by cervical dislocation upon completion of the experiment.
Cell Culture and Treatment
The rat cardiac myoblast H9C2 was obtained from the European Collection of Authenticated Cell Cultures and cultured in Dulbecco’s Modified Eagle Medium (DMEM)-high glucose (Gibco) containing 10% certified fetal bovine serum (FBS, 04–001-1A-AUS, BI) and 1% penicillin‒streptomycin (Gibco) in a 5% CO2 incubator at 37°C. H9C2 cells were treated with DOX (1 μM) for 24 h and set aside for further analysis.
Biosafety of ICG@VCAM-1 NBs
In vitro Cytotoxicity
The cells were plated in 96-well plates at a density of 5×103 cells/well and cultured in a 5% CO2 incubator at 37°C. After 24 h, the cells were treated with serial concentrations of ICG@VCAM-1 NBs at 1 × 105, 1 × 106, 1 × 107, and 1×108 bubbles/mL (n = 3). After 24 h, 10 μL of Cell Counting Kit-8 (CCK-8) solution (MedChemExpress, NJ) was added to each well. The cells were then incubated in a 5% CO2 incubator at 37°C. After a 2-h incubation period, the absorbance at 450 nm was determined using a microplate reader (Varioskan Flash, Thermo Fisher Scientific).
In vivo Safety of ICG@VCAM-1 NBs and ICG@ NBs
The in vivo safety of ICG@VCAM-1 NBs and ICG@ NBs was evaluated by tail vein injection in healthy C57BL/6 mice, followed by routine blood tests after 7 days and liver and kidney function tests after 14 days. The mice were anesthetized and the heart, liver, spleen, lungs, and kidneys were subjected to hematoxylin and eosin (HE) staining to analyze the biosafety of ICG@VCAM-1 NBs and ICG@ NBs (n = 3).
Hemolysis Test
Fresh heparin-treated mouse blood was mixed with a three-fold volume of PBS and centrifuged at 2,000 rpm for 10 min to obtain red blood cells. PBS and 0.1% Triton X-100 served as negative and positive controls, respectively. Different concentrations (1 × 108/mL, 5 × 108/mL, 1 × 109/mL) of ICG@VCAM-1 NBs, PBS, and 0.1% Triton X-100 were co-incubated with 2% volume red blood cells at 37 ±1°C for 1 h. Then, the tubes were subjected to centrifugation at 10,000 rpm for 5 min and the absorbance of the supernatants at 540 nm was measured using a microplate reader (n=4). The hemolysis ratio was calculated as follows:
Hemolysis ratio (%) = (ODnanobubbles − ODnegative control)/(ODpositive control − ODnegative control) × 100%
All experiments were repeated at least three times.
VCAM-1 Expression
To further investigate cardiotoxicity, immunocytochemistry (ICC) and Western blotting were performed to measure the expression of VCAM-1 protein in DOX-treated H9C2 cells and C57BL/6 mice. The mice in the 3W, 4W, 5W, and 6W groups were injected with DOX to 3, 4, 5, and 6 weeks respectively, and samples of the cardiac tissues were collected 1 week after the injection. The mice in the 8W group had their tissues sampled 3 weeks after receiving 6 injections of DOX (n > 6).
Targeted Binding Ability
Targeted binding ability of ICG@VCAM-1 NBs was measured via flow cytometry (ACEA novocyte, Agilent) and CLSM (LSM 880, ZEISS, Germany).
For cytometry, H9C2 cells (2 × 105 cells) were added to six-well plates for 24 h and divided into four groups: CON + ICG@ NBs, CON + ICG@VCAM-1 NBs, DOX + ICG@ NBs, and DOX + ICG@VCAM-1 NBs (n =4). The DOX + ICG@VCAM-1 NBs and DOX + ICG@ NBs groups were treated with 1 μM DOX whereas the CON + ICG@VCAM-1 NBs and CON + ICG@ NBs groups were not treated. After 24 h of incubation, the media was exchanged for media supplemented with ICG@ NBs or ICG@VCAM-1 NBs (FITC-labeled VCAM-1 binding peptide), respectively. After 1 h of coincubation, the cells were rinsed three times with PBS and assessed via flow cytometer.
H9C2 cells in logarithmic growth phases (7 × 104 cells) seeded in a 24-well plate were divided into four groups: DOX + ICG@VCAM-1 NBs, DOX + VCAM-1 antibody + ICG@VCAM-1 NBs, DOX + ICG@ NBs, and CON + ICG@VCAM-1 NBs. The relative concentrations of the NBs and DOX were 1 × 107/mL and 1 μM, respectively.
After culture overnight, the DOX + ICG@VCAM-1 NBs, DOX + VCAM-1 antibody + ICG@VCAM-1 NBs, and DOX + ICG@ NBs groups were treated with 1 μM DOX whereas the CON + ICG@VCAM-1 NBs group was treated with PBS. After another 24 h, cells in the DOX + VCAM-1 antibody + ICG@VCAM-1 NBs group was blocked with VCAM-1 antibody (1:200 Invitrogen), whereas the other groups were blocked with 5% BSA for 1 h. Afterward, the media was exchanged for media supplemented with DiI-labeled ICG@ NBs or ICG@VCAM-1 NBs.
Following incubation for 1 h, the cells were washed three times with PBS and stained with DAPI. Finally, the targeted binding of each group was observed via CLSM, respectively (DiI, Ex/Em = 405/455 nm; DAPI, Ex/Em = 549/565 nm, n = 5 or 6).
In vitro Imaging
In vitro Ultrasound Imaging
Ultrasound images of ICG@VCAM-1 NBs, ICG@ NBs, and clinical ultrasound contrast agents (SonoVue MBs) were acquired using a Vevo 2100 small-animal ultrasound imaging system (FujiFilm Visualsonics, Canada) (VEVO 2100; VisualSonics, Toronto, Canada). Different concentrations (5.0 × 108/mL, 1.0 × 108/mL, 5.0 × 107/mL, 1.0 × 107/mL, and 5.0 × 106/mL) of ICG@VCAM-1 NBs, ICG@ NBs, and MBs were placed in a 1% agarose gel well model (n > 6). B-mode ultrasound and contrast-enhanced ultrasound (CEUS, mechanical index MI = 0.12, frequency of the probe = 18 MHz) were performed, and then we continuously recorded the images of ICG@VCAM-1 NBs and ICG@ NBs for 20 min at a concentration of 1.0 × 108/mL (n = 3). The in vitro imaging intensity of two kinds of NBs or MBs was analyzed using ImageJ software (National Institutes of Health). The relationship between ultrasound gray value and concentration, as well as the decay rate, were analyzed.
In vitro Fluorescence Imaging
Fluorescent images of ICG@VCAM-1 NBs and ICG@ NBs were scanned by an IVIS Imaging Spectrum System (PerkinElmer) with a 745-nm excitation wavelength and an 840-nm emission wavelength. Different concentrations (5.0 × 108/mL, 1.0 × 108/mL, 5.0 × 107/mL, 1.0 × 107/mL, and 5.0 × 106/mL) of ICG@VCAM-1 NBs and ICG@ NBs were placed in 1.5-mL EP tubes to acquire fluorescent images (n = 3 or 4). The Living Image IVIS software (PerkinElmer) was used to quantitatively analyze the fluorescence intensities by setting ROIs of the same size.
In vitro Photoacoustic Imaging
Photoacoustic images were acquired using a Vevo LAZR (FujiFilm, Visualsonics, Canada), equipped with a transducer (Vevo LAZR LZ250) with a broadband ultrasound frequency of 13–24 MHz, producing an axial resolution of 75 μm. For the PAI in vitro experiments, ultrasound frequency was set at 21 MHz, and PA gain at 30 dB. Photoacoustic spectra were acquired in the wavelength ranging 680–970 nm with a step size of 5 nm. ICG@VCAM-1 NBs (1.0 × 108/mL), SonoVue MBs, and free ICG solution were placed in the 1% gel well model, and the best photoacoustic excitation wavelength of ICG@VCAM-1 NBs was determined by full wavelength scanning with the photoacoustic transducer (n = 6).
Then, photoacoustic images of ICG@VCAM-1 NBs (5.0 × 108/mL, 1.0 × 108/mL, 5.0 × 107/mL, 1.0 × 107/mL, and 5.0 × 106/mL) were acquired at the best photoacoustic excitation wavelength (n > 6). VevoLAB software package (VisualSonics) was used to analyze photoacoustic images. The average signal intensity within the regions of interest (ROIs) was reported as photoacoustic signal. The relationship between the photoacoustic signal value and concentrations of ICG@VCAM-1 NBs and ICG@ NBs was also analyzed.
In vivo Imaging
In vivo Ultrasound Imaging
A Philips Healthcare EPIQ 7 Ultrasound System equipped with a high-resolution linear US probe (eL18-4, 4–18 MHz) was used to perform myocardial contrast echocardiography (MCE). The C57BL/6 mice were anesthetized with 2% isoflurane and the probe was fixed by a clamp holder in position to acquire the optimal left ventricular long-axis view at the largest section. Depth (1 cm) and gain settings (when the myocardium is clear enough) were maintained throughout the whole experiment. After the B-mode ultrasound was collected, the imaging mode was adjusted to contrast-enhanced imaging mode with a low mechanical index (MI = 0.4).
To differentiate ultrasound signals from NBs bound to injured myocardium and those from NBs freely circulating in the bloodstream, we used the destruction-replenishment method:20,200 μL of ICG@VCAM-1 NBs or ICG@ NBs (1 × 108 NBs/mL) was randomly injected into the mice via the tail vein. After the contrast agent filled the myocardium, imaging was performed and a high energy flash pulse (MI > 1) was triggered manually to destroy all bound and unbound NBs. Subsequently, the system automatically switched to the real-time contrast-enhanced imaging mode. After the clearance of NBs from previous injections (more than 30 min), another type of NBs was injected in the same manner (n = 3).
The MCE videos were analyzed online using the QLab workstation. An ROI of approximately 1.5 cm2 was delineated in the middle segment of the left ventricular anterior wall. Frame-by-frame manual tracking was performed to avoid the influence of the sternum and myocardial movement. The images of the first frame before and after the flash, and 3 s after the flash were obtained, and the time-intensity curves were analyzed.
In vivo Fluorescence Imaging
Fluorescence images of mice were collected using an IVIS small-animal live imaging system, with the excitation and emission wavelengths set at 745 nm and 840 nm, respectively. Subsequently, 200 μL of ICG@VCAM-1 NBs or ICG@ NBs (200 μL, 1×108 NBs/mL) was injected into the mice via the tail vein. Fluorescence images were collected at different time periods (Pre-, 2 min, 5 min, 10 min, 15 min, and 30 min) for quantitative analysis (n = 3).
In vivo Photoacoustic Imaging
PAI of left ventricle of the mice were performed using a Vevo LAZR photoacoustic imaging system. Before random bolus injection of ICG@VCAM-1 NBs or ICG@ NBs (200 μL, 1×108 NBs/mL) via the tail vein, the PAI parameters were adjusted: laser wavelength = 780 nm, central frequency of the probe = 21 MHz, PA gain = 30 dB. The photoacoustic signal was collected continuously and dynamically at different time points (Pre-, 2 min, 5 min, 10 min, 15 min, and 30 min). After the in vivo PAI signal disappeared, another type of NBs was injected in the same manner (n = 3). Quantitative analysis of the photoacoustic signal in the ROI was performed, and the time-photoacoustic signal intensity curve was plotted.
A single-blind design and standardized ROI selection criteria were employed to minimize observer bias during quantitative analysis of in vivo ultrasound, fluorescence, and photoacoustic imaging.
Immunofluorescence Staining
In vitro expression of VCAM-1 protein induced by DOX was evaluated by immunofluorescence staining. H9C2 cells (7.5 × 104 cells) were seeded on sterilized coverslips in 24-well plates in triplicate for 24 h, and the cells were treated with PBS or DOX (1 μM). After further incubation for 24 h, the cells were fixed with 4% paraformaldehyde for 15 min, permeabilized with 0.1% Triton X-100 for 20 min, and blocked with 5% BSA for 2 h. Subsequently, the cells were incubated with anti-VCAM-1 (PA5-86042, Invitrogen) overnight at 4°C, followed by further incubation at 37°C temperature with Alexa Fluor 488-conjugated secondary antibodies for 1 h. The cells were stained with DAPI and observed by CLSM.
Western Blotting
Western blotting was conducted with the proteins collected from lysates from the heart tissue in DOX-induced C57BL/6 mice. The heart tissues were ground with liquid nitrogen, homogenized mechanically with RIPA buffer, and centrifuged at 4°C at 12,000 rpm for 15 min. The supernatant was collected carefully without disturbing the pellet and used for Western blot analysis to measure the VCAM-1 protein content. The protein concentrations were estimated by bicinchoninic acid protein assay kit, where equal concentrations of protein were loaded in each well, and tubulin was used as a control (n > 6).
Statistical Analysis
Our results are expressed as the mean ± standard deviation (SD). For in vitro experiments, n ≥ 3 independent biological replicates were performed, with three technical replicates per sample. For in vivo studies, n ≥ 3 mice per group were used. Statistical analyses included Student’s t-test for two-group comparisons, one-way analysis of variance (ANOVA) followed by Dunnett’s test for multiple-comparisons and two-way analysis of variance for grouped analysis. All the statistical analyses, line plots, and histograms were conducted using GraphPad Prism 6.0 software (GraphPad Software, San Diego, CA, USA). The significance level for all of the tests was set at P < 0.05.
Results
Synthesis and Characteristics of ICG@VCAM-1 NBs
Targeted NBs were prepared based on the mechanical vibration method to encapsulate the ICG, and the Biotin–Avidin System to bind VCAM-1 peptides to the lipid shell (Figure 1a). The chosen lipids exhibited excellent biocompatibility and superior stability, and were able to self-assemble to form the NB shell with a core filled with perfluoropropane. The obtained general NBs and ICG@VCAM-1 NBs appeared as a white and green milky solution, respectively (Figure S1). Under optical microscopy, the ICG@VCAM-1 NBs were uniformly distributed with identical size and shape, without evident aggregation and under transmission electron microscopy (TEM) were spherical with smooth boundaries (Figure 1b and c).
Figure 1.
Characteristics of the ICG@VCAM-1 NBs. (a) Illustration of nanobubbles containing VCAM-1 targeting peptide. (b) Optical microscopy image (the black arrows indicate ICG@VCAM-1 NBs, scale bar = 10 μm) and (c) transmission electron microscopy (TEM) image of the ICG@VCAM-1 NBs (scale bar = 500 nm). (d) Size distribution of the ICG@VCAM-1 NBs. (e) Particle size and f) zeta potential of NBs, ICG@VCAM-1 NBs, and ICG@VCAM-1 NBs. (g) Stability of ICG@VCAM-1 NBs (n = 4–6). (h) Characteristic absorption peaks of ICG and ICG@VCAM-1 NBs. (i) Hemolysis test (n =4). (j) VCAM-1 targeting peptides were bound to the surfaces of the ICG@VCAM-1 NBs (n = 3). The lipid shells of NBs were labeled with DiI (red), and the VCAM-1 targeting peptide was labeled with FITC (green). Datas are expressed as mean ± standard deviation (SD). Statistical analysis was performed using one-way ANOVA followed by Dunnett’s multiple-comparisons test (**p < 0.01 and ***p < 0.001: particle size vs. day 0; &&p < 0.01 and &&&p < 0.001: PDI vs. day 0).
DLS analysis revealed average diameter and zeta potential of blank NBs. ICG@ NBs and ICG@VCAM-1 NBs were approximately 358.64 ± 6.89 nm, 465.29 ± 20.57 nm, 515.82 ± 12.28 nm, and −12.45 ± 0.42 mV, −14.84 ± 0.51 mV, −20.08 ± 0.46 mV, respectively (Figure 1d–f). The dispersion coefficients were all < 0.3, indicating suitable dispersibility in aqueous solution.
The concentration of the NBs was 2.0×1011 NBs/mL via optical microscopy by hemocytometer. Moreover, the sizes, zeta potentials, and dispersion coefficient of blank NBs, ICG@ NBs, and ICG@VCAM-1 NBs are summarized in Table S1.
UV–VIS spectroscopy revealed distinct absorption peaks at 780 nm for ICG and ICG@VCAM-1 NBs, which are the characteristic absorption peaks of ICG, confirming the successful encapsulation of ICG within the NBs (Figure 1g). By plotting the standard curve of the ICG solution, quantitative measurements revealed an EE% of 80.18 ± 6.11%, and a drug loading capacity (LC%) of 3.49 ± 0.27%.
After storage for 3 days at 4°C, the particle size showed significant differences compared with day 0 (P < 0.01) (Figure 1h). The bottom liquid became noticeably turbid at day 7, indicating that some of the NBs had decomposed and ruptured.
An in vitro hemolysis assay demonstrated the biocompatibility of the NBs during intravenous injection, with hemolysis rates below 5% for different concentrations (5 × 107, 1 × 108, 5 × 108, 1×109 NBs/mL) of ICG@VCAM-1 NBs coincubated with red blood cells for 1 h, indicating that the NBs were not significantly toxic to red blood cells and were safe for intravenous injection (Figure 1i).
The ICG@VCAM-1 NBs were observed under CLSM. The ICG was encapsulated in the lipid shell. The lipids marked by DiI show red fluorescence, while the FITC-labeled VCAM-1 targeting peptide show green fluorescence. The complete overlap of the two types of fluorescence resulted in yellow fluorescence (Figure 1j), indicating that the VCAM-1 binding peptides were coupled to the surfaces of the ICG@VCAM-1 NBs.
Biosafety of ICG@VCAM-1 NBs
Furthermore, the toxic effects of the ICG@VCAM-1 NBs on H9C2 cells were analyzed using CCK-8. The results showed that ICG@VCAM-1 NBs at different concentrations (1 × 105, 1 × 106, 1 × 107, 1×108 NBs/mL) did not significantly affect the cell viability of H9C2 cells (P > 0.05). The effect of ICG@ NBs on cell viability was similar to that of the ICG@VCAM-1 NBs (Figure 2a).
Figure 2.
The in vitro and in vivo safety of ICG@VCAM-1 NBs. (a) In vitro cytotoxicity of ICG@VCAM-1 NBs and ICG@ NBs (n = 3). (b and c) hepatic function indicators and, (d and e) renal function indicators performed 14 days after vein injection of saline, ICG@ NBs or ICG@VCAM-1 NBs (n = 3). (f–i) Results of routine blood tests performed 7 days after vein injection for CON group, ICG@ NBs group, and ICG@VCAM-1 NBs group (n = 3). (j) Representative HE staining images of major organs (heart, liver, spleen, lung, and kidney) from the CON group, ICG@ NBs group, and ICG@VCAM-1 NBs group (n = 3). Datas are expressed as mean ± standard deviation (SD). Statistical analysis was performed using one-way ANOVA (NS, not significant: p > 0.05).
Moreover, the in vivo safety of ICG@VCAM-1 NBs was evaluated by tail vein injection in healthy C57BL/6 mice, followed by routine blood tests after 7 days, liver function tests, and kidney function tests after 14 days. There were no significant differences in hepatic function (Figure 2b and c), renal function (Figure 2d and e) or routine blood (Figure 2f–i) indicators among the CON group, ICG@ NBs group, and ICG@VCAM-1 NBs groups. Additionally, no significant pathological changes were observed in the heart, liver, spleen, lungs, or kidneys at 14 days postinjection between the CON group, ICG@ NBs group, and ICG@VCAM-1 NBs groups (Figure 2j).
Upregulation of VCAM-1 Expression
To evaluate VCAM-1 expression in vitro and in vivo, we performed immunocytochemistry (ICC) with DOX-treated H9C2 cells and Western blotting with DOX-induced C57BL/6 mice. Animal experiment scheme for Western blotting is shown in the Figure 3a. Immunofluorescence results indicated that the VCAM-1 expression in DOX-treated H9C2 cells was significantly higher compared with the control group (P < 0.001, Figure 3b and c). Meanwhile, the Western blotting data showed that the expression of the VCAM-1 protein greatly increased when treated with DOX to the third week (P < 0.01, Figure 3d), and the overexpression of VCAM-1 persisted continuously with the increment of treatment duration until the eighth week (2 weeks after chemotherapy; P < 0.001, Figure 3e).
Figure 3.
VCAM-1 expression in vitro and in vivo. (a) Animal experiment scheme for Western blotting. (b) Representative immunocytochemistry images and (c) quantitative analysis result of VCAM-1 protein expression level in H9C2 cells treated with or without DOX (n = 7, scale bar = 50 μm): DAPI (blue), VCAM-1 (green). (d) Western blotting results and (e) quantitative analysis of VCAM-1 expression in C57BL/6 mice treated with DOX for different time (n > 6). Datas are expressed as mean ± standard deviation (SD). Statistical analyses included Student’s t-test for two-group comparisons and one-way ANOVA followed by Dunnett’s test for multiple-comparisons (**p < 0.01; ***p < 0.001).
Binding Capacity
For the flow cytometry test, the results were consistent with those obtained by CLSM. The affinity of ICG@VCAM-1 NBs for H9C2 cells treated with DOX was significantly higher than that of the ICG@ NBs (P < 0.05, Figure 4a–c), whereas no significant difference in the affinity for H9C2 cells in control group was found between ICG@VCAM-1 NBs and ICG@ NBs (P > 0.05, Figure 4a–c).
Figure 4.
Binding capacity of ICG@VCAM-1 NBs. (a and b) Flow cytometry results. (c) Quantitative analysis of flow cytometry results (n = 4): FITC+ (fluorescently positive cells targeted by ICG@VCAM-1 NBs). (d) Representative CLSM images and (e) quantitative analysis of fluorescence intensities in each group (n = 5 or 6). DAPI (blue), NBs (DiI, red). Blocking: H9C2 cells were preblocked by anti-VCAM-1 monoclonal antibody. “+” = addition of ICG@ NBs, ICG@VCAM-1 NBs, VCAM-1 Antibody or DOX; “-” = no addition. Datas are expressed as mean ± standard deviation (SD). Statistical analysis was performed using one-way ANOVA followed by Dunnett’s multiple-comparisons test (*p < 0.05; **p < 0.01; ***p < 0.001).
After the H9C2 cells were incubated with the ICG@VCAM-1 NBs or ICG@ NBs, they were washed with PBS, and the binding of the ICG@VCAM-1 NBs and ICG@ NBs to the H9C2 cells was observed under a CLSM. A large number of ICG@VCAM-1 NBs accumulated at the periphery of the H9C2 cells treated with DOX (Figure 4d), while no obvious ICG@VCAM-1 NBs were present near the H9C2 cells in the control group (Figure 4d). After the H9C2 cells treated with DOX were incubated with the ICG-NBs, only a small amount of the NBs accumulated around the cell membranes (Figure 4d). When the cells were blocked with anti-VCAM-1 monoclonal antibody, the targeting binding ability of ICG@VCAM-1 NBs to H9C2 cells significantly decreased (P < 0.05, Figure 4d and e).
Ultrasound Imaging of ICG@VCAM-1 NBs
The ultrasound imaging results in vitro showed that the signal intensity of MBs (SonoVue), ICG@VCAM-1 NBs, and ICG-NBs increased as the concentration increased (Figure 5a and b). During 20 min of continuous dynamic imaging, the intensities of ICG@VCAM-1 NBs and ICG-NBs gradually diminished over time. After 20 min, the imaging ability remained relatively good, laying the foundation for subsequent in vivo ultrasound imaging. Linear fitting of the intensity data revealed that ICG@VCAM-1 NBs exhibited a similar decay rate with ICG@ NBs (Figure 5c).
Figure 5.
Ultrasound imaging in vitro and in vivo. (a) In vitro ultrasound imaging of MBs, ICG@ NBs, and ICG@VCAM-1 NBs (Red circles indicate ROIs of the same size). (b) Quantitative analysis of average ultrasound gray value for each group (n > 6). (c) Linear fitting of decay rate for ICG@ NBs and ICG@VCAM-1 NBs in vitro (n = 3). (d) Quantitative analysis of signal intensity of the attached NBs in CON and DOX groups after ICG@ NBs or ICG@VCAM-1 NBs injection (n = 3). (e) Representative in vivo ultrasound images and time–intensity curve in each group. Datas are expressed as mean ± standard deviation (SD). Statistical analysis was performed using two-way ANOVA followed by Tukey’s multiple-comparisons test (***p < 0.001).
Based on the characteristics of real-time NB ultrasound imaging,21 we examined the in vivo ultrasound imaging performance of the ICG@VCAM-1 NBs. The results of the comparative analysis of the ultrasonic imaging and contrast effects in the three groups are presented in Figure 5d and e.
After injection of the ICG@VCAM-1 NBs or ICG@ NBs through the tail vein, the B-mode ultrasound imaging showing the ventricle and myocardium of the mouse heart was performed. The CEUS imaging exhibited that the ICG@VCAM-1 NBs or ICG@ NBs appeared at the heart site rapidly in all three groups and filled the entire cardiac chamber, showing strong echoes and suitability for cardiac imaging (Figure 5e). After a flash burst, a significant decrease of the contrast signal in the myocardium was observed, followed by a rapid reflow of NBs.
The signal intensity (SI, expressed in dB) of the attached NBs was calculated by subtracting the SI postflash from preflash obtained from the time-intensity curve (Figure 5e). Quantitative analysis revealed that the SI of the ICG@VCAM-1 NBs was significantly higher compared with ICG@ NBs in DOX-induced model mice (7.34 ± 3.21 dB versus 0.63 ± 0.34 dB, P < 0.001, Figure 5d and e). However, there was no significant difference for ICG@VCAM-1 NBs and ICG@ NBs in normal mice (0.54 ± 0.35 dB versus 0.30 ± 0.10 dB, P > 0.05, Figure 5d and e). The results confirmed the targeted attachment of the ICG@VCAM-1 NBs to the DOX-injured myocardium in vivo.
FLI of ICG@VCAM-1 NBs
The fluorescence signal was strongest when the concentration of the ICG@VCAM-1 NBs and ICG-NBs was 5×107 NBs/mL, weakening at higher concentrations because of fluorescence quenching.22,23 Although the strongest fluorescence signal was observed at a concentration of 5×107 NBs/mL in vitro (Figure 6a and b), a concentration of 1×108 NBs/mL was chosen to account for NB dilution by blood in vivo and the results of in vitro ultrasound imaging.
Figure 6.
Fluorescence imaging in vitro and in vivo. (a) Representative in vitro fluorescence imaging of ICG@ NBs and ICG@VCAM-1 NBs at different concentrations, and (b) quantitative analysis of average fluorescence signal in each group (n = 3 or 4). (c) The time-intensity curve of average fluorescence signal in ROI for each group after injection of ICG@ NBs or ICG@VCAM-1 NBs (n = 3). (d) Representative in vivo fluorescence images in each group at different time points (Blue circles indicate ROIs of the same size). Datas are expressed as mean ± standard deviation (SD). Statistical analysis was performed using two-way ANOVA followed by Tukey’s multiple-comparisons test (*p < 0.05; **p < 0.01).
The targeting ability and fluorescence imaging effect of ICG@VCAM-1 NBs on DOX-damaged myocardium were evaluated by observing the in vivo distribution of fluorescence after ICG@VCAM-1 NBs injection through the tail vein using a FLI system. FLI in vivo showed that strong fluorescence signals were observed in all three groups in the liver area at 2 min postinjection. The intensity-time curve of the fluorescence signal was parabolic, featuring a rise followed by a fall (Figure 6c). At 30 min postinjection, the fluorescence signals in the CON + ICG@VCAM-1 NBs and DOX + ICG@ NBs groups lost intensity, while that signal in the DOX + ICG@VCAM-1 NBs group remained strong, showing a significant difference (Figure 6d). Quantitative analysis showed a significant difference in the fluorescence intensity in the precordial region between the DOX + ICG@VCAM-1 NBs and other groups (DOX + ICG@VCAM-1 NBs group v.s CON + ICG@VCAM-1 NBs group, P > 0.05; DOX + ICG@VCAM-1 NBs group v.s DOX + ICG@ NBs group, P > 0.01, Figure 6c). These results indicate that ICG@VCAM-1 NBs could precisely locate the mouse heart injury tissue through VCAM-1 peptides on the nanobubbles.
PAI of ICG@VCAM-1 NBs
To effectively minimize interference from other light sources and enhancing the signal-to-noise ratio, according to the all-spectrum imaging, a wavelength corresponding to ICG’s primary absorption peak at 780 nm was chosen for subsequent PAI. The ICG@VCAM-1 NBs and free ICG solution exhibited significantly stronger PA signals than blank NBs, which indicated the important role of carrying ICG in PAI (Figure 7a and b). The effects of ICG@VCAM-1 NBs concentration on the PA signal were observed in vitro. With the increasing concentration, the PA signal intensities of the ICG@VCAM-1 NBs increased at first and then decreased (Figure 7c and d).
Figure 7.
Photoacoustic imaging in vitro and in vivo. (a) In vitro Photoacoustic imaging, and (b) quantitative analysis of PA signal intensity of NBs, ICG@VCAM-1 NBs, and ICG solution (n = 6). (c) Representative in vitro photoacoustic images of ICG@VCAM-1 NBs at different concentrations, and (d) quantitative analysis of average photoacoustic signal in each group (n > 6). (e) The time-intensity curve of average photoacoustic signal in ROI for each group after injection of ICG@ NBs or ICG@VCAM-1 NBs (n = 3). (f) Representative in vivo photoacoustic images in each group at different time points (Yellow dashed areas indicate outline of left ventricular wall). Datas are expressed as mean ± standard deviation (SD). Statistical analysis was performed using two-way ANOVA followed by Tukey’s multiple-comparisons test (***p < 0.001).
To investigate the targeting effect of ICG@VCAM-1 NBs in the injured myocardium of DOX model mice, we utilized a noninvasive tracking method using PA signals. The signals were collected after vein injection at different times to assess the ability of the NBs to target the DOX-damaged myocardial muscle. Prior to NBs injection, no PA signals were observed in all groups (Figure 7e and f). PA signals in the DOX + ICG@VCAM-1 NBs group were observed at 2 min postinjection. The intensity-time curve of the PA signal was also parabolic, in agreement with fluorescence imaging (Figure 7f). At 30 min postinjection, the PA signals in the DOX + ICG@VCAM-1 NBs group remained strong, compared with the other two groups. Quantitative analysis showed that he PA signal intensity in the DOX + ICG@VCAM-1 NBs group significantly exceeded those observed for the other two groups (DOX + ICG@VCAM-1 NBs group v.s CON + ICG@VCAM-1 NBs group, P > 0.001; DOX + ICG@VCAM-1 NBs group v.s DOX + ICG@ NBs group, P > 0.001, Figure 7e). This might be due to the binding of the VCAM-1 targeting peptide to the damaged myocardium.
Discussion
Doxorubicin is a broad-spectrum chemotherapeutic agent; however, its clinical utility remains severely limited by cumulative dose-dependent cardiotoxicity, which poses a critical challenge in cardio-oncology. Therefore, real-time dynamic and accurate diagnosis of DOX-induced cardiotoxicity (DIC) is essential. The rapid development of multimodal NBs as contrast agents has brought opportunities to address these challenges. In this study, we constructed an ICG-loaded targeted nanobubble platform (ICG@VCAM-1 NBs) for visualized monitoring of DIC, which integrated the high spatial resolution and deep tissue penetration of ultrasound with the strong optical contrast of photoacoustic imaging (PAI) and the high sensitivity of fluorescence imaging (FLI). The ICG@VCAM-1 NBs exhibited excellent biocompatibility, superior stability, and strong targeting capability, possessing the suitability for simultaneous fluorescence, ultrasound, and photoacoustic imaging in vitro and in vivo. The results of this study represents a promising strategy for pre-functional diagnostic monitoring of DOX cardiotoxicity.
VCAM-1 loading increases the negative charge of NBs, which attribute to the suitable dispersibility of ICG@VCAM-1 NBs in aqueous solution. Common absorption peaks at 780 nm for ICG and ICG@VCAM-1 NBs confirmed the successful encapsulation of ICG within the NBs. Complete overlap of fluorescence indicated that the VCAM-1 binding peptides were coupled to the surfaces of the ICG@VCAM-1 NBs. When blocked by anti-VCAM-1 monoclonal antibody, the binding ability of ICG@VCAM-1 NBs to DOX treated H9C2 cells significantly decreased, indicating that the targeting capability of ICG@VCAM-1 NBs depends on the VCAM-1 binding peptide on its surface.
DIC causes the overexpression of VCAM-1 protein in the heart. Therefore, we regarded VCAM-1 protein as an inflammatory biomarker induced by DOX, establishing a foundation for subsequent multimodal imaging of ICG@VCAM-1 NBs. The excessive expression of VCAM-1 persists after chemotherapy, suggesting that the inflammatory response in the myocardial tissue has not been effectively alleviated. The continuous presence of inflammation may be one of the important driving factors for the progression of cardiomyopathy.
The similar signal intensity with clinical ultrasound contrast agents (SonoVue) laid the foundation for subsequent in vivo ultrasound imaging. Furthermore, a similar decay rate of ICG@ NBs and ICG@VCAM-1 NBs indicated that the connection of VCAM-1-targeting peptides did not affect its stability in vitro.
Although the strongest fluorescence and PA signals were both observed at a concentration of 5×107 NBs/mL in vitro, a concentration of 1×108 NBs/mL was chosen to account for NB dilution by blood in vivo and the results of in vitro imaging.
The in vitro imaging results demonstrated that the ICG@VCAM-1 NBs possess ultrasound, photoacoustic, and fluorescence multimodal imaging capabilities. In vivo results confirmed that ICG@VCAM-1 NBs can be precisely located by targeting the myocardial injury, thus realizing real-time dynamic monitoring in DOX-induced mice. The results of in vivo competitive blocking experiments reveal that the enhanced fluorescence imaging signal indeed originates from the specific binding of VCAM-1, rather than from hemodynamic, vascular permeability or non-specific factors such as EPR effects (Figure S2).
Several limitations should be acknowledged, including a moderate sample size in the animal experiments and a relatively simplified disease model (especially only male mice were studied), which may not fully replicate the complexity of human DOX-induced cardiotoxicity. Further investigation is needed on the long-term biological safety of ICG@VCAM-1 NBs. The current imaging approach mainly focused on the chronic chemotherapy-induced cardiotoxicity rather than the acute chemotherapy-induced cardiotoxicity. Future studies should increase the sample size, adopt more complex disease models (including female mice to enhance the applicability of the research results), and systematically evaluate the pharmacokinetics of the NBs. Our research lacks direct comparative data to prove that multimodal imaging is superior to single-modal imaging. In future, we will conduct additional relevant experiments to quantitatively evaluate the synergistic effect. By combining imaging with therapeutic interventions to explore therapeutic diagnostic applications and extending to other chemotherapy drug-induced cardiac toxicity models, the translational potential of this approach will be significantly enhanced.
Conclusions
The close relationship between cancer chemotherapy and cardiovascular toxicity has been extensively researched. DOX chemotherapy-developed cardiomyopathy is a dose-dependent and irreversible side effect. Although current clinical guidelines recommend global longitudinal strain (GLS) and cardiac troponins for the early identification of subclinical cardiotoxicity, these methods mainly reflect myocardial function or non-specific myocardial injury. The VCAM-1 targeted multimodal imaging platform constructed in this study can serve as a complement to existing approaches, enabling visualized monitoring of chemotherapy-induced myocardial inflammatory molecular changes (such as VCAM-1 overexpression), and providing new molecular-level information for early warning and mechanistic research of cardiotoxicity. The VCAM-1 targeting peptide-decorated surface of the ICG@VCAM-1 NBs enable the targeting of DOX cardiomyopathy and the use as a contrast agent to diagnose DOX cardiotoxicity. The results of this study will contribute to pre-functional diagnosis of DOX cardiotoxicity, offering an effective supplementary monitoring method for diagnosis.
Funding Statement
This study was supported by the National Natural Science Foundation of China (No. 82572254), the Natural Science Foundation of Chongqing (No, CSTB2024NSCQ-MSX0974), the Chongqing Special Key Project of Technology Innovation and Application Development, China (No. CSTB2022TIAD-KPX0153) and the Talent Plan Program of Chongqing (No. 425Z2K1).
Abbreviations
DIC, DOX-induced cardiotoxicity; CEUS, Contrast-enhanced ultrasound; PAI, Photoacoustic imaging; FLI, Fluorescence imaging; GLS, Global longitudinal strain; CLSM, Confocal laser scanning microscopy; DLS, Dynamic light scattering; EE, Encapsulation efficiency; LE, Loading efficiency; HE, Hematoxylin and eosin; MCE, Myocardial contrast echocardiography; PBS, Phosphate-buffered saline; ROI, Regions of interest; TEM, Transmission electron microscopy; CTR-CVT, Cancer treatment-related cardiovascular toxicity.
Data Sharing Statement
Data is available from the corresponding author.
Ethical Approval and Informed Consent
The study was approved by the Laboratory Animal Welfare and Ethics Committee of the Army Medical University (NO. AMUWEC20252089). All institutional and national guidelines for the care and use of laboratory animals were followed.
Author Contributions
All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.
Disclosure
The authors report no conflicts of interest in this work.
References
- 1.Wild CP, Weiderpass E, Stewart BW. World Cancer Report: Cancer Research for Cancer Prevention. Lyon (FR): International Agency for Research on Cancer; 2020. [PubMed] [Google Scholar]
- 2.J LJ, H WZ, Chen L, et al. Efficacy and safety of neoadjuvant SHR-A1811 with or without pyrotinib in women with locally advanced or early HER2-positive breast cancer: a randomized, open-label, Phase II trial. Ann Oncol. 2025;36(6):651–19. doi: 10.1016/j.annonc.2025.02.011 [DOI] [PubMed] [Google Scholar]
- 3.Chen Y, Ye Z, He RQ, et al. Landscape of non-coding RNAs in cancer treatment-induced cardiovascular toxicity: from mechanistic insights to clinical implications. Semin Cancer Biol. 2025;115:16–39. doi: 10.1016/j.semcancer.2025.07.009 [DOI] [PubMed] [Google Scholar]
- 4.Bisht A, Avinash D, Sahu KK, Patel P, Das Gupta G, Kurmi BD. A comprehensive review on doxorubicin: mechanisms, toxicity, clinical trials, combination therapies and nanoformulations in breast cancer. Drug Deliv Transl Res. 2025;15(1):102–133. doi: 10.1007/s13346-024-01648-0 [DOI] [PubMed] [Google Scholar]
- 5.Avagimyan A, Madonna R, Sheibani M, et al. Translational aspects of doxorubicin-induced cardiotoxicity: what we have omitted for the past decades? Vascul Pharmacol. 2025;160:107526. doi: 10.1016/j.vph.2025.107526 [DOI] [PubMed] [Google Scholar]
- 6.Zhang S, Liu X, Bawa-Khalfe T, et al. Identification of the molecular basis of doxorubicin-induced cardiotoxicity. Nat Med. 2012;18(11):1639–1642. doi: 10.1038/nm.2919 [DOI] [PubMed] [Google Scholar]
- 7.Čelutkienė J, Pudil R, López-Fernández T, et al. Role of cardiovascular imaging in cancer patients receiving cardiotoxic therapies: a position statement on behalf of the Heart Failure Association (HFA), the European Association of Cardiovascular Imaging (EACVI) and the Cardio-Oncology Council of the European Society of Cardiology (ESC). Eur J Heart Fail. 2020;22(9):1504–1524. doi: 10.1002/ejhf.1957 [DOI] [PubMed] [Google Scholar]
- 8.Lyon AR, Lopez-Fernandez T, Couch LS, et al. 2022 ESC Guidelines on cardio-oncology developed in collaboration with the European Hematology Association (EHA), the European Society for Therapeutic Radiology and Oncology (ESTRO) and the International Cardio-Oncology Society (IC-OS) Developed by the task force on cardio-oncology of the European Society of Cardiology (ESC). Eur Heart J. 2022;43(41):4229–4361. doi: 10.1093/eurheartj/ehac244 [DOI] [PubMed] [Google Scholar]
- 9.Pudil R, Mueller C, Čelutkienė J, et al. Role of serum biomarkers in cancer patients receiving cardiotoxic cancer therapies: a position statement from the Cardio-Oncology Study Group of the Heart Failure Association and the Cardio-Oncology Council of the European Society of Cardiology. Eur J Heart Failure. 2020;22(11):1966–1983. doi: 10.1002/ejhf.2017 [DOI] [PubMed] [Google Scholar]
- 10.Bekbossynova M, Saliev T, Mukarov M, et al. Longitudinal myocardial deformation as an emerging biomarker for post-traumatic cardiac dysfunction. Life. 2025;15(7):1052. doi: 10.3390/life15071052 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Addison D, Neilan TG, Barac A, et al. Cardiovascular imaging in contemporary cardio-oncology: a scientific statement from the American heart association. Circulation. 2023;148(16):1271–1286. doi: 10.1161/CIR.0000000000001174 [DOI] [PubMed] [Google Scholar]
- 12.Lee JH, Park JH. Clinical applications of speckle-tracking echocardiography in heart failure: from diagnosis to prognostication. Int J Heart Failure. 2025;7(4):201–215. doi: 10.36628/ijhf.2025.0052 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Afrin H, Huda MN, Islam T, et al. Detection of anticancer drug-induced cardiotoxicity using VCAM1-targeted nanoprobes. ACS Appl Mater Interfaces. 2022;14(33):37566–37576. doi: 10.1021/acsami.2c13019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.F TM, Ortiz-Quintero J, Garrido-Moreno V, et al. VCAM-1 as a predictor biomarker in cardiovascular disease. Biochimica Et Biophysica Acta. 2021;1867(9):166170. doi: 10.1016/j.bbadis.2021.166170 [DOI] [PubMed] [Google Scholar]
- 15.Chen J, Wang B, Dasgupta A, et al. Aminolysis-mediated single-step surface functionalization of poly (butyl cyanoacrylate) microbubbles for ultrasound molecular imaging. J Nanobiotechnol. 2024;22(1):528. doi: 10.1186/s12951-024-02806-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Luo Z, Zheng Y, Sun R, et al. Photoacoustic imaging: an emerging tool for intraoperative margin assessment in breast-conserving surgery. Photoacoustics. 2025;47:100788. doi: 10.1016/j.pacs.2025.100788 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Song J, Wang H, Meng X, Li W, Qi J. A hypoxia-activated and microenvironment-remodeling nanoplatform for multifunctional imaging and potentiated immunotherapy of cancer. Nat Commun. 2024;15(1):10395. doi: 10.1038/s41467-024-53906-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Li Y, Liu W, Li Y, Peng T, Wang J, Zou T. Preparation and antitumor of hyaluronic acid-human serum albumin-indocyanine green-paclitaxel nanoparticles. Int J Biol Macromol. 2025;318(Pt 3):145131. doi: 10.1016/j.ijbiomac.2025.145131 [DOI] [PubMed] [Google Scholar]
- 19.Zhong C, Chen J, Ling Y, et al. Indocyanine green-loaded nanobubbles targeting carbonic anhydrase ix for multimodal imaging of renal cell carcinoma. Int J Nanomed. 2023;18:2757–2776. doi: 10.2147/IJN.S408977 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Liu J, Chen Y, Wang G, et al. Ultrasound molecular imaging of acute cardiac transplantation rejection using nanobubbles targeted to T lymphocytes. Biomaterials. 2018;162:200–207. doi: 10.1016/j.biomaterials.2018.02.017 [DOI] [PubMed] [Google Scholar]
- 21.Liu F, Mao Y, Yan J, et al. Bionic microbubble neutrophil composite for inflammation-responsive atherosclerotic vulnerable plaque pluripotent intervention. Research. 2022;2022:9830627. doi: 10.34133/2022/9830627 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Zhang X, Gao J, Tang Y, et al. Bioorthogonally activatable cyanine dye with torsion-induced disaggregation for in vivo tumor imaging. Nat Commun. 2022;13(1):3513. doi: 10.1038/s41467-022-31136-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Yu B, Pang Z, Zhao J, et al. Ischemic myocardium targeting peptide-guided nanobubbles for multimodal imaging and treatment of coronary microvascular dysfunction. Adv Healthc Mater. 2025;14(13):e2404477. doi: 10.1002/adhm.202404477 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data is available from the corresponding author.