Ultrasound Molecular Imaging of Blood Vessel Walls and Vulnerable Plaques via CXCR4-Targeted Nanoscale GVs

Introduction

Atherosclerosis Vulnerable Plaque (ASVP) is an atherosclerotic plaque with a thin fibrous cap, large necrotic core rich in lipids, internal bleeding, neovascularization, and active inflammation.¹ The insidious and unpredictable rupture of the ASVP can lead to secondary thrombosis, acute cardiovascular and cerebrovascular events, and even death.^2–4 Consequently, the prompt and precise identification of ASVP proves of utmost significance in pinpointing high - risk patients, alleviating disease - related risks, and effectually governing the progression of the disease.

Ultrasound molecular imaging (UMI) uses specific molecular targets to track the biological processes of lesions and visually display the internal information of the ASVP.^5,6 Moreover, the noninvasive nature and convenience of ultrasound technology suggest its potential as a valuable diagnostic tool.⁷ SonoVue, a microbubbles contrast agent commonly used clinically in combination with ultrasound technology to enhance imaging resolution and sensitivity, is typically between 2–5 μm in diameter, which makes them too large to extravasate and accumulate in the perivascular space.^8–10 ASVP primarily conducts an analysis of surface ulcers by virtue of the filling defect phenomenon of the contrast agent, and the contrast signal is introduced into neovascularization for the purpose of diagnosis.¹¹ In the small plaque period, due to the overflow of angiographic signals, blood pool imaging could not identify vascular wall lesions or provide internal information of lesions, resulting in false-negative results and missed diagnoses.¹² In the previous work, our research group has successfully prepared new nanoscale biological bubbles GVs from Halo, which can generate stable contrast-enhanced ultrasound signals in the blood vessel wall, and by surface modification of GVs, a targeted molecular probe can be constructed for diagnosing ASVP.^13–17 In the pathological process, 99.5% of neovascularization in the plaque originates from the outer membrane vasa vasorum.^18,19 As the disease progresses, an increasing amount of neovascularization breaks through the media and enters the intima, providing a channel for the entry of inflammation and making the plaque vulnerable.^20–22 Targeted nanobubbles constructed by selecting CXCR4 related to ASVP inflammation recruitment as the target are expected to enter the lesion site through neovascularization in the plaque, provide plaque information, and simultaneously display the ASVP ulcer surface, realizing multi-angle assessment of ASVP and improving diagnostic efficiency.^23–26

⁶⁸Ga-pentixafor, a specific CXCR4 ligand designed for positron emission tomography (PET), can precisely recognize and bind to the CXCR4 receptor, enabling the identification of macrophage infiltration in atherosclerotic plaques at the molecular level and has been widely used in clinical settings.^27–30

Materials and Methods

Cell Culture, Cell Model and ASVP Model

Mouse macrophage cells (RAW264.7), human umbilical vein endothelial cells (HUVEC), and mouse aortic vascular smooth muscle cells (MOVAS) were purchased from the American Type Culture Collection and cultured in DMEM high-glucose medium supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (PS) at 37 °C and 5% CO₂. Cells were seeded in 6-well plates. To induce foam cells, HUVECs and MOVAS were given 80 μg/mL Oxidized Low-density Lipoprotein (ox-LDL) (Shenzhen Chuangxin Science and Technology Ltd., Shenzhen, China) and cultured for 72 h, RAW264.7 were given 80μg/mL ox-LDL and cultured for 48 h. Animal experiments were approved by the Ethics Committee of Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences (SIAT-IACUC-211224-HCS-YF-A2101), and were performed in strict accordance with the national standards of the People’s Republic of China (Laboratory animal: Guideline for ethical review of animal welfare [GB/T 35892–2018]. Male Sprague-Dawley rat (SD rat) (8 weeks old, body weight 200 ± 20 g) were supplied by Zhanyan Biotechnology Co., Ltd., Shanghai. The rats were kept in an SPF room on a 12/12 hour light–dark cycle with freely available food and water. An accelerated atherosclerotic rat model was induced by a combination of a high-fat, high-cholesterol diet (cholesterol 1%, fat 10%, yolk powder 3%) and 2F Fogarty balloon catheter endothelial injury. Atherosclerosis was initiated after one week of a high-fat, high-cholesterol diet. Anesthesia was maintained with isoflurane inhalation. The left carotid arteries were injured with a 2.0 Fogarty balloon catheter, as described earlier. Briefly, a balloon catheter was gently advanced through the external carotid artery into the left common carotid artery. The balloon was gently inflated to 2 atm and then retracted. This procedure was repeated thrice for each rat. The balloon catheter was removed, the incision closed with sutures, and the rats were allowed to recover. Plaques in these rats were observed and monitored using weekly ultrasound.

Preparation of GVs, PEG-GVs, Con-GVs, and CXCR4-GVs

Halo bacteria were cultured in ACTT culture medium (250 g NaCl, 20 g MgSO_4, 3 g Trisodium citrate, 2 g KCl, 3 g Yeast extract, 5 g Tryptone, 1 L purified water) at 37 °C continuously for 7–9 days at 220 rpm/min, these cultured bacteria were transferred to the separatory funnel for 1–2 weeks to collect floated bacteria. Floating bacteria were lysed using TMC lysate buffer (10 mm Tris-HCl, 2.5 mm MgCl2, 2 mm CaCl2, pH 7.5), and GVs were centrifuged at 300×g at 4 ◦C. The isolated GVs were washed with phosphate buffered saline (PBS), purified three times by centrifugation at 250 g for 3 h, and finally stored in PBS at 4 °C. GV concentration was estimated using a full-wavelength microplate reader (Scientific Multiskan GO Thermo Fisher, Waltham, MA, USA) at an OD₅₀₀.

In the presence of 1-ethyl-(3—dimethyl aminopropyl) carbamide (EDC) and N-hydroxysuccinimide (NHS), PEG was coupled to the GVs surface through an amidation reaction. EDC (4 mg) and NHS (6 mg) were dissolved in GVs solution and activated at room temperature for 2–4 h. Subsequently, the activated GVs were slowly added to a methoxypolyethylene glycol amine (PEG-amine, molecular weight = 5 kDa) (523 mg) solution and incubated at 4 °C overnight to ensure full combination of PEG-amine and GVs. The obtained solution was centrifuged and rinsed with PBS 3–4 times to remove excess EDC, NHS, and PEG-amine.

The GVs were coupled to alpha-maleimide-omega-N-hydroxysuccinimide ester polyethylene glycol (Mal-PEG-NHS, molecular weight = 2 kDa) through an amidation reaction, and the targeted peptides (2-Nal-Gly-Tyr-N-me-Orn(Cys)-Arg³¹) or the control peptide sequence, 2-Nal-N-me-Gly-Arg-Orn(Cys)-Tyr, were linked to Mal-PEG-GVs through a thioether reaction. Briefly, Mal-PEG-NHS (40 mg) was dissolved in 1 mL GVs (OD₅₀₀:3.5) and incubated at 4 °C for 2–4 h, then centrifuged and rinsed with PBS 3–4 times to remove unbound Mal-PEG-NHS. Next, a PBS solution containing 1 mg of the targeted peptides was added to the Mal-PEG-GVs and incubated at 4 °C overnight. The reaction mixture was centrifuged and rinsed with PBS 3–4 times. FITC-labeled CXCR4-GVs were obtained by adding AF488-NHS (Beyotime Biotechnology, Shanghai, China). Similar rinse steps were used to remove free AF488.

Flow Cytometry and Immunofluorescence Staining

For flow cytometry analysis, 1×10⁶ in vitro cultured ox-LDL-induced RAW264.7 cells were directly stained with anti-CXCR4 antibody for 1 h at 4 ◦C and Alexa Fluor 488-conjugated secondary antibody for 45 min at room temperature, followed by analysis with flow cytometry (BD FACSAria™III, BD, Franklin Lakes, USA). The RAW264.7 cells were not induced by ox-LDL using the same antibody, and a secondary antibody was used as the control. HUVEC and MOVAS were treated in the same manner.

Plaque and normal blood vessel tissues from rats were collected after euthanasia according to the IACUC and embedded in OCT. Frozen sections were made a microtome cryostat (Leica CM1950, Germany) (thickness =8µm). The tissue sections were blocked with 3% BSA for 30 min and stained with an anti-CXCR4 antibody overnight at 4 °C. The sections were then incubated with an Alexa Fluor 488-conjugated secondary antibody (Beyotime Biotechnology, Shanghai, China) for 1 h at room temperature. 4’6-diamidino-2-phenylindole (DAPI) was used for staining nuclei. Fluorescence images were acquired using an inverted microscope (IX73, Olympus).

The Characterization and in vitro Imaging of CXCR4-GVs

The GVs solution was diluted, placed on a copper net, negatively stained with 2% phosphotungstic acid, and dried at room temperature. GVs morphology were examined using transmission electron microscopy (TEM) (Hitachi H-7650, Hitachi Limited, Tokyo, Japan). The absorbance peaks of CXCR4-GVs and Con-GVs were detected using a Fourier infrared spectrometer (FT-IR, Bruker, Germany). The particle size and potential of GVs, PEG-GVs, Mal-PEG-GVs, and CXCR4-GVs were measured using a Zetasizer analyzer (Zetasizer Nano S90, Malvern, Worcestershire, UK).

Targeting Ability Testing of CXCR4-GVs

RAW264.7 cells were cultured overnight in a 24-well plate (5×10⁴ cells/well), and RAW264.7-derived foam cells were induced using the above method after the cells were attached to the wall. After washing three times with PBS, FITC-labeled CXCR4-GVs or Con-GVs were added to the walls, and the cells were incubated darkly for 5–6 min. Free CXCR4-GVs or Con-GVs were rinsed with PBS 3–5 times, after which the cells were fixed with 4% paraformaldehyde for 15 min and stained with DAPI. The adhesion of CXCR4-GVs or Con-GVs to the cells was observed by random field selection under an inverted microscope. Simultaneously, a competitive binding inhibition experimental group was established (the cells were pre-incubated with excess free targeting peptides at room temperature for 30 min to occupy the CXCR4 receptor on the cell surface, and FITC-labeled CXCR4-GVS was added to observe cell adhesion to the cells).

Vascular Wall Imaging Capabilities of GVs

Healthy rats were randomly injected with SonoVue (300 μL, 45 µg/mL), GVs (300 μL, OD ₅₀₀ = 3.5), or PEG-GVs (300 μL, OD ₅₀₀ = 3.5) via the tail vein. Ultrasound imaging was performed using a 3.0–11.0 MHz line array transducer equipped with ultrasound diagnostic equipment (Mindray Resona 7, Mindray, Shenzhen, China). After this number completely disappeared, the next injection was performed at intervals of more than half an hour, and all parameters (acoustic power, 5.13%; mechanical index, 0.167; contrast gain, 65 dB) remained unchanged during imaging.

In vivo Ultrasound Molecular Imaging of CXCR4-GVs

During imaging, rats (n = 6) were anesthetized with 2% isoflurane and warmed using a heating pad. Model rats were randomly injected with Con-GVs (300 μL OD ₅₀₀ = 3.5) or CXCR4-GVs (300 μL OD ₅₀₀ = 3.5) via the tail vein. Ultrasound imaging was performed using a 3.0–11.0 MHz line array transducer equipped with a Mindray Resona 7. After the signals completely disappeared, the next injection was performed at 30 min intervals, and all parameters (acoustic power, 5.13%; mechanical index, 0.167; contrast gain, 65 dB) remained unchanged during imaging.

Histological Examination

To confirm that GVs were able to pass through the endothelial gaps of plaques, ASVP rats (n = 6) were injected with 300 μL FITC-labeled Con-GVs or FITC-labeled CXCR4-GVs via the caudal vein, and plaques were collected after 10 min and used as frozen sections. Tumor sections were fixed with 4% PFA and stained with DAPI for nuclei and anti-CD31 antibody (Abcam, Cambridge, UK) for plaque vessels. Cy5-conjugated goat anti-rabbit secondary antibody (Abcam, Cambridge, UK) was used to visualize vessels.

Biosafety Testing of CXCR4-GVs

The Cell Counting Kit-8 (CCK-8) (Beyotime Biotechnology, Shanghai, China) was used to determine cell viability. Briefly, RAW264.7 cells were cultured in 96-well cell plates (5 × 10⁴ cells per well) and exposed to 100 μL GVs, PEG-GVs, or CXCR4-GVs (OD ₅₀₀ =2.0, 2.5, 3.0, or 3.5). CCK-8 solution was then added to the plates and incubated for 12 h. The absorbance was measured at a wavelength of 450 nm.

Twelve rats were systemically administered PBS, GVs, PEG-GVs, or CXCR4-GVs (300 μL, OD ₅₀₀ = 3.5). Blood samples were collected from the ophthalmic arteries after 1 and 7 days to detect liver function markers (alanine aminotransferase [ALT], aspartate aminotransferase [AST]) and kidney function markers (blood urea nitrogen, BUN, creatinine, and CREA). After blood collection, the mice were sacrificed immediately and the main organs, including the heart, liver, spleen, lungs, and kidneys, were acquired and fixed with paraformaldehyde (4%, w/v) for H&E staining.

Statistical Analysis

Quantitative data on the vascular wall imaging capability of the GVs are expressed as the mean ±SD. GraphPad Prism software (version 6.0) was used for plotting graphs and data analysis. Differences between two groups were compared using an independent sample t-test. P<0.05.

Results

Characterization of GVs, PEG-GVs and CXCR4-GVs

TEM examination clearly showed that the GVs were monodisperse, with a rugby-ball-shaped structure (Figure 1A). In contrast to the GVs and Mal-PEG-GVs, CXCR4-GVs and Con-GVs were observed the absorbance peaks at 2327.33 cm⁻¹(Figure 1B). The zeta potential of GVs, PEG-GVs, Mal-PEG-GVs and CXCR4-GVs were −29.10± 2.14 mV, −1.70± 0.14 mV, −20.20± 1.65 mV and −11.83± 1.55, respectively (Figure 1C). The average particle hydrodynamic diameters were 217.45±4.65 nm for GVs, 272.65±8.95 nm for PEG-GVs, 243.50±3.50 nm for Mal-PEG-GVs and 265.25±3.05 nm for CXCR4-GVs, respectively (Figure 1D).

Figure 1 Preparation and characterization of GVs, PEG-GVs, and CXCR4-GVs. (A) TEM images of isolated GVs (left and right). The scale bars are 100 nm (left), 50 nm (right), respectively; (B) infra-red spectrogram of GVs, Mal-PEG-GVs, Con-GVs and CXCR4-GVs; (C) Zeta potential of GVs, PEG-GVs, Mal-PEG-GVs and CXCR4-GVs; (D) Size distribution of GVs, PEG-GVs, Mal-PEG-GVs and CXCR4-GVs; Data of (C and D) represent the mean ± SD from 3 independent experiments.

CXCR4 Expression in Cell and Tissue

The expression of CXCR4 in RAW264.7, HUVEC, and MOVAS-derived foam cells was detected using flow cytometry. The CXCR4-positive rate on the surface of RAW264.7-derived foam cells, HUVEC-derived foam cells, and MOVAS-derived foam cells were 64.9%, 0.14%, and 2.42%, respectively (Figure 2A). CXCR4 expression in plaque and normal carotid tissues was also evaluated. From Figure 2B and C, we can see a strong green fluorescence signal in plaque tissue, whereas almost no green fluorescence signals were observed in normal carotid tissue.

Figure 2 CXCR4 expression in foam cells and plaque tissues. (A) Flow cytometry analysis for the foam cell surface expression level of CXCR4 protein on RAW264.7 cells, MOVAS cells and HUVEC cells. Scale bars are 50 nm; (B) Immunofluorescence Staining in normal vessels and plaque tissues; (C) Quantification analysis of mean fluorescence intensity from (B). *p < 0.05.

Binding of CXCR4-GVs to RAW264.7-Derived Foam Cells

To determine the ability of CXCR4-GVS to bind to CXCR4 receptors on the cell surface, cultured RAW264.7-derived foam cells were incubated with CXCR4-GVS or Con-GVs (OD ₅₀₀ = 2.0). We could see that after 5 min incubation, the binding ability of FITC-labeled CXCR4-GVs to RAW264.7-Derived Foam Cells was significantly higher than that of FITC-labeled Con-GVs (Figure 3A). The ability of CXCR4-GVs to bind to RAW264.7-foam cells decreased because of the pre-blocking of excessively targeted peptides. Quantitative analysis showed that the average fluorescence intensity of Con-GVs, CXCR4-GVS and blocking groups was 29.92 a.u± 1.80 a.u, 99.97 a.u± 21.46 a.u and 38.19 a.u± 4.74 a.u, respectively (Figure 3B).

Figure 3 In vitro binding of CXCR4-GVs to RAW264.7-derived foam cells. (A) Representative fluorescent microscope images of RAW264.7-derived foam cells incubated with FITC-labeled Con-GVs, FITC-labeled CXCR4-GVs and free targeting peptide + FITC-labeled CXCR4-GVs. Green represents FITC-labeled GVs and blue represents cell nuclei stained with DAPI. Scale bar: 50 µm. (B) Quantification of fluorescence intensity from (A). The data in (B) represent the mean ± SD of three independent experiments. **p < 0.01, nsno significance.

In vitro Imaging of CXCR4-GVs

To evaluate the contrast imaging performance of CXCR4-GVs, different concentrations of CXCR4-GVs from OD₅₀₀ 0.5 to 2.0 were imaged in contrast mode. Figure 4A clearly shows that GVs, Con-GVs, and CXCR4-GVs exhibit significantly enhanced contrast signals. The higher the GVs, Con-GVs and CXCR4-GVs concentrations, the stronger the contrast signals of GVs, Con-GVs, and CXCR4-GVs. Quantitative analysis of average signal intensity revealed that GVs, Con-GVs, and CXCR4-GVs had comparable contrast signals at the same concentration (Figure 4B).

Figure 4 In vitro ultrasound contrast imaging of GVs, con-GVs, and CXCR4-GVs. (A) Ultrasound contrast images of GVs, Con-GVs, and CXCR4-GVs at different concentrations (OD500 = 0.5–2.0). (B) Quantification of the mean ultrasound signal intensities from (A). Data represent the mean ± SD of three independent experiments.

Vascular Wall Imaging Capability of GVs

In this study, SonoVue, GVs, and PEG-GVs were injected intravenously into rats at intervals of > 30 min to compare their imaging abilities in blood vessels. Angiography was performed in CEUS mode. As shown in Figure 5A, GVs generated stable angiographic signals on the carotid artery wall of rats, and SonoVue only imaged the carotid blood pool of rats. Notably, PEG-GVs could obtain longer-lasting angiographic signals in the carotid artery walls of rats.

Figure 5 Vascular wall imaging capabilities of GVs. (A) Contrast-enhanced images of SonVue, GVs, and PEG-GVs after intravenous injection into the normal carotid artery. (B) Contrast-enhanced images of SonVue, GVs, and PEG-GVs after intravenous injection into the plaque carotid artery.

SonoVue, GVs, and PEG-GVs were injected into SD rats in a plaque model for in vivo imaging, and it was found (Figure 5B) that SonoVue could be used for vascular lumen imaging. Owing to the overflow of imaging signals, we could not observe any difference in the imaging signals between the plaque in the anterior wall of the carotid artery and the lumen. By contrast, GVs can generate angiographic signals from the blood vessel wall and accurately outline the shape of the plaque located in the anterior wall of the carotid artery. After injection of PEG-GVs, the contrast signal and cycle time in the vascular wall and plaque were stronger than those of the GVs.

In vivo UMI

Next, we evaluated the targeting ability of CXCR4-GVs in plaque rats with plaques. The results (Figure 6A) showed no contrast signals in the intravascular plaque prior to injection of CXCR4-GVs or Con-GVs. As shown in Figure 6B, there was no significant difference in the peak signal intensities between the Con-GVs (190.80± 8.22) and CXCR4-GVs (207.30± 4.43) in the plaque. However, the signal intensity in the plaques of the CXCR4-GVs was significantly higher than that of the Con-GVs after 0.5 min (Figure 6C). The quantitative analysis of signal intensity in the plaque received with CXCR4-GVs achieved 14.91-fold higher than those of Con-GVs at 5 min (69.19 ± 4.64 vs 7.52 ± 10.53, P < 0.001), 11.94-fold higher than those of Con-GVs at 10 min (34.52 ± 14.76 vs 2.89 ± 3.47, P < 0.05), respectively.

Figure 6 In vivo ultrasound molecular imaging of plaques. (A) Non-linear contrast images of Con-GVs and CXCR4-GVs (OD500 = 3.5) at different time points after intravenous injection. (B) Time-intensity curves of Con-GVs and CXCR4-GVs after intravenous injection. (C) Contrast signal intensities of tumors treated with Con-GVs and CXCR4-GVs at 1, 3, 5, and 10 minutes. Data (B and C) represent the mean ± SD of three independent experiments. *p < 0.05, ***p < 0.001.

Distribution of CXCR4-GVs

To further verify the distribution of CXCR4-GVs in rat plaques, FITC-labeled CXCR4-GVs or Con-GVs were injected into rats via the caudal vein and the sections were sacrificed after 10 min. From the Figure 7A, it can be observed that numerous FITC-labeled CXCR4-GVs were not only present in the vascular system, but also in the extravascular space of the plaque, indicating that CXCR4-GVs is leaking from the neovasculature within the plaque. Three fields were randomly selected for quantitative fluorescence intensity analysis (Figure 7B), and the fluorescence signal of the CXCR4-GVs was significantly stronger than that of the Con-GVs (P < 0.001).

Figure 7 Histological analysis. (A) Representative fluorescence images of plaque sections from rats injected with FITC-labeled Con-GVs or FITC-labeled CXCR4-GVs. Green indicates GVs, red indicates anti-CD31 antibodies, and blue indicates cell nuclei stained with DAPI. Scale bar: 50 µm. (B) Mean fluorescence intensity of plaque sections in three random view fields. Data represent the mean ± SD of three independent experiments. ***p < 0.001.

In vitro and in vivo Biosafety of GVs, PEG-GVs and CXCR4-GVs

The results of the in vitro cytotoxicity assay were assessed by incubating different concentrations of GVs, PEG-GVs and CXCR4-GVs with MB49 cells, followed by viability assay using the CCK- 8 kit. Supplementary Figure 1 clearly shows that no obvious cytotoxicity to RAW264.7 cells was found at all tested concentrations (OD500 = 2.0, 2.5, 3.0, and 3.5). H&E staining showed no pathological damage to the heart, liver, spleen, lungs, and kidneys after injection of GVs, PEG-GVs, and CXCR4-GVs compared with the PBS control (Figure 8A–D). Hematologic data of GVs, PEG-GVs, and CXCR4-GVs injected intravenously into healthy C57BL/6 mice on the first day and on the seventh day were also analyzed, revealing that the indices of (alanine aminotransferase (ALT), aspartate aminotransferase (AST), and renal function (alkaline phosphatase, ALP; Blood urea nitrogen, BUN; Serum creatinine, CREA) were still within the normal range (Figure 8E).

Figure 8 Biosafety analysis. (A–D) Liver and kidney function indices of rats treated with the same volume of PBS, PEG-GV, and CXCR4-GVs. The units of ALT, AST, and ALP were U/L, and those of BUN and CREA were μmol/L. (E) Representative H&E sections of the main organs (heart, liver, spleen, lung, and kidney) from mice treated with PBS, GVs, PEG-GVs, or CXCR4-GVs for seven days. Scale bars: 100 µm. Data represent the mean ± SD of three independent experiments.

Discussion

The precise identification of ASVP assumes a pivotal role in determining the progression of the disease and its prognosis. In the present study, we engineered an innovative, targeted molecular probe, namely CXCR4 - GVs, specifically designed for the diagnosis of ASVP. Our results showed that the targeting of CXCR4-GVs can be visualized in contrast-enhanced ultrasound mode and has good sensitivity for identifying ASVP. Successful identification of ASVP may contribute to several aspects. First, because of their small particle size, GVs have inherent characteristics in vascular wall imaging, especially in vascular wall lesions; second, the targeting peptide, 2-Nal-Gly-Tyr-N-me-Orn(Cys)-Arg from 68Ga-pentixafor, designed as a specific CXCR4 ligand for positron emission tomography (PET), is able to precisely recognize and bind to the CXCR4 receptor. As shown in Figure 5, our study substantiated the efficacy of gas-filled GVs for imaging vascular wall lesions by conducting in vivo ultrasound imaging experiments on the carotid artery, confirming that GVs outperformed SonoVue in diagnosing vascular wall lesions. Furthermore, the data indicated that surface modification of nanoparticles with polyethylene glycol (PEG) effectively mitigates Reticuloendothelial System (RES) uptake and prolongs blood circulation time, thereby improving imaging outcomes.³² We further opted for the vascular wall imaging capabilities of GVs by conjugating them with PEG for their application in vascular wall imaging. This modification was intended to mitigate the immunogenicity associated with protein-shelled GVs and to minimize non-specific binding and clearance within the body, thereby improving imaging efficacy at vascular wall lesion sites.

To achieve molecular-targeting function of GVs and extend their circulation time, we fabricated the targeted molecular probe CXCR4-GVs through a two-step modification strategy. The first step was to link MAL-PEG-NHS with GVs, with one end of the molecule reacting with the protein shell of the GVs and the other end retaining the Mal group, providing an active site for subsequent connections. Next, the medial end of MAL-PEG-NHS connects with the thiol group of the modified peptide, forming a “bridge” to transform GVs into CXCR4-GVs with targeting functionality. We compared the imaging of Con-GVs and CXCR4-GVs in vitro and found that the contrast signals of both Con-GVs and CXCR4-GVs increased with increasing concentration of GVs. The contrast signals of Con-GVs and CXCR4-GVs were similar at the same GV concentration. This indicates that the modified GVs still had good imaging capabilities. As shown in Figure 6, in vivo animal experiments demonstrated that CXCR4-GVs produced stronger signals in the plaque tissue. In conclusion, we combined a new contrast agent capable of imaging the vessel wall to achieve early and accurate identification of ASVP, which may shorten the time required for clinical translation.

Neovascularization in ASVP is an important channel for transporting inflammatory mediators, regulating plaque metabolism, and affecting plaque progression owing to its high permeability.³² At the same time, plaque neovascularization promoted the growth of outer membrane microvessels at the site of wall rupture, resulting in the formation of immature vascular networks, and thus making vulnerable plaques fragile and highly permeable.^33,34 The translocation of nanoparticles into the atherosclerotic endothelium relies on microvascular permeability.³⁵ Owing to neovascularization within vulnerable plaques, they also exhibit a tumor-like EPR effect, allowing small nanoparticles to penetrate the vascular wall and accumulate within the vulnerable plaques.^36–39 Phagocytosis within atherosclerotic lesions and the EPR-like effect provide a passive mechanism for CXCR4-GVs to enter plaques, thus enabling the identification of vulnerable plaques. Meanwhile, the surface ligands of CXCR4-GVs provided an active role in binding with plaques, resulting in better targeting imaging of CXCR4-GVs relative to Con-GVs.

In the future, by using this molecular - targeted probe to load drugs, we can achieve the integration of diagnosis and treatment. Additionally, the real - time detection of drug release during the treatment process can be realized, which is expected to provide new strategies for the prevention and treatment of clinical cardiovascular and cerebrovascular events.^40,41

Conclusion

In summary, we used biosynthetic GVs to develop a molecular probe, CXCR4-GVs, with imaging and targeting properties. We discovered a unique vascular wall imaging technique. For the first time, we developed a novel targeted molecular probe, CXCR4-GVs, which provides a new method for early diagnosis of atherosclerotic vulnerable plaques and for future clinical translation.