T1-weighted MRI of targeting atherosclerotic plaque based on CD40 expression on engulfed USPIO’s cell surface

Atherosclerosis is a chronic inflammatory disease characterized by the accumulation of cholesterol within the arterial wall. Its progression can be monitored via magnetic resonance imaging (MRI). Ultrasmall Superparamagnetic Particles of Iron Oxide (USPIO) (<5 nm) have been employed as T1 contrast agents for MRI applications. In this study, we synthesized USPIO with an average surface carboxylation of approximately 5.28 nm and a zeta potential of −47.8 mV. These particles were phagocytosed by mouse aortic endothelial cells (USPIO-MAECs) and endothelial progenitor cells (USPIO-EPCs), suggesting that they can be utilized as potential contrast agent and delivery vehicle for the early detection of atherosclerosis. However, the mechanism by which this contrast agent is delivered to the plaque remains undetermined. Our results demonstrated that with increasing USPIO concentration during 10–100 μg ml−1, consistent change appeared in signal enhancement on T1-weighted MRI. Similarly, T1-weighted MRI of MAECs and EPCs treated with these concentrations exhibited a regular change in signal enhancement. Prussian blue staining of USPIO revealed substantial absorption into MAECs and EPCs after treatment with 50 μg ml−1 USPIO for 24 h. The iron content in USPIO-EPCs was much higher (5 pg Fe/cell) than in USPIO-MAECs (0.8 pg Fe/cell). In order to substantiate our hypothesis that CD40 protein on the cell surface facilitates migration towards inflammatory cells, we utilized AuNPs-PEI (gold nanoparticles-polyethylenimine) carrying siRNACD40 to knockout CD40 expression in MAECs. It has been documented that gold nanoparticle-oligonucleotide complexes could be employed as intracellular gene regulation agents for the control of protein level in cells. Our results confirmed that macrophages are more likely to bind to MAECs treated with AuNPs-PEI-siRNANC (control) for 72 h than to MAECs treated with AuNPs-PEI-siRNACD40 (reduced CD40 expression), thus confirming CD40 targeting at the cellular level. When USPIO-MAECs and MAECs (control) were delivered to mice (high-fat-fed) via tail vein injection respectively, we observed a higher iron accumulation in plaques on blood vessels in high-fat-fed mice treated with USPIO-MAECs. We also demonstrated that USPIO-EPCs, when delivered to high-fat-fed mice via tail vein injection, could indeed label plaques by generating higher T1-weighted MRI signals 72 h post injection compared to controls (PBS, USPIO and EPCs alone). In conclusion, we synthesized a USPIO suitable for T1-weighted MRI. Our results have confirmed separately at the cellular and tissue and in vivo level, that USPIO-MAECs or USPIO-EPCs are more accessible to atherosclerotic plaques in a mouse model. Furthermore, the high expression of CD40 on the cell surface is a key factor for targeting and USPIO-EPCs may have potential therapeutic effects.


Introduction
Atherosclerosis is a prevalent cardiovascular disease.Plaques are formed as a large amount of cholesterol accumulate on the surface of the blood vessel endothelium, leading to arterial stenosis [1].
Endothelial damage caused by atherosclerosis attracts lipids in the blood and invite them to accumulate at the injured site, leading to aggregation of platelets and monocytes.The balance between endothelial injury and repair is essential to maintain endothelial integrity and function to prevent thrombotic complications [2].Endothelial cells are a single layer of cells lining the inner surface of arteries that play a crucial role in the formation of atherosclerosis and changes in intimal integrity and permeability [3].Endothelial progenitor cells (EPCs) possess homing capabilities, which could be recruited to the plaque site to repair damaged endothelium [4].Our previous studies affirm the ability of endothelial cells to target atherosclerotic plaques [5] while the mechanism underlying plaques-targeting by endothelial cells and EPCs remain obscure, thus we speculate the high CD40 expression on the cell surface serve as a crucial role.
Molecular imaging technologies have been utilized to monitor the development of atherosclerosis.One such tool is magnetic resonance imaging (MRI), which has been used as mainstream clinical diagnostic technique due to its excellent capacity of distinguishing soft tissues [6].To ensure accuracy, MRI requires contrast agents such as gadolinium (GD) agents and ultra-small superparamagnetic iron oxide particles (USPIO).The long halflife of USPIO in blood qualifies them as an MRI contrast agent [7].Compared to conventional GDbased contrast agents, USPIO has low toxicity and im-proved biocompatibility, and an ability to significantly reduce transverse relaxation time in tissue [8].Studies have shown that USPIO is readily phagocytosed by macrophages in organs or tissues with high inflammatory cell infiltration, such as atherosclerotic plaque [9] and myocardial infarction [10].Consequently, some researchers utilized USPIO to integrate into the lesion through blood circulation and subsequently employ MRI to detect the lesion [11].USPIO (<5 nm) have been used as T1 contrast agents for MR imaging applications [12].We previously used this USPIO with peptide linkage to achieve targeting inflammation sites with significant difference in T1-weighted MRI [13], which raise this question-could phagocytosed USPIO's EPCs (USPIO-EPCs) and mouse aortic endothelial cells (USPIO-MAECs) perform T1-weighted MRI?This query warrants worth further research.
Expression of CD40 and CD40L promotes the accumulation of atherosclerotic plaque [14,15].CD40, a key target for immunotherapy for over 20 years [16], is implicated in atherosclerotic plaque, an immune disease [17].Therefore, we speculate that blood endothelial cells may be enriched in atherosclerotic plaques through the interaction of the high CD40 expression on the cell surface and the high CD40L expression on the plaque.The study aims to discern whether USPIO-EPCs or USPIO-MAECs can diagnose ath-erosclerosis clinically, and provide preliminary evidence for future treatment of atherosclerosis with EPCs.In this study, we confirm that high CD40 expression on the cell surface serves as a key factor enabling MAECs and EPCs to target atherosclerotic plaques.

Ethical approval
All animal procedures were performed in accordance with the Guidelines for Care and Use of Laboratory Animals of South China Normal University and approved by the Animal Ethics Committee of South China Normal University.

Synthesis of USPIO and AuNPs-PEI (gold nanoparticles-polyethylenimine), Gel retardation assay
Synthesis of USPIO can be seen in our previous experimental methods [13].USPIO is synthesized by the following steps: 8 mM anhydrous FeCl 3 was dissolved in 40 ml of diglycol.3.2 mM sodium citrate was added to the above solution in 80 • C water bath, then 24 mM anhydrous sodium acetate was added at 40 • C. Then sodium acetate was added to the above mixture and mixed to dissolve.In the next step, the dissolved mixture was transferred to a 100 ml capacity Teflon stainless steel autoclave.The autoclave was placed in an oven at 200 • C, and the mixture was allowed to fully react for 4 h.After the autoclave is cooled to room temperature, the reactant is taken out.The reactant is added to absolute ethanol and centrifuged at 13 200 × g for 15 min, and the black precipitate after centrifugation is collected.Repeat the centrifugation process three times to remove excess reactants and byproducts.The resulting black product was dissolved in water and lyophilized to obtain USPIO powder for the next step.USPIO shape with transmission electron microscopy (TEM) observation, zeta potential and dynamic light scattering meas-urements, specific method please refer to previous articles published by our laboratory [13].
AuNPs-PEI synthesis method could be referred to previous articles published by our laboratory [18].In short, 1 ml HAuCl 4 (Aladdin, Shanghai, China) solution (2 mM) with deionized water and 14 ml PEI (branched polyethylenimine, 25 kD, Shyuanye, Shanghai, China) solution (0.2 mM) were prepared respectively.Two solutions were slowly mixed and heated to 85 • C until become ruby red, then stir it at room temperature, that is AuNPs-PEI solution.Centrifuged at 7040 × g for 15 min to remove the free PEI for three times and resuspended the pellet in 200 µl deionized water, the concentration of AuNPs-PEI solution is about 1 mM.

Establishment of atherosclerosis model and evaluation
APOE -/-(T001458) mice were purchased from Nanjing Biomedical Research Institute, Nanjing University, and were kept at a stable temperature and humidity during a 12 hour dark/light cycle.Disinfect the water and the cage, and change the cage every week.For consistency, we used male mice in all experiments.APOE -/-mice induced by high fat diet (HFD) are widely recognized and used in the study of athero-sclerosis [19].8 week-old APOE -/-mice were fed with HFD (Shoobree, Nanjing, Jiangsu, China) for 20 weeks (APOE -/-mouse-HFD), which include 1% cholesterol, 15% lard, 0.25% sodium cholate, 5% yolk powder, 2.5% sugar and basal diet, were fed with basal diet for 20 weeks as the control (APOE -/-mouse-ND).
Oil red O (ORO) Staining.Distributions of vascular plaques were determined by ORO staining reagent (Sigma-Aldrich, St. Louis, MO, USA) and observed by the Stereo Microscope (OlympusSZ51, Tokyo, Japan).ORO staining reagent was dissolved in distilled water in a ratio of 3:2 and filtered with a slow filter.The aortic arch to the root of the abdominal blood vessels is separated, and the vessels were longitudinally dissected.The samples were stained by ORO solution for 30 min, destained with 80% isopropanol for 30 s, and finally washed with a phosphate buffer for 5 min.
Histological Examination.Hematoxylin eosin staining (H&E staining) is used to detect the presence of plaque in atherosclerosis model mice after 4 months of HFD.The deeply anesthetized APOE -/- mice were dissected, blood vessels were carefully removed, and they were immersed in formalin for 24 h.It was dried with gradient alcohol, em-bedded in a paraffin block, and cut into 4 µm sections with a cryostat.The sections were continuously collected on gelatin-coated glass slides and stained with H&E.Then, observe the H&E-stained sections under a microscope and take pictures [20].

Phagocytosis of USPIO by MAECs and EPCs, prussian blue staining
In order to confirm the cellular uptake of USPIO, 5 × 10 5 MAECs and 5 × 10 5 EPCs were incubated with different concentrations of USPIO for 24 h, without FBS.After that, the adherent cells were washed 3 times with PBS and incubated with 4% paraformaldehyde for 20 min to fix their shape and stained with Prussian blue kit (Beijing Solarbio, G1422) for Prussian blue staining.

Cell viability
Cell Counting Kit-8 (CCK-8, Dojindo Laboratories, Kumamoto, Japan) was used to evaluate cells' viability.1 × 10 4 cells well −1 were plated in the 96-well plates (Corning Incorporated, NY, 128 USA).After incubation with different concentration of USPIO for 24 h and removal of the supernatant, each well was washed for three times, added by 100 µl fresh DMEM without FBS and 10 µl reagents and further incubated for 0.5-4 h.The viability of cells was estimated by measurement of absorbance at 450 nm (A 450 ) that was read with a microplate reader (INFINITE M200, Tecan, Switzerland) [21].

Determination of iron content
MAECs and EPCs were seeded in a six-well plate at a density of 1 × 10 6 , and treated with 50 µg ml −1 USPIO for 24 h.Then wash with PBS 3 times to wash away the US-PIO that has not been phagocytosed by the cells, then digest the cells with trypsin solution, and collect the cells by centrifugation at 3000 rpm (633 × g) for 5 min.Resuspend the cells in 1 ml PBS, and then digest the cells with aqua regia.Then the digested solution was made up to 10 ml with deionized water, and then detected by ICP-OES (Inductively Coupled Plasma Optical Emission Spectrometer, Thermo E. IRIS Duo) [12] MAECs and EPCs not treated with USPIO were used as controls.The iron content is measured by ICP-OES, for the conversion of the relationship between iron content and USPIO, please refer to our previously published article [13].

Western blotting
For Western blotting (WB) analysis, cells samples are collected and put on ice with a lysis buffer containing a complete protease inhibitor cocktail (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1% TritonX-100, 100 µg ml −1 PMSF) lysis for 60 min, then collect the lysate and centrifuge at 12 000 rpm (10 142 × g) for 15 min.Then the supernatant was taken, and the protein concentration in the supernatant was determined by the Bio-Rad protein assay.The protein samples were denatured and Loaded on a 12.5% sodium-dodecyl sulfate polyacrylamide gel, and then transferred to PVDF Membrane (CAT.NO.PVH00010 PORE SIZE: 0.45 µm, Merck Millipore Ltd.).Use Tris Buffered Saline with Tween-20 (TBST) solution containing 5% skim milk (150 mmol L −1 NaCl, 10 mmol L −1 Tris-HC [pH 7.4], 0.1% Tween 20) to block the membrane for 2 h, and then incubate with the specific primary antibody Anti-CD40 Polyclonal Antibody (Servicebio) and secondary antibody Goat Anti-Rabbit IgG (MultiSciences, Hangzhou, China); Anti-GAPDH (GAPDH was used as loading control, Absin, Shanghai, China) and secondary anti-body Goat Anti-Mouse IgG (Dingguo, Beijing, China).Signals were detected using an image analyzer (Tanon-5200, shanghai, China).The experiment was repeated three times and results were analyzed using ImageJ software [24].

Cell and tissue targeting experiments
Cell targeting experiment.For the cell adhesion experiment, macrophages were stained with 10 µM dioctadecyloxacarbocyanine (DiO) dye (Beyotime C1038) in living cells for 30 min [25].After staining, the excess dye was washed with PBS, and then macrophages and MAECs (AuNPs-PEI-siRNA NC and AuNPs-PEI-siRNA CD40 were treated with MAECs respectively for 72 h) were co-cultured for 30 min.After 30 min the macrophages not bound to MAECs were gently washed with PBS, and then observed under a confocal fluorescence microscope (Zeiss LSM 880).DiO (Cell Membrane Green Fluorescent Probe), ex-citation of 488 nm wavelength, emission of 540 nm wavelength, detection of 493-586 nm wavelengths.
Tissue targeting experiment.Using high-fat-fed APOE -/-mice as material, the mice were injected via the tail vein with MAECs that phagocytosed USPIO (MAECs were injected as control).The mice were euthanized 24 h later, and the tissue aortic arch was removed.The iron content of vascular plaques was determined by ICP-OES method.

Blood lipids Examination
Normally fed APOE -/-mice and high-fat-fed APOE -/- mice were deeply anesthetized with 1% sodium pentobarbital, the eyeballs were removed, and the mice's whole-body blood was obtained from the eye sockets.After standing at room temperature for 30 min, centrifuge at 650 g for 10 min at 4 • C. Take the blood supernatant and send it to the South China Normal University School Hospital for testing indicators: glucose, triglycerides (TG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C) [26].

USPIO and Cells phagocytized USPIO's MRI, in vivo MRI
Different concentrations of USPIO were placed in 1.5 ml Eppendorf tubes at different USPIO concentration levels (10, 25, 50, 100 µg ml −1 ).After incubated with 10, 25, 50, 100 µg ml −1 USPIO for 24 h, the cells were collected and washed with 1 × PBS for three times.1 × 10 6 cells of each sample were fixed for MRI.MRI were obtained with a 3.0 T Achieva (Philips Healthcare, Best, The Netherlands) with the parameters: T1 weighted imaging (T1WI): field of view (FOV) = 180 × 161 mm, matrix = 376 × 269, slice thick-ness = 3.5 mm, time of repetition (TR) = 582 ms, and time of echo (TE) = 20 ms.T1 signal intensity measurement was performed in T1WI, expressed as signal noise ratio (SNR) [5], SNR = SI tissue/ SD background, SI tissue represents the average of the tissue signal in-tensity in the area of interest.The SD background represents the standard deviation (S.D) of the background signal of the same area, and the area without tissue structure at the same level as the SI tissue in the phase encoding direction is often selected.For image analysis, all images were analyzed via MRI viewing station.
The mice were imaged under general anesthesia to assess the distributions of the USPIO-labeled cells.MRI were obtained using 9.4 T superconducting magnet system T1 weighted sequence imaging (BioSpec 94/30 USR 9.4 T, Bruker, Germany).The specific parameters of the MRI scan: T1 WI: spin echo sequence is used for scanning, repetition time [TR]/echo time [TE] 400/12.5 ms, FOV = 3 cm × 3 cm, matrix = 196 × 196, layer thickness 0.7 mm, layer spacing 0.3 mm.The number of scanning layers is 15, scanning range: mouse head and neck (aortic archbrain).The scanning position is the prone position with the head advanced.For image analysis, all images were analyzed via MRI viewing station.

Data analysis
Statistical analysis was performed using GraphPad Prism 7.0 software.All experiments were repeated three times and expressed as mean standard error of the mean.Other comparisons between two or more groups were made using student unpaired t-test, twoway analysis of variance, and one-way analysis of variance.P < 0.05 is considered a statistically significant difference, p < 0.01 extreme significant difference, the smaller the p-value (the more asterisks), the more significant the difference is.The bars in the graph represent the S.D.

Synthesis, characterization and cytotoxicity assessment of USPIO
The USPIO were synthesized in alignment with our previous experimental methods [13].The explicit synthesis method is comprehensively described in the Material and Methods section 2.2.The synthesized USPIO were then characterized by means of zeta potential and TEM.The average zeta potential of the USPIO was 47.8 mV (figure 1(a)), indicative of an excellent dispersion.The TEM analysis revealed that the statistical average diameter of the particles was approximately 5.28 nm (figures 1(b) and (c)), with the hydrodynamic sizes of USPIO primarily falling within the range of 1.5 nm-6.5 nm (figure 1(d)).The CCK-8 experiment results demonstrated that the survival rate of the MAECs was approximately 100% following treatment with 50 µg ml −1 USPIO for 24 h, while the cell survival rate fell below 80% upon treatment with 100 µg ml −1 USPIO (figure 1(e)).When EPCs were treated with 100 µg ml −1 USPIO for 24 h, a significant inhibition of EPCs proliferation was observed.However, when the USPIO concentration was less than or equal to 50 µg ml −1 , USPIO notably promoted the proliferation of EPCs (figure 1(f)).These findings are in accordance with the cytotoxicity prerequisites proposed in previous reports [27][28][29], indicating that 50 µg ml −1 USPIO does not exhibit significant cytotoxicity to MAECs and EPCs.Consequently, we use 50 µg ml −1 as the operative concentration for incubating USPIO with MAECs (henceforth referred to as USPIO-MAECs) and EPCs (henceforth referred to as USPIO-EPCs).

Comparative analysis of USPIO phagocytosis in MAECs and EPCs
To examine the phagocytosis of MAECs towards USPIO, we employed Prussian blue staining to ascertain the iron content of MAECs post incubation with 50 µg ml −1 USPIO for 12 h, 16 h and 24 h.Initial observations revealed no significant iron accumulation in MAECs for 12 h, while a significant iron accumulation in MAECs treated for 16 h could be detected.As incubation time stretched to 24 h, this accumulation in MAECs manifested more evidently (figure 2(a)).
Furthermore, we utilized an inductively coupled plasma optical mission spectrometer (ICP-OES) to quantify the specific iron content in 10 6 MAECs incubated with USPIO for 24 h (figure 2(b)).Post a 24 h treatment with 50 µg ml −1 USPIO, 10 6 MAECs were collected and digested thus USPIO not  engulfed by MAECs were rinsed off with PBS.The results revealed that the final iron content settled around 0.8 pg cell −1 .In contrast, the iron content in MAECs without USPIO treatment measured only 0.069 pg cell −1 , which were far lower than that in USPIO-MAECs.
We probed the impact of LPS treatment on USPIO-MAECs, fuelled by prior reports that CD40 expression in macrophages exhibited an increase treated with LPS or TNF-α [30].Unexpectedly, results from Prussian blue staining and ICP-OES illustrated that LPS treatment could significantly enhance the uptake of USPIO by MAECs (appendixs (a) and (b)).This indicates that LPS treatment significantly augments the phagocytic capacity of MAECs towards USPIO.Similarly, we employed Prussian blue staining to investigate the phagocytosis of USPIO by EPCs.The results showed that no significant accumulation of iron was detected in the EPCs post 12 h treatment with USPIO.However, large amounts of iron were detected in EPCs following a 24 h treatment with USPIO (figure 2(c)).To delve deeper into the uptake of USPIO by EPCs, we quantified the uptake of USPIO by 10 6 EPCs following a 24 hour incubation period with USPIO.The results showed that the after treated with 50 µg ml −1 USPIO for 24 h, unabsorbed USPIO were washed away.ICP-OES results illustrated that the iron content was approximately 5 pg cell −1 , indicating a higher absorption of USPIO by EPCs, while the iron content of EPCs without USPIO treatment was only 0.1 pg cell −1 (figure 2(d)).
Prior research suggests a positive correlation between the expression of tumor necrosis factorα (TNF-α) and CD40 in dendritic cells [31].Accordingly, our ideal method is to treat cells with TNF-α to secure high CD40 expression, anticipating a subsequent elevation in USPIO phagocytosed by cells.As evidenced by Prussian blue staining and ICP-OES, TNF-α treatment amplified the uptake of USPIO by EPCs (appendixs (c) and (d)).
Treatments with LPS and TNF-α enhanced the capacity of cells to phagocytize USPIO, indicating that in an environment rich in LPS and TNF-α, MAECs and EPCs could accelerate the absorption of USPIO.Surprisingly, comparative analysis of USPIO uptake by MAECs and EPCs revealed that presence of USPIO was significantly heightened in USPIO-EPCs (5 pg cell −1 ) than in USPIO-MAECs (0.8 pg cell −1 ) (figures 2(b) and (d)).This prompts the question: Do regular changes occur in T1-weighted MRI with varying concentrations USPIO, or the amount of phagocytosed USPIO in MAECs and EPCs fluctuate under different USPIO concentration treatments?

T1-weighted MRI of USPIO and phagocytosed USPIO-incorporating cells
USPIO's excellent MRI performance is very important with its super-paramagnetic [32], USPIO showing superparamagnetic property both shorten the T1 and T2 relaxation times that is on the basis of MRI signal enhancement.USPIO are generally used as T2weighted (negative) contrast agents since they more effectively shorten T2 rather than T1.When USPIO diameter is <4-5 nm, USPIO is considered to be the next generation of magnetic resonance contrast agent [33].The smaller the particle size of USPIO nanoparticles, the more effective it is in shortening the T1 time, and the more adapted it is to enhanced T1weighted MRI [34].Our findings also showed that USPIO performed good T1-weighted MRI ability by correlating stronger T1 signal with increasing USPIO concentration (10-100 µg ml −1 ) (figures 3(a) and (d) line 1).
To ascertain if cells harbouring phagocytosed USPIO still remain superparamagnetic, we assessed the T1-WI effects (figure 3  (figures 3(b) and (c)).These findings suggested that both EPCs and MAECs of phagocytosed USPIO still maintain excellent T1-weighted MRI performance.This observation revealed that phagocytosed USPIO's EPCs could serve as an optimal contrast agent, potentially suitable for the integration of T1-weighted MRI tracing and therapy.

Expression of CD40 on the surface of MAECs and EPCs
Prior research has shown that CD40 expressed by MAECs facilitates the accumulation of inflammatory cells within plaques while promoting monocytes and endothelial cells adhesion [35].Building on this understanding, we utilized 50 µg ml −1 USPIO for a 24 h treatment of MAECs to generate USPIO-MAECs thus we could test our hypothesis that endothelial cells expressing high levels of CD40 could be targeted to atherosclerotic plaques.Previous research has unveiled that treatment with LPS or TNF-α could significantly upregulate CD40 expression in macrophages [30].Given these findings, we speculate that LPS or TNF-α might function similarly in the context of MAECs or EPCs.Therefore, we subjected MAECs to 1 µg ml −1 LPS for 48 h then extracted RNA and total protein for CD40 detection.Contrary to our expectations, the qPCR (figure 4(a)) and Western Blot (WB) (figures 4(b) and (c)) results suggested that LPS treatment did not induce CD40 expression in MAECs, while slightly diminished the CD40 content on the cell surface (figures 4(a)-(c)).In a parallel experiment, we utilized 10 ng ml −1 TNF-α to treat EPCs for 48 h.Cells collection followed by RNA and total protein extraction revealed that while TNFα treatment significantly increased the CD40 mRNA expression (figure 4(d)) and it did not significantly alter CD40 protein level (figures 4(e) and (f)).

Knockout of CD40 on the surface of MAECs and cell targeting experiment
In order to substantiate our hypothesis that CD40 protein on the cell surface facilitates migration towards inflammatory cells, we utilized AuNPs-PEI carrying siRNA CD40 to knockout CD40 expression in MAECs.It has been documented that gold nanoparticle-oligonucleotide complexes could be employed as intracellular gene regulation agents for the control of protein level in cells [36].We encapsulated siRNA CD40 with AuNPs-PEI to diminish CD40 expression in MAECs.Gel retardation assay was utilized to ascertain the optimal binding ratio of AuNPs-PEI and siRNA CD40 .The optimal volume binding ratio of AuNPs-PEI solution (1 mM) and siRNA CD40 (20 µM) was 4:1 (figure 5(a)).Post-transfection of AuNPs-PEI-siRNA CD40 for 60 h, qPCR result indicated a significant downregulation in CD40 expression (figure 5(b)).WB experiments also showed that CD40 expression were significantly reduced after transfection for 72 h (figures 5(c) and (d)).However, compared to the control (AuNPs-PEI-siRNA NC ), CD40 expression did not decrease but increased in the cells post transfection of AuNPs-PEI-siRNA CD40 for 36 h and 48 h respectively (figures 5(c) and (d)).We speculate that phagocytosis of USPIO by cell may stimulate cell surface CD40 expression and the cells adapt over time before significantly reducing cell surface CD40 expression.Nevertheless, we could reduce the expression of CD40 on the cell surface of MAECs through treatment by AuNPs-PEI-siRNA CD40 for 72 h.
Following this, we utilized AuNPs-PEI-siRNA CD40 to treat MAECs for 72 h and performed cell targeting experiments due to low CD40 expression.We co-incubated 3,3 ′ -DiO perchlorate -labeled macrophages with MAECs treated with AuNPs-PEI-siRNA CD40 (siRNA CD40 -MAECs) for 30 min and MAECs treated with AuNPs-PEI-siRNA NC (siRNA NC -MAECs) served as the control group.Stained macrophages with DiO were assessed to define the difference in adhesion of macrophages to CD40 knockout and non-knockout MAECs.The results decipts a greater number of DiO dyed macrophages recruited near MAECs treated with AuNPs-PEI-siRNA NC (figure 5(e) line1), while few macrophages recruited near MAECs treated with AuNPs-PEI-siRNA CD40 (figure 5(e) line 2).This indicated that knocking out CD40 expression in MAECs reduced the adhesion of macrophages to MAECs.In essence, high expression of CD40 on the surface of MAECs promote macrophages to target MAECs.

Construction of atherosclerosis model mice and vascular plaques targeting experiment
The APOE -/-mice in the control group were fed a normal basic diet (ND), while the APOE -/-mice in the second experimental group were fed a HFD for 20 weeks.Analysis of fresh serum unveiled a significant difference in blood composition between APOE -/-mouse-HFD and APOE -/-mouse-ND (figure 6(a)).The results showed that TC (TG), LDL-C and HDL-C of HFD mice were significantly elevated compared to ND mice, which is a typical feature of atherosclerosis.To further validate the formation of atherosclerosis, the longitudinally dissected aortic arch was stained with ORO and H&E.ORO staining showed obvious lipid droplets in APOE -/-mouse-HFD, while not As shown in figure 4, MAECs and EPCs manifested high expression of CD40.To verify that MAECs and EPCs can target atherosclerotic plaques due to high CD40 expression, we initially performed in vivo experiments utilizing the atherosclerosis model mice [37].To test the potential capacity of USPIO-MAECs to be delivered to plaque sites and thereby demonstrate their targeting, USPIO-MAECs were administered to mice via tail vein injection for 24 h, while MAECs alone were injected as a control.We analyzed the digested and dissolved iron content of aortic tissue and found that USPIO-MAECs treatment resulted in a higher iron content (average iron 1.05 mg per gram aorta) compared to control, corresponding to the average iron content per gram aorta of merely 0.34 mg.This result suggested that USPIO-MAECs indeed could target plaques (figure 6(d)).To further confirm that EPCs may target atherosclerotic plaque due to high CD40 expression, we employed T1-weighted MRI based on our results that USPIO and phagocytosed USPIO's cells have better T1-weighted MRI (figures 3(b) and (d), to verify our hypothesis.

T1 WI of targeted atherosclerotic plaque in APOE -/-mice
We established PBS, USPIO (50 µg mL −1 ), EPCs and USPIO-EPCs (50 µg ml −1 USPIO and each group was treated for 24 h, non-phagocytosed USPIO by EPCs was washed away).Four groups were delivered to atherosclerotic mice via tail vein injections respectively.Post 72 hour injection, T1-weighted MRI values of the atherosclerotic plaque were captured under the 9.4 T MRI (figure 7(a)).According to the statistical results of the T1 WI, we found that the strongest signals were detected in USPIO-EPCs group, indicating that USPIO-EPCs were indeed accumulating in the plaques (figure 7(b)).This finding demonstrates that USPIO-EPCs are capable of targeting plaque sites due to the high CD40 expression on the cell surface of the EPCs.
We also noted that the T1 weighted MRI signal intensity of the USPIO group, while lower than that of the USPIO-EPCs group, was significantly higher than that of the PBS and EPCs groups.We speculate that this could be due to the presence of substantial amounts of inflammatory factors (including LPS and TNF-α) at the atherosclerotic plaque sites.As illustrated in figure 2 and appendix, LPS and TNF-α can stimulate EPCs to phagocytose more USPIO, suggesting that MAECs and EPCs at the plaque site (where there is a large accumulation of EPCs [38] can phagocytose more USPIO when USPIO passes through plaque site via blood flow.

Discussion
We originally hoped to identify a cell exhibiting low CD40 surface expression to serve as a control for the paving our experiment.However, we found that a majority of cells expressed CD40 at high levels.To investigate whether the CD40/CD40L interaction mediates the accumulation of endothelial cells on the plaques, we employed the pristine MAECs (high-expression CD40) and the low-expression CD40-MAECs (AuNPs-PEI-siRNA CD40 treatment) as materials and confirmed that macrophages are more likely to target MAECs with high CD40 expression at the cellular level (figure 5(e)).After USPIO was taken up by MAECs with high expression of CD40, we delivered this into model mice by tail vein injection.Subsequent measurement of the iron content in the aortic tissue revealed a greater accumulation of iron at the plaque site (figure 6(d)), indicating that USPIO-MAECs may demonstrate their targeting capabilities at the tissue level due to their high CD40 surface expression.Then we injected USPIO-PECs into atherosclerotic plaque model mice via the tail vein.USPIO-EPCs with high expression of CD40 yielded a stronger T1-weighted MRI signal at the plaque site (figure 7), which illustrated the targeted characteristics of USPIO-EPCs in vivo.
Several studies have shed light on that new blood vessels occur in the early stages of atherosclerosis [39,40].Moreover, a heightened detection of EPCs in the unstable plaques of patients was demonstrated [38].These findings support that the T1-weighted MRI signal at the plaque site in the USPIO-treated group was higher than in the PBS and EPCs-treated groups owing to that endothelial cells and EPCs at the plaque have the capacity to phagocytose more USPIO, hence forming endothelial cells and EPCs cells that resembled phagocytosed USPIO.T1-weighted MRI signal are the highest in the USPIO-EPCs group, which can be attributed to the high surface CD40 expression on EPCs.We noticed a USPIO accumulation at the plaque site when injected USPIO alone (figure 7).We hypothesized that the cells at the atherosclerotic plaque possess the capacity to naturally phagocytize USPIO, possibly due to inflammatory factors such as TNF-α at the plaque.Our results showed that TNF-α treatment led to a large increase of iron uptake in MAECs and EPCs.Our actual injection of USPIO (1 mg ml −1 ) is 100 µl (the iron content is actually 44.6 µg), and the actual iron content of EPCs phagocytosed USPIO is 5 µg (10 6 EPCs).That is to say, there is an 8.92 times difference in iron content between USPIO (44.6 µg) and USPIO-EPCs (5 µg), however, the T1WI values were significantly higher for the USPIO-EPCs group than for USPIO.This reinforces that USPIO-EPCs possess the remarkable capacity to target atherosclerotic plaques.
Studies have shown that bone marrow-derived EPCs, under the influence of ap-propriate cytokines and/or hormones, can contribute to vascular wound healing [41] and amelioration of limb ischemia [42], repair of myocardial infarction [43] and promotion of vascular endothelialization [44], thereby offering therapeutic effects for vulnerable plaques.Repairment of damaged vascular endothelium and promotion of vascular re-endothelialization have become important topics in the preventive strategies against atherosclerosis.Utilizing tail vein injection targeting of USPIO-EPCs and the leveraging therapeutic effect of EPCs, our cell-targeted therapy holds an advantage in simplicity, which can be obtained without the necessity for complex instrumentation in future applications.

Conclusions
We synthesized a simple USPIO capable for T1weighted MRI; USPIO-MAECs and USPIO-EPCs exhibited a good performance in T1-weighted MRI.Our work demonstrated that high CD40 expression on the surface of MAECs facilitates targeting of macrophages; high CD40 expression on the surface of EPCs may be one of the important reasons for the significantly enhanced T1-weighted MRI signal of USPIO-EPCs treatment; The potential future of USPIO-EPCs utilization for T1-weighted MRI diagnosis and therapeutic effects is promising.

Figure 3 .
Figure 3. MR imaging of USPIO and phagocytosed USPIO's cells.(a) T1-weighted imaging (T1WI) SNR of different concentrations of USPIO for 24 h; (b) T1WI SNR of EPCs treated with different concentrations of USPIO for 24 h; (c) T1WI SNR of MAECs treated with different concentrations of USPIO for 24 h (d) T1-weighted MRI of different concentrations USPIO, EPCs and MAECs treated with different concentrations of USPIO for 24 h (Removal of USPIO not phagocytosed by cells).(at least three individual experiments, mean ± SEM, one-way ANOVA).

Figure 4 .
Figure 4. CD40 expression on the surface of MAECs and EPCs.(a) A representative qPCR experiment used to detect the level of CD40 in MAECs by LPS; (b) A representative western blot test used to detect the level of CD40 in source MAECs by LPS; (c) gray value statistics of figure 2(b); (d) a representative qPCR experiment used to detect the level of CD40 in EPCs by TNF-α; (e) a representative western blot test used to detect the level of CD40 in source EPCs by TNF-α; (f) gray value statistics of (e).(at least three individual experiments, mean ± SEM, one-way ANOVA, ns indicates no significant difference, * p < 0.05).