Ultrasound molecular imaging for early detection of acute renal ischemia–reperfusion injury

Abstract Background Microcirculatory perfusion disorder and inflammatory response are critical links in acute kidney injury (AKI). We aim to construct anti‐vascular cell adhesion molecule‐1(VCAM‐1) targeted microbubbles (TM) to monitor renal microcirculatory perfusion and inflammatory response. Methods TM carrying VCAM‐1 polypeptide was constructed by biological coupling. The binding ability of TM to human umbilical vein endothelial cells (HUVECs) was detected. Bilateral renal ischemia–reperfusion injury (IRI) models of mice were established to evaluate microcirculatory perfusion and inflammatory response using TM. Thirty‐six mice were randomly divided into six groups according to the different reperfusion time (0.5, 2, 6, 12, and 24 h) and sham‐operated group (Sham group). The correlation of TM imaging with serum and histopathological biomarkers was investigated. Results TM has advantages such as uniform distribution, regular shape, high stability, and good biosafety. TM could bind specifically to VCAM‐1 molecule expressed by tumor necrosis factor‐alpha (TNF‐α)‐treated HUVECs. In the renal IRI‐AKI model, the area under the curve (AUC) of TM significantly decreased both in the renal cortical and medullary after 2 h of reperfusion compared with the Sham group (p < 0.05). Normalized intensity difference (NID) of TM at different reperfusion time was all higher than that of blank microbubbles (BM) and the Sham group (p < 0.05). Ultrasound molecular imaging of TM could detect AKI early before commonly used renal function markers, histopathological biomarkers, and BM imaging. AUC of TM was negatively correlated with serum creatinine (Scr), blood urea nitrogen (BUN), and Cystatin C (Cys‐C) levels, and NID of TM was linearly correlated with VCAM‐1, TNF‐α, and interleukin‐6 (IL‐6) expression (p < 0.05). Conclusions Ultrasound molecular imaging based on TM carrying VCAM‐1 polypeptide can accurately evaluate the changes in renal microcirculatory perfusion and inflammatory response, which might be a promising modality for early diagnosis of AKI.


| INTRODUCTION
Acute kidney injury (AKI) is a clinical syndrome characterized by a rapid decline in renal function over a short period. 1 AKI is characterized by high morbidity, rapid progression, and high mortality.3][4] In clinical practice, urine output and serum creatinine (Scr) are the commonly used indicators for detecting AKI.However, urine output is easily affected by diuretics and urinary tract diseases.6][7] Therefore, urine output or Scr cannot accurately detect AKI at the early stage, leading to missing optimal treatment opportunities and irreversible kidney damage.Therefore, there is an urgent need for reliable and accurate early detection means for AKI.
Ischemia-reperfusion injury (IRI) is one of the most common causes of AKI.The overproduction of reactive oxygen species (ROS), hemodynamic disorders, and inflammatory responses are the primary causes of IRI-AKI. 8,9Numerous studies have confirmed that changes in renal microcirculatory perfusion and inflammatory response are the early manifestations of AKI before the change of serum biomarkers, such as Scr and blood urea nitrogen (BUN). 10,11Therefore, early and accurate assessment of renal microcirculatory perfusion disorders and inflammatory response is crucial for improving renal function in AKI.
Contrast-enhanced computed tomography (CT)/ magnetic resonance imaging (MRI) has made some progress in evaluating renal perfusion and inflammatory response.Still, its clinical application is limited due to the nephrotoxicity of contrast agents.
Ultrasound molecular imaging has been widely used in the study of tumors, thrombosis, and atherosclerosis, as it can be enhanced at the molecular level by targeted ultrasound contrast agents that specifically aggregate in specific tissues to reflect changes in molecular expression in the tissue. 12,13Some scholars have conducted experimental studies on evaluating AKI using ultrasound molecular imaging.However, it is still in the exploratory stage, and there are few studies on the early assessment of AKI.Vascular cell adhesion molecule-1 (VCAM-1) is a critical molecular in the development of AKI, which promote the aggregation of white blood cells in the renal, leading to damage to the endothelial cell and obstruction of microvessels, accelerating changes in renal microcirculatory perfusion and inflammatory response.5][16] Therefore, anti-VCAM-1 targeted ultrasound contrast agents can achieve ultrasound molecular imaging of AKI to show changes in renal microcirculatory perfusion and inflammatory responses.
In this study, we constructed targeted microbubbles (TM) carrying

| Main reagents
Details are provided in the Data S1.

| Preparation of TM
The 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), 1,2-distearoylsn-glycero-3-phosphoethanolamine-N-[biotinyl (polyethylene glycol) 2000] (DSPE-PEG2000-Biotin), and 1,2-dipalmitoyl-sn-glycero-3-phosphate (DPPA) in a certain proportion were dissolved in a mixture of glycerol and phosphate-buffered saline (PBS).The mixture was heated in a water bath at 45 C for 30 min, then transferred into a penicillin bottle, filled with perfluoropropane gas, and shaken horizontally for 30 s in an ST-B series amalgamator (TUVRheinland, Cologne, Germany) at a speed of 4500 revolutions/min.After resting in a refrigerator at 4 C for 2 h, the sample was centrifuged in a 300 g centrifuge for 3 min to remove the insoluble lipid material at the bottom.The blank microbubbles (BM) were obtained.Streptavidin was added to the biotinylated microbubbles suspension at the ratio of 3 μg per 10 7 microbubbles and incubated at 4 C for 1 h.Then 0.32 μg of biotinylated VCAM-1 polypeptide was added and incubated at 4 C for 1 h.TM carrying anti-VCAM-1 polypeptide was obtained (Figure S1 shows the route pattern diagram of the preparation of TM).BM without carrying anti-VCAM-1 polypeptide was used as control group microbubbles.To directly verify that the biotin-streptavidin method can stably and effectively connect biotinylated polypeptides to the surface of the TM, Fluorescein Isothiocyanate (FITC)-modified biotinylated polypeptides and BM were bound by the same process, and the lipid membrane of TM was labeled with DiI dye.Then the fluorescence intensity of the surface of TM was observed under laser confocal scanning microscopy (LCSM) (Olympus Corporation, Kyoto, Japan).

| Characterization and detection of TM
The concentrations of TM and BM were counted using the Hemocytometer.The particle size and zeta potential of TM and BM were measured using a dynamic light scattering analyzer (Nano-ZS; Malvern, UK).The distribution of TM and BM was observed under an optical microscope (Olympus Corporation, Kyoto, Japan), and their morphology was observed under a transmission electron microscope (TEM) (HT7700; Hitachi, Tokyo, Japan).The FACSCelesta flow cytometry (BD Biosciences, San Jose, CA) was used to detect the binding efficiency between TM and anti-VCAM-1 polypeptide.The biosafety of TM was assessed by a Cell Counting Kit-8 (CCK-8) assay and a hemolysis assay.The prepared TM were stored in a refrigerator at 4 C, and their stability was evaluated by measuring the change in concentration and particle size at 0, 1, 2, 3, and 5 days after preparation.

| Evaluation of the binding ability of TM in vitro
Details are provided in the Data S1.

| In vitro imaging experiments of TM
An enhanced ultrasound imaging model was prepared using 1% agarose powder.Mindray Resona R9 color Doppler ultrasonic diagnostic instrument and L11-3U linear array probe were used to collect the images of enhanced ultrasound imaging in vitro.After fixing the probe and adjusting to the contrast-enhanced ultrasound (CEUS) mode (mechanical index of 0.072, the gain of 65 dB), the CEUS images of TM and BM were acquired at different concentrations (1.0 Â 10 8 /ml, 5.0 Â 10 7 /ml, 1.0 Â 10 7 /ml, 5.0 Â 10 6 /ml, and 1.0 Â 10 6 /ml).The contrast intensity of TM at the concentration of 5.0 Â 10 6 /ml was compared before and after high mechanical index ultrasound irradiation.

| Establishment of animal models
All animal experiments and methods were performed in accordance with the relevant guidelines and regulations approved by the Animal Ethics Committee of the Chinese People's Liberation Army (PLA) General Hospital (No. 2022-X18-91).Thirty-six male C57BL/6J mice, aged 6-8 weeks, were purchased from the Medical Experimental Animal Center of Chinese PLA General Hospital.
The mice were subjected to bilateral renal IRI through an established protocol described previously. 17In brief, the mice were anesthetized by intraperitoneal injection of 100 μl 1% pentobarbital sodium per 10 g of body weight.After the lateral abdominal opening, the renal hilum was separated, and the renal arteries and veins were clamped using atraumatic vascular clamps.The clips were released 35 min later, and the abdominal wall and muscles were sutured.The animals were randomly divided into five groups according to the different reperfusion time (0.5, 2, 6, 12, and 24 h).The sham-operated group (Sham group) was not subjected to clipping treatment, and other operations were the same as other groups.

| Ultrasound imaging of TM in the IRI-AKI model
The Mindray Resona R9 color Doppler ultrasound diagnostic instrument and the L11-3U line array probe were applied to perform in vivo ultrasound imaging in mice.CEUS was performed when each group of mice reached the corresponding reperfusion time.The largest coronal section of the right kidney was selected, and the probe was fixed and switched into CEUS mode (mechanical index of 0.072, the gain of 65 dB, image depth of 1.5 cm, frame rate of 12 fps).TM and BM with a concentration of 5 Â 10 6 /ml were injected through the tail vein of mice in random order.
The dynamic storage and timing keys were activated simultaneously to save the vibrant images for 3 min continuously.The quantitative analysis of the CEUS image was carried out using the built-in analysis software of the ultrasound diagnostic instrument.The regions of interest (ROI) in cortical and medullary were selected, respectively, and time-intensity curves (TICs) were plotted to obtain the area under the curve (AUC) to analyze the microcirculatory perfusion disorder of AKI.
Inflammatory response evaluation of AKI was performed using a destruction-complementation approach.After acquiring the first 3 min of contrast images, all microbubbles were destructed for 3 s using a "flash" with a high mechanical index of 0.503.After the destruction, ultrasound imaging was performed for 10 s to acquire signals from free-circulating microbubbles supplemented by an external imaging plane.Quantitative analysis was performed using the built-in analysis software.The contrast intensity before the destruction pulse was the total contrast signal from target-bound and free-circulating microbubbles.In contrast, the contrast intensity after the destruction was from re-perfused free-circulating microbubbles only. 18,19The quantification of the targeting contrast signal was calculated as the normalized intensity differences (NIDs [%]) = (pre-destruction contrast intensity À post-destruction contrast intensity)/pre-destruction contrast intensity Â 100%]. 20

| Renal function and serum biomarkers of AKI
After the CEUS examination, blood was taken from the inferior vena cava of the mice.Scr, BUN, and Cystatin C (Cys-C) levels were measured using the biochemical analyzer Hitachi-8000 (Hitachi, Tokyo, Japan).The levels of kidney injury molecule-1 (KIM-1) and neutrophil gelatinase-associated lipocalin (NGAL) were measured using enzymelinked immunosorbent assay (ELISA) kits.

| Renal histopathology evaluation
The kidney was treated with 4% paraformaldehyde fixation, dehydration, and paraffin embedding, and the tissue was sectioned and stained with periodic acid-Schiff (PAS).Ten slices with 400-fold non-overlapping visual fields were randomly selected, including five in the cortical region and five in the corticomedullary junction.Acute renal tubular injury score was performed blind to evaluate the loss of brushy margins, tubular dilatation, cast formation, tubular necrosis, neutrophil infiltration, etc.The tubular injury was classified into six grades 21 : 0, normal; 1 score, mild injury (involving 0% to 10%); 2 scores, moderate damage (11%-25% involvement); 3 scores, severe injury (26%-49% involvement); 4 scores, highly severe injury (50%-75% involvement); 5 scores, extensive damage (>75% involvement).All assessments were carried out by two pathologists who did not know the experimental conditions.Immunohistochemical staining of paraffin sections was used to analyze the protein expression of VCAM-1, and CD31 staining was used to assess renal vessel density.After routine dewaxing, endogenous peroxidase was removed using 3% H 2 O 2 , and antigen repair was performed.
After goat serum was blocked at room temperature for 30 min, the sections were incubated overnight at 4 C with rabbit anti-mouse VCAM-1 antibody (1:500 dilution) and rabbit anti-mouse CD31 antibody (1:2000 dilution), respectively.After rinsing, the sections were incubated with horse radish peroxidase (HRP)-labeled goat anti-rabbit secondary antibody for 50 min at room temperature, followed by DAB development and hematoxylin re-staining of cell nuclei.After dehydration, clearing, and neutral gum sealing, the slices were microscopically observed and comparatively analyzed for VCAM-1 expression and vessel density in kidney tissue at different reperfusion time.Five non-overlapping fields of view at 400Â were randomly selected.The quantification was completed by two researchers unaware of the experimental conditions.
Western blot assay was used to analyze the protein expression of VCAM-1 (detailed procedures are provided in the Data S1).
Cell apoptosis in kidney paraffin sections was detected by TUNEL staining.Five randomly selected non-overlapping fields of view at 400Â were used to count the number of apoptotic cells.The levels of tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6) in kidney tissue were measured using ELISA kits.

| In vivo distribution of TM
Detailed procedures are provided in the Data S1.

| The characterization of TM
The concentrations of BM and TM were (10.87 ± 1.10) Â 10 8 /ml and (9.30 ± 0.56) Â 10 8 /ml, respectively, and there was no significant difference between them (t = 2.199, p = 0.093).The detection results of the dynamic light scattering instrument showed that the particle sizes of BM and TM were (2280.23 ± 446.55) nm and (2728.61± 514.11) nm, respectively (Figure 1a, b).The potential of BM was (À15.72 ± 3.62) mV (Figure 1c) because of the presence of anionic phospholipid DPPA in the lipid components used to prepare BM.The potential of TM was (À7.67 ± 2.51) mV (Figure 1d), which was higher than that of BM, indicating that the positive charge of the amino group on VCAM-1 polypeptide could neutralize part of the negative charge of anionic phospholipid, which also indirectly confirmed that TM carried anti-VCAM-1 polypeptide.Under the transmission electron microscope, TM were regular black circles with clear boundaries, and the particle size was about 2600 nm, consistent with the particle size measured by the dynamic light scattering instrument (Figure 1e).
Under the optical microscope, BM and TM were evenly distributed and similar in particle size.There was no apparent aggregation or burst phenomenon (Figure 1f, g).
Under the laser confocal scanning microscopy (LCSM), the lipid membrane of DiI-labeled TM showed red fluorescence (Figure 1h).FITC-labeled anti-VCAM-1 polypeptide carried on the surface of TM showed green fluorescence (Figure 1i), and the two fluorescence could completely overlap and showed yellow (Figure 1j), directly confirming that anti-VCAM-1 polypeptide was carried on the surface of TM.The binding efficiency between TM and anti-VCAM-1 polypeptide detected by flow cytometry was (95.48 ± 1.98) %, significantly higher than that of BM (1.83 ± 0.74) %, and the difference was statistically significant ( p < 0.05) (Figure 1k, l).The CCK8 experiment showed no significant cytotoxicity of BM and TM (Figure 1m).The hemolysis experiment showed that BM and TM had no significant hemolytic effect (Figure 1n).
The concentration of TM stored at 4 C decreased continuously, with concentrations of (8.23 ± 0.60) Â 10 8 /ml, (6.43 ± 0.55) Â 10 8 /ml, (4.40 ± 1.31) Â 10 8 /ml, and (2.27 ± 1.25) Â 10 8 /ml on days 1, 2, 3, and 5, respectively.After 2 days of storage, the concentration of TM was significantly lower than that of freshly prepared TM (p < 0.05) (Figure S2a).The particle size of TM gradually increased over time with particle sizes of (3219.30± 862.10) nm, (4742.04 ± 919.00) nm, (7274.72 ± 1418.09) nm, and (9074.04± 1546.35) nm on days 1, 2, 3, and 5, respectively.The particle size of TM stored for 3 days was significantly larger than that of freshly prepared TM (p < 0.05) (Figure S2b).Flow cytometry was used to detect the binding ability of TM and BM to HUVECs treated with TNF-α.The results showed that HUVECs treated with TNF-α were significantly bound by TM.At the same time, no significant binding was observed in other groups (Figure 2g), which was consistent with the detection results of cellular immunofluorescence technology.Further quantitative analysis showed that the binding efficiency of TM to HUVECs after TNF-α treatment was as high as (87.67 ± 5.24) %, which was significantly higher than the other groups (p < 0.05) (Figure 2h).

| In vitro enhanced ultrasound imaging of TM
In vitro, the enhanced ultrasound imaging model showed that there was no statistical difference in contrast intensity of enhanced ultrasound imaging between TM and BM in each concentration gradient (p > 0.05) (Figure 3a, b).After high mechanical index ultrasound irradiation, the ultrasound contrast intensity significantly decreased (p < 0.05) (Figure 3c, d).decrease in the medullary phase.In addition, quantitative analysis results showed the cortical AUC of TM in the reperfusion group decreased with time.The cortical and medullar AUCs of TM in the reperfusion 2 h group were significantly lower than that in the Sham group, and the 24 h reperfusion group had the lowest AUC (p < 0.05).BM's cortical and medullary AUC trends were consistent with TM's.However, the cortical AUC of BM in the reperfusion 0.5 h group was significantly lower than that in the Sham group (p < 0.05) (Figure 4b-e).

| Enhanced ultrasound imaging of TM in vivo
Figure 5a showed renal inflammatory response evaluation using TM and BM in the AKI mice model.The peak intensity of TM in the reperfusion groups was higher than that of the BM.Quantitative analysis showed that compared with the Sham group, the NID of TM in all reperfusion groups was increased to varying degrees, and the highest level was found in the reperfusion 2 h group, with statistical significance ( p < 0.05).There was no significant difference in the NID of BM between the different reperfusion time groups and the Sham group (p > 0.05).The Paired analysis comparison showed no statistically significant difference in the Sham group's NID between TM and BM ( p > 0.05).In contrast, the NID of TM in the reperfusion groups was significantly higher than that of BM ( p < 0.05) (Figure 5b).

| Renal function and early biomarkers of AKI
Compared with the Sham group, the Scr and BUN levels in the reperfusion 6 h group were significantly increased, and the highest level was in the reperfusion 24 h group, with statistical significance (p < 0.05).There were no significant differences in Scr and BUN levels in the reperfusion 0.5 and 2 h groups compared with the Sham group (p > 0.05) (Figure 6a, b).
Compared with the Sham group, the Cys-C levels in the reperfusion 12 and 24 h groups were increased, with a statistically significant difference ( p < 0.05).The differences were not statistically significant in the reperfusion 0.5, 2, and 6 h groups compared to the Sham group (p > 0.05) (Figure 6c).The KIM-1 and NGAL levels increased significantly from the reperfusion 2 h group compared to the Sham group and were highest in the reperfusion 24 h group, a statistically significant difference ( p < 0.05).The differences in KIM-1 and NGAL levels in the reperfusion 0.5 h group compared to the Sham group were not statistically significant (p > 0.05) (Figure 6d, e).

| Renal histology evaluation
PAS staining showed that compared with the Sham group, there was no significant tubular injury in the reperfusion 0.5 and 2 h groups, and the damage was gradually aggravated in the reperfusion 6 h group.
The tubular injury in the 24 h reperfusion group was the most severe, mainly at the corticomedullary junction.The renal tubules were characterized by the loss of brushy margins, tubular dilatation, and cast formation (Figure 7a).Renal tubular injury scores showed no statistically significant difference in the reperfusion 0.5 h and 2 h groups compared to the Sham group ( p > 0.05).The difference was statistically significant in the reperfusion 6, 12, and 24 h groups compared to the Sham group ( p < 0.05) (Figure 7b). Figure 7a showed TUNEL staining for the detection of cell apoptosis.
Compared with the Sham group, the degree of apoptosis was the least in the 0.5 h reperfusion group and gradually increased in the 2 and 6 h reperfusion groups.The apoptosis of cells reached a peak in the reperfusion 12 h group.The reperfusion 24 h group showed decreased detected cell apoptosis due to severe cell shedding.This result further indicated that the degree of AKI in the reperfusion 24 h group was the most severe.Further quantitative analysis showed that compared with the Sham group, there was no statistically significant difference in the number of apoptotic cells in the reperfusion 0.5 h group (p > 0.05).In contrast, the number of apoptotic cells in the other reperfusion groups was significantly higher than that in the Sham group (p < 0.05) (Figure 7c).
As shown in Figure 7d,e, the levels of TNF-α and IL-6 in renal tissue of each reperfusion group had statistical significance compared with those of the Sham group ( p < 0.05).

| Renal vessel density and VCAM-1 expression
Figures 8a and S4 show the immunohistochemical staining results of CD31 and the protein expression of VCAM-1 in different groups.Quantitative analysis showed that there was no statistically significant difference in renal vessel density between the reperfusion groups and the Sham group (p > 0.05) (Figure 8b).Compared with the Sham group, VCAM-1 expression was elevated to different degrees in all reperfusion groups, and the difference was statistically significant (p < 0.05) (Figures 8c and   S4).In addition, Pearson's correlation analysis showed a linear positive correlation between the NID of TM and VCAM-1 expression (r = 0.8457, p < 0.05) (Figure 8d).

| Correlation analysis between ultrasound variables and serum and tissue indicators
Pearson's correlation analysis showed that NID and AUC of TM were linearly correlated with the levels of TNF-α and IL-6 in renal tissue of all groups ( p < 0.05) (Figure S5a-d).In addition, the AUC of TM was linearly negatively correlated with serum Scr, BUN, and Cys-C levels (p < 0.05) (Figure S5e-g).There was no significant correlation between the NID of TM and serum levels of Scr, BUN, and Cys-C ( p > 0.05).In contrast, a small number of fluorescence signals were observed in the kidneys of mice in other groups (Figure S6a).Quantitative analysis showed that the fluorescence intensity of the kidneys in the IRI group was significantly higher than that in the other groups, and the difference was statistically significant ( p < 0.05) (Figure S6b).

| In vivo distribution of TM
The distribution of DiR-labeled TM in vivo was similar to that of BM.The fluorescence signal was first distributed throughout the body and then concentrated in the liver and spleen (Figure S6c,d).

| DISCUSSION
In our study, TM carrying anti-VCAM-1 polypeptide for ultrasound molecular imaging was successfully constructed.TM had homogenous particle size, round morphology, high stability, and exemplary safety.In this study, we established a bilateral renal IRI model in mice.In our experiments, the presence of AKI was defined by a 1.5-fold and higher increase in Scr according to the AKI staging proposed by the KDIGO definition. 1,22As reported in previous literature, the bilateral renal IRI model in mice is the most stable compared with unilateral renal IRI (with or without contralateral renal resection) and is most relevant to human pathology. 17Regarding the selection of ischemia time, previous studies have reported that kidney ischemia time exceeding 30 min can cause severe kidney injury, and most studies have focused on ischemia time around 30-45 min.The longer the ischemia time, the more severe the kidney injury and the higher the mortality rate of the mice.We ultimately chose 35 min of ischemia to construct the AKI model to ensure the success rate.
In our study, to assess the dynamic process of AKI inflammatory response, we set up more detailed reperfusion time points to study.
The reperfusion time began at 0.5 h, followed by 2, 6, and 12 h, until 24 h of reperfusion.Previous studies were limited to 4 and 24 h after reperfusion to evaluate inflammatory response. 23The results showed the NID of TM in different reperfusion groups increased to varying  degrees, with statistically significant differences compared with the Sham group.Moreover, all the NID of TM were also higher than those of BM in the same reperfusion groups.Correlation analysis also showed that the NID of TM was positively correlated with the VCAM-1 protein expression.In addition, studies have confirmed that TNF-α can induce the expression of VCAM-1 and IL-6. 24Our research showed that NID of TM was linearly and positively correlated with the levels of TNF-α and IL-6 in renal tissues, which is consistent with previous findings.Endothelial damage and inflammatory reactions promote the occurrence and development of AKI. 5,257][28] These results indicate that ultrasound molecular imaging based on TM carrying anti-VCAM-1 polypeptide can effectively evaluate endothelial injury and inflammatory response in AKI in the early stage.
In the present study, we first investigated the value of TM in quantitatively assessing the microcirculatory perfusion disorder of AKI.We found that TM and BM were consistent in assessing renal microcirculation reperfusion in IRI-AKI.The contrast intensity of both TM and BM rapidly peaked in the cortical phase, followed by a decrease in the medullary phase in AKI.Most studies have suggested that AUC has a specific value in assessing blood flow perfusion. 29th the severity of IRI-AKI, renal blood perfusion becomes worse.In the cortex and medulla, both TM and BM exhibited a lower AUC in the reperfusion group than in the Sham group, consistent with the results of renal perfusion changes in canine renal IRI-AKI assessed by Lee using SonoVue. 30CD31 staining showed no difference in renal vessel density in the reperfusion group compared with the Sham group, which is consistent with previous studies. 31The destruction of the vascular endothelial structure increases vascular permeability and intense vasoconstriction, resulting in changes in renal microcirculatory perfusion. 29,32,33In short, ultrasound molecular imaging based on TM could reflect the microcirculation disorder of AKI.
There are some limitations in our study.
anti-VCAM-1 polypeptide to explore their specific binding ability and target-enhanced imaging ability during the development of AKI, investigate the changes of microcirculatory perfusion and inflammatory response in IRI-AKI, and clarify the correlation of TM in ultrasound molecular imaging with serum and histopathological biomarkers of AKI.Overall, TM carrying anti-VCAM-1 polypeptide may have clinical application potential for ultrasound molecular imaging and improve the accuracy of early detection of AKI.

2. 11 |
Statistical analysisStatistical analyses were performed using GraphPad Prism 8.0 (GraphPad Software, Inc., San Diego, CA) and SPSS 25.0 statistical software.All quantitative data are expressed as the mean ± SD.Kolmogorov-Smirnov test was used to test the data for normal distribution.Student's t-test or one-way ANOVA were used to compare and analyze all quantitative parameters.When the variance homogeneity test p > 0.05, Tukey's multiple comparison post-test was used.Otherwise, Dunnett's T3 postmortem test was applied.Pearson's correlation analysis was performed between ultrasound variables and serum and histopathological parameters.p < 0.05 was considered a statistically significant difference.

Figure
FigureS3shows the VCAM-1 protein expression in HUVECs under the LCSM, with nuclei in blue and coralite488-conjugated VCAM-1 in green.VCAM-1 protein expression was observed in HUVECs treated with TNF-α but not untreated cells.The binding of microbubbles to HUVECs was observed under the LCSM.The cell nucleus was blue, and microbubbles labeled with DiI were red.The TNF-α treated HUVECs reacted with TM, and many TM was seen around the cells, while only a tiny number of microbubbles were observed around the cells of the other groups (Figure2a-f).

Figure
Figure 4a demonstrated the renal blood flow perfusion evaluation in the AKI mice model using TM and BM.The TICs trend of TM was consistent with that of BM during the assessment of blood flow perfusion.Both TM and BM showed a rapid peak in the cortical phase, followed by a

F
I G U R E 2 The binding ability of microbubbles in vitro.(a-f) The binding ability of microbubbles to HUVECs under laser confocal scanning microscopy.Blocked indicates the HUVECs were pre-blocked with VCAM-1 polypeptide.Blue fluorescence indicates cell nuclei, and red fluorescence indicates microbubbles.(g) Flow cytometry analysis of the binding ability of microbubbles to HUVECs.(h) The binding efficiency of microbubbles to HUVECs (n = 6 per group).Data in the graph represent the mean ± SD, and p value was determined by one-way ANOVA and Dunnett's T3 post hoc test.*p < 0.05 compared with the other groups.BM, blank microbubbles; HUVEC, human umbilical vein endothelial cell; TM, targeted microbubbles; TNF-a, tumor necrosis factor-alpha.
In vitro imaging of TM.(a) Ultrasound images of microbubbles at different concentrations in vitro.(b) Quantification of contrast intensity of microbubbles in vitro (n = 3 per group).(c) Ultrasound images of TM before and after destruction.(d) Quantification of contrast intensity of TM before and after destruction (n = 3 per group).Data in the graphs represent the mean ± SD, and p values were determined using t test.*p < 0.05.BM, blank microbubbles; TM, targeted microbubbles.
Small animal in vivo fluorescence imaging showed that after intravenous injection of DiR-labeled TM, a large amount of aggregated fluorescent signals was observed in the kidneys of mice in the IRI group.

F
I G U R E 4 Quantitative evaluation of renal blood perfusion.(a) CEUS images for assessing renal blood perfusion by TM and BM in different reperfusion time groups and sham group.The yellow and red lines represent the time-intensity curve of renal cortical and medullary ROI, respectively.(b) Comparison of BM in the cortical by AUC (n = 6 per group).(c) Comparison of TM in the cortical by AUC (n = 6 per group).(d) Comparison of BM in the medullary by AUC (n = 6 per group).(e) Comparison of TM in the medullary by AUC (n = 6 per group).Data in the graphs represent the mean ± SD, and p values were determined by one-way ANOVA and Tukey's or Dunnett's T3 post hoc test.*p < 0.05.AUC, area under the curve; BM, blank microbubbles; TM, targeted microbubbles.
TM can successfully bind to the VCAM-1 molecule expressed by TNF-α treated HUVECs through anti-VCAM-1 polypeptide on their surface.In the IRI-AKI model, TM had a good enhanced ultrasound imaging effect, can accurately demonstrate the change of microcirculatory perfusion and inflammatory responses, and can be used as an agent for the early detection of AKI.Compared with common markers, including renal function markers (Scr and BUN), early biomarkers (Cys-C, KIM-1, and NGAL), pathological damage, and ultrasound imaging of BM, ultrasound molecular imaging based on TM carrying anti-VCAM-1 had higher sensitivity.Moreover, ultrasound molecular imaging based on TM carrying anti-VCAM-1 has good consistency with inflammatory factors (VCAM-1, TNF-α, and IL-6).

F I G U R E 5
Quantitative evaluation of inflammatory response.(a) Ultrasound images for assessment of inflammatory response by TM and BM in different reperfusion time groups and sham group.(b) The comparison of NID of TM and BM (n = 6 per group).Data in the graphs represent the mean ± SD, and p values were determined by one-way ANOVA and Tukey's post hoc test.*p < 0.05.BM, blank microbubbles; NID, normalized intensity difference; TM, targeted microbubbles.F I G U R E 7 Renal histology evaluation.(a) Histopathologic images of the kidney using PAS and TUNEL staining; scale = 20 μm.(b) Comparison of the renal tubular injury score (n = 6 per group).(c) Comparison of the number of apoptotic cells (Â400) (n = 6 per group).(d) Comparison of kidney tissue TNF-a levels (n = 6 per group).(e) Comparison of kidney tissue IL-6 levels (n = 6 per group).Data in the graphs represent the mean ± SD, and p values were determined by one-way ANOVA and Tukey's or Dunnett's T3 post hoc test.*p < 0.05.IL-6, interleukin-6; PAS, periodic acid-Schiff; TNF-a, tumor necrosis factor-alpha.

F I G U R E 6
Serum biomarkers of AKI.(a) Comparison of the Scr levels (n = 6 per group).(b) Comparison of the BUN levels (n = 6 per group).(c) Comparison of the serum Cys-C levels (n = 6 per group).(d) Comparison of the serum KIM-1 levels (n = 6 per group).(e) Comparison of the serum NGAL levels (n = 6 per group).Data in the graphs represent the mean ± SD, and p values were determined by one-way ANOVA and Tukey's or Dunnett's T3 post hoc test.*p < 0.05.BUN, blood urea nitrogen; Cys-C, Cystatin C; KIM-1, kidney injury molecule-1; NGAL, neutrophil gelatinase-associated lipocalin; Scr, serum creatinine.
First, the early detection of TM in IRI-AKI was validated only after 35 min of ischemia.Hence, further studies are needed to include different ischemia time mouse models.Second, renal IRI can cause inflammation and injury in distal organs, and we did not evaluate the impact of other organs on TM.Third, oxidative stress also plays a vital role in the occurrence and development of AKI, and future experimental research in this area will be conducted to detect AKI early from multiple perspectives.Finally, the mice model of AKI cannot fully reflect the characteristics of clinical AKI patients.Large animal models are needed to confirm the early diagnostic effect of TM for clinical translation.

5 |
CONCLUSION In summary, TM carrying anti-VCAM-1 polypeptide was successfully constructed, which possessed the attractive features of high safety, high specificity, and excellent enhanced imaging effect.TM-based ultrasound molecular imaging can early and accurately demonstrate the change of microcirculatory perfusion and inflammatory response in AKI earlier than commonly assessed markers of renal function, biomarkers, and histopathological indicators.These may provide a sensitive and specific technique to achieve early warning of AKI, thereby promptly guiding AKI treatment and improving the prognosis of AKI.

F
I G U R E 8 Immunohistochemical staining results.(a) Immunohistochemical image of CD31 and VCAM-1.(b) Quantitative analysis of vessel density (n = 6 per group).(c) Quantitative analysis of VCAM-1 expression (n = 6 per group).(d) Pearson's correlation analysis between the NID of TM and VCAM-1 expression.Data in the graphs represent the mean ± SD, and p values were determined by one-way ANOVA and Tukey's or Dunnett's T3 post hoc test.*p < 0.05.NID, normalized intensity difference; VCAM-1, vascular cell adhesion molecule-1.