In vivo visualization and characterization of inflamed intestinal wall: the exploration of targeted microbubbles in assessing NF‐κB expression

Abstract NF‐κB, a critical cytokine of inflammatory bowel diseases (IBD), is a viable marker to reflect the inflammatory activity of the intestine. We aimed to develop NF‐κB‐targeted microbubbles (MBs) and perform molecular contrast‐enhanced ultrasound (CEUS) to quantify NF‐κB expressions on the intestinal wall in IBD mice in vivo. In this study, NF‐κB‐targeted MBs were fabricated by connecting biotin‐loaded NF‐κB antibodies and avidin‐loaded MBs. NF‐κB‐targeted MBs presented as transparent and round bubbles with an average diameter of 1.03/μm±0.01. The specific binding of targeted MBs and inflammatory cells was validated by in vitro experiments, including flow cytometry, Western blot and immunofluorescence, which revealed the specific binding of targeted MBs and inflammatory cells. Subsequently, NF‐κB‐targeted CEUS imaging was performed on mice with chemical‐induced colitis, and the peak intensity (PI) and time‐to‐peak (TTP) were quantified. Pathological and immunohistochemical (IHC) examinations were further implemented. For the target CEUS group, fast enhancement followed by slow subsiding was observed. The PI of target CEUS of the IBD mice was significantly higher than that of non‐target CEUS of the IBD mice, healthy controls and target CEUS of the treated IBD mice (34835%[13379–73492%] VS 437%[236–901%], 130%[79–231%], 528%[274–779%], p<0.0001), in accordance with the IHC results of NF‐κB expressions. The TTP of target CEUS of the treated mice was significantly higher than that of untreated mice (35.7s [18.1–49.5s] VS 8.3s [4.2–12.5s], p<0.0001). Therefore, we suggested that NF‐κB‐targeted CEUS could accurately detect and quantify NF‐κB expressions on the intestinal walls of IBD, enabling the evaluation of intestinal inflammation.


| INTRODUC TI ON
Inflammatory bowel diseases (IBD) are systematic, chronic inflammatory diseases, including ulcerative colitis (UC) and Crohn disease (CD). 1,2 The disease often begins in adolescence and has a high recurrence rate, causing a huge economic and social burden. Chronic inflammation resulting in mucosal damage is essential to the pathogenesis of IBD. 3,4 The NF-κB pathway has been proved to be highly involved in the chronic inflammation process of IBD. 5 NF-κB is an essential family of inducible transcription factors that is intensively involved in the inflammatory processes and immune responses, including p50 (NF-κB1) and p52 (NF-κB2), p65 (RelA), RelB and c-Rel (Rel). The activity of NF-κB is triggered by the TNF family through different signalling pathways. 6 NF-κB family has alsso been proved to present a high-level expression in the intestinal mucosa of IBD and play a vital role in the mucosal healing of IBD. 7,8 Consequently, NF-κB has been deemed as an essential target of diagnosis and treatment of the disease. 9,10 A variety of natural and novel synthetic agents are developed to ameliorate colitis by targeting NF-κB, in addition to conventional anti-inflammatory agents. 11 Studies have shown that the expression level of NF-κB on intestinal epithelial cells is not only highly associated with the disease activity, but also indicative of the treatment response and the risk of developing colitis-associated cancer. [12][13][14] Hence, an accurate as- To note, due to its noninvasiveness and convenience, bowel ultrasound (US) has been recommended as a conventional imaging method for IBD in clinical practice. 15,16 Hence, the application of molecular CEUS in evaluating IBD can be promising in clinical practice. Previous studies have proved the feasibility of target CEUS in evaluating murine colitis by depicting the expressions of inflammatory molecules, including P-selectin and mucosal addressin cellular adhesion molecule (MAdCAM)-1. 17,18 Therefore, it is reasonable to assume that CEUS can also be used to evaluate the NF-κB expression in intestinal epithelial cells in vivo, thus facilitating further disease management.
In this study, we fabricated NF-κB-p65-targeted MBs by decorating NF-κB-p65 antibody molecules onto the surface of MBs, and NF-κB-targeted CEUS was performed on murine models with chemically induced colitis to quantify NF-κB-p65 expressions of colon tissues, aiming to explore its potentials in the evaluation and treatment monitoring of IBD.
Unbinding Abs were removed after washing with PBS solution. The upper layer of the mixture was obtained and subjected to volume measurement. In this way, the NF-κB-p65-targeted MBs were fabricated. The solution of targeted MBs was put in a counting plate to count the targeted MBs and measure the particle sizes (Multisizer 4e, Beckman, USA). We also observed the targeted MBs under a microscope (BX53, Olympus, Tokyo, Japan).

| The detection of binding rate of NF-κB-Abs and MBs
A total of 400μl NF-κB-p65-targeted MBs loaded with different volumes of NF-κB-p65 Abs (5μg, 10μg, 40μg) were used for detection of the binding rate using flow cytometry of Annexin

| Confocal laser scanning
We further performed confocal laser scanning to observe the intracellular distributions of the NF-κB-targeted MBs. The cells were added with NF-κB-FITC-targeted MBs and incubated for 30 min.
Then we used DAPI fluorescent dyes to stain the nuclei of cells.
The confocal laser scanning microscopy (Nikon Ti-E+C2, Japan) was performed to observe the intracellular distributions of the FITC-NF-κB-targeted MBs. For DAPI microscopic imaging, the excitation wavelength was 358 nm, and the emission wavelength was 461 nm.
For FITC on the NF-κB microscopic imaging, the excitation wavelength was 495 nm, and the emission wavelength was 520 nm.

| DSS-induced colitis model and grouping
The animal experiment was approved by the Institutional Animal Care and Use Committee of Peking Union Medical College Hospital.
A total of 30 female C57BL/6 mice (6-8 weeks of age, 18-24g) (Beijing, China) were prepared and acclimatized for 2 weeks before modelling at a room temperature. Twenty-four mice were distributed to the model groups and were fed with 3% dextran sodium sulphate (DSS) (Thermo Fisher, Shanghai, China) solution for nine days.
Six mice were fed with normal drinking water for nine days as the control.
Twelve mice of the 24 modelling mice were used as the treatment group. On Day 9, the DSS solution was switched to drinking water. And six mice were administrated with Mesalazine Enemas (Dr. Folk Pharma GmbH, Freiburg, Germany) through enema for 3 consecutive days (Mesalazine Enemas solvent, 0.1ml per mouse, once per day, 3 days in total). Another six mice in the experimental group were used as the non-treatment group. After drinking DSS water on Day 9, they switched to ordinary drinking water and were treated with 0.1ml saline for 3 consecutive days (0.1ml per mouse, once a day, 3 days in total). The mice were divided into five groups, with 6 mice in each group, listed as follows: Group A: NF-κB-targeted CEUS +IBD mice with colitis on Day 9; Group B: non-target CEUS +IBD mice with colitis on Day 9; Group C: NF-κB-targeted CEUS +healthy mice on Day 9; Group D: NF-κBtargeted CEUS +Mesalazine-treated IBD mice on Day 12; Group E: NF-κB-targeted CEUS +untreated IBD mice on Day 12.

| Clinical assessment
The clinical assessments of the mice were implemented on Days 0, 6, 9 and 12, including weight measurement and stool evaluation.

| Quantification of CEUS parameters
The SonoLiver Software (Philips, Andover, USA) for quantitative evaluation of CEUS was applied to delineate the time-intensity curves (TICs) and calculate the CEUS-related parameters, including time-to-peak (TTP) and peak intensity (PI) of the regions of interest (ROIs). For each animal, a segment of colon wall with 0.5cm length was selected as the ROI. After determining the ROI, the calculation would be done by the software automatically. For each CEUS examination, we performed three times of ROI drawing for calculation on the same bowel segment and took the mean value. The mean values of the parameters of different groups were utilized for analysis.

| Pathological and immunohistochemical (IHC) examination
Groups A-C were sacrificed on Day 9, and Groups D-E were sac-

| Characterizations of the NF-κB-p65-targeted microbubbles
The targeted MBs had an average diameter of 1.03/μm±0.01 and a concentration of 1.5×10 9 /ml. The histogram illustrating the particle size distribution and the microscope image is shown in Figure 1.

| The binding of NF-κB-p65 Abs and MBs
The MBs with different concentrations were successfully conjugated with biotin-FITC double-labelled NF-κB-p65 antibodies, depicted as compact globular stereo-structures with a uniform green aura by fluorescence microscopy (Figure 2). The morphology and concentration remained unchanged after 30 minutes of incubation.
According to the results of flow cytometry (Figure 3), the biotin-FITC double-labelled NF-κB-p65 antibodies with different concentrations and IgG antibodies specifically bond to avidin-labelled MBs, and the binding rate reached saturation at 40μg of antibodies.

| Cellular uptake of targeted MBs in vitro
The results of in vitro experiments are shown in Table 1 and

| Western blot
The image of protein charts of Western blot examination and the grey value analysis histogram is illustrated in Figure 5, from which an enhanced expression of NF-κB-p65 could be observed in G4 and G6, indicating that the RAW264.7 cells had higher expression level of NF-κB than NIH 3T3 cells.

| Immunofluorescence
The immunofluorescence images of the six groups are shown in

| Clinical assessment
The timeline for in vivo experiment is presented in Figure 8.

| CEUS results
The schematics of CEUS imaging and TICs are presented in Figure 9, and the CEUS parameters of the five groups are listed in Table 2.
Group A: After injection of MB bolus, significant enhancement of the hypertrophic colon wall was observed, followed by a slow subsiding, and the enhancement was kept at a high level ( Figure 9A1-4).
Group B: Significant enhancement and slow subsiding were also observed after injection. The contrast signals were decreased to a relatively low level at the end of the imaging process, but higher than surrounding tissues ( Figure 9B1-4).
Group C: Significant enhancement and fast subsiding were presented, and the peak intensity was equivalent with surrounding tissues. After subsiding, the signal intensity was lower than the surroundings ( Figure 9C1-4).
Group D: Slow increment of signals was visualized, and the peak intensity was slightly higher than surrounding tissues. The enhancement remained for a relatively long period and decreased slowly ( Figure 9D1-4).
Group E: The enhancement pattern was similar to Group 1 ( Figure 9E1-4).
The PI of Group A (target CEUS of IBD mice with colitis on Day 9) was significantly higher than Group B (non-target CEUS of IBD mice),

| Histological and IHC results
The HE and IHC staining results are presented in Figure 9, and quantitative IHC results of NF-κB-p65 expressions are shown in Table 2.
On Day 9 after DSS drinking, the intestinal villi became shorter, and the submucosa was highly oedema. The necrosis and shedding of

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest. The TTP was about 6.1s. Then, the signal intensity decreased slowly, but still maintained at a high level, significantly higher than surrounding tissues (A1-4). The inflammatory signs could be observed on HE staining (A5), and a high level of NF-κB-p65 expression was identified (A6). B. Mice with colitis receiving non-target CEUS imaging using naked MBs on Day 9. Significant enhancement was presented (PI: 582%). The TTP was about 8.4s. Then, the signal intensity decreased slowly to a relatively low level, but higher than surrounding tissues (B1-4). Inflammatory signs of the intestinal wall and enhanced NF-κB-p65 expression could also be detected in pathological results (B5-6). C. Healthy controls receiving CEUS on Day 9. A rapid enhancement was observed, followed by a relatively slow subsiding. The intensity was equal to the surroundings (C1-4). Normal bowel tissues were identified by pathology (C5-6). D. Mice with colitis and treated with anti-inflammatory drugs at the 12th day receiving target CEUS imaging using NF-κB-targeted MBs on Day 12. The signal intensity grew slowly (PI: 579%) and remained at a relatively high level with minimal decrease. The TTP was about 40.5s (D1-4). In pathology, minimal inflammatory changes were identified after treatment (D5). The NFκB-p65 expressions on segments of intestinal wall decreased after treatment (D6). E. Untreated mice with colitis receiving target CEUS imaging using NF-κB-targeted MBs on Day 12. Rapid enhancement and high level of peak intensity were obtained, similar to the results of A (E1-4). Severe inflammation of the intestinal wall and high level of NF-κB-p65 expression were illustrated by pathology (E5-6)

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.