Establishment of Coloproctitis Cancer Model in Mice and Evaluation of Therapeutic Effect of Chinese Medicine

Colorectal cancer (CRC) is a common gastrointestinal malignancy, the third most common malignancy, and the second most common cause of death in the world, accounting for 10% of the global cancer incidence and 9.4% of the total cancer-related death^1,2. Genetic factors, chronic inflammation, high-fat diet, diabetes, and abnormal intestinal flora are risk factors for CRC^3,4. Among them, inflammatory bowel disease, especially ulcerative coloproctitis (UC), is a clear risk factor for CRC^5,6. UC-associated CRC (UC-CRC) is a transition process of inflammation, atypical hyperplasia, and cancer based on chronic inflammation of the colorectum, which is different from the typical adenoma-adenocarcinoma development model of CRC^7,8. Compared with the general population, the risk of CRC is approximately 10-40 times higher in patients with inflammatory bowel disease⁹.

Currently, surgery is still the standard treatment for CRC, and depending on the location and stage of the tumor, radiation therapy, systemic drug therapy, or a combination of both are possible¹⁰. Although these traditional treatment modalities have made great progress, due to the high heterogeneity and recurrence rate of CRC, the prognosis is poor, and the treatment effect is not ideal^11,12. Therefore, early detection, early diagnosis, and comprehensive treatment are key to improving the survival rate of CRC patients, and it is particularly important to pay attention to the transformation of UC to CRC. Over the years, traditional Chinese medicine (TCM) has attracted much attention in the treatment of UC-CRC or chronic gastritis due to its limited side effects and significant efficacy. Based on dialectical treatment, famous Chinese medicine practitioners of various generations have created a large number of classic prescriptions, such as Huangqi Jianzhong decoction¹³, Sijunzi decoction¹⁴, and Sishen pill¹⁵.

Liujunzi decoction (LJZD) originated from the works of Yi Xue Zheng Zhuan compiled in the Ming Dynasty and is a classic prescription in TCM¹⁶. As shown in Table 1, LJZD consists of six traditional herbs, including Codonopsis pilosula (Franch.) Nannf. (Dangshen), Poria cocos (Schw.) Wolf (Fuling), Atractylodes macrocephala Koidz. (Baizhu), Glycyrrhiza uralensis Fisch. (Gancao), Citrus reticulata Blanco (Chenpi) and Pinellia ternata (Thunb.) Breit (Banxia), which has the effect of replenishing qi and strengthening the spleen, drying dampness, and resolving phlegm. In modern clinical practice, it is often used to treat chronic gastritis, gastric ulcers, and duodenal ulcers. Modern pharmacological research has shown that LJZD and modified LJZD have high application value in the adjuvant treatment of UC and digestive tract cancer^17,18,19.

At present, there are many ways to construct UC-CRC mouse models, but the azoxymethane (AOM)/dextran sulfate sodium (DSS) induced mouse model is the most widely used UC-CRC model; the clinical symptoms, morphological, and pathological observations have proved that the model is very similar to human UC-CRC^20,21. The basic principle is to first induce carcinogenesis with chemical carcinogen AOM and then continuously expose mice to the inflammatory stimulation environment of DSS to simulate the continuous damage and repair of intestinal mucosal epithelium, thereby constructing a UC-CRC mouse model²². The aim of this study is to establish a mouse model of UC-CRC by intraperitoneal injection of AOM and cyclic stimulation of DSS in the short term and to evaluate the effect of the drug and the molecular mechanism of LJZD on UC-CRC in order to provide a scientific basis for the treatment of UC-CRC.

The animal procedure has been approved by the Ethics Committee of Changchun University of Chinese Medicine (Record number: 2021214). Specific pathogen-free C57BL/6J mice (8-10 weeks, weight 18-22 g), male and female, were housed in independently ventilated cages at 22 °C and 65% relative humidity. The mice began the experiment after 7 d of adaptive feeding, during which they had free access to water and diet.

1. Drug preparation

1. Preparation of LJZD
   ​NOTE: The Chinese medicine used was purchased from the Affiliated Hospital of Changchun University of Chinese Medicine and was identified as genuine Chinese medicine (see Table 1).
   1. Put Dangshen (12 g), Baizhu (12 g), Gancao (6 g), Chenpi (12 g), Jiangbanxia (9 g) into a specialized ceramic pot (see Table of Materials). Add 1000 mL of distilled water and soak at room temperature for 1 h (see Figure 1A).
   2. Powder 12 g Fuling into fine powder with a grinder and soak in 300 mL of distilled water in another container for 1 h at room temperature.
   3. Mix the above Chinese medicine in the specialized ceramic pot. Boil the mixture and then keep on medium flame until only 300 mL of decoction remains. Use medical gauze for filtering and preserve the filtrate at room temperature.
   4. Add 1000 mL of distilled water and repeat the above decoction operation once more.Filter again using medical gauze. Combine the filtrate and boil it until only 150 mL remains.
   5. Centrifuge the concentrated liquid at 10,000 x g for 5 min, and further concentrate the obtained supernatant to 30 mL on medium heat. Transfer the final concentrate to a dish and dry it using an electric drying oven until only the solute remains as a powder.
   6. Weigh the above solid and dissolve it in sterile distilled water to obtain a solution containing 22.85 mg of drug per 0.2 mL (114.16 mg/mL), which is the daily dosage of mice.
2. Preparation of 5-amino salicylic acid
   ​NOTE: 5-amino salicylic acid (5-ASA; see Table of Materials) has a good preventive effect on UC-CRC, and it was used as a positive drug in this study²³.
   1. Dissolve 64 mg 5-ASA powder in 200 mL of sterile distilled water to obtain 1.82 mg/mL 5-ASA solution. The daily dosage for a single mouse was 0.2 mL.
3. Preparation of AOM injection solution
   1. Add 2.5 mL of sterile distilled water into 25 mg AOM powder (see Table of Materials), and mix by vortex mixer (see Table of Materials) to make 10 mg/mL AOM stock solution, store at -20 °C until use.
   2. Prepare the AOM injection solution by diluting the AOM stock solution with sterile distilled water at 10:1 (1 mg/mL).

2. Establishment of UC-CRC model

NOTE: The experiment was divided into 4 groups: control, model, LJZD, and 5-ASA group, 10 mice in each group. Except the control group, the other groups were treated with AOM and DSS.

1. Intraperitoneal injection of AOM injection solution
   NOTE: After adaptive feeding for 7 d, mice were given AOM injection solution (1 mg/mL) by intraperitoneal injection (see Figure 1B).
   1. Hold the mouse with its belly up and head slightly down. Grip the back skin to tighten the abdominal skin and pierce the skin approximately 1 cm to the right of the intersection of the root line of both thighs on the mid-abdomen line with a 1 mL syringe (see Table of Materials).
   2. Push the 1 mL syringe needle to a distance of 3-5 mm under the skin, keeping it parallel to the abdominal midline, and insert the needle 0.3-0.5 mm into the abdominal cavity at a 45°.
   3. After the tip passes through the abdominal muscle, the operator senses a sudden loss of resistance. Subsequently, pull the syringe outward and backward to observe whether liquid seepage occurs. If not, push the AOM injection solution slowly into the mice at 0.1 mL/10 g.
2. Cyclic stimulation of 2% DSS solution
   ​NOTE: Each AOM-induced mouse was given 500 mL of DSS solution at weeks 3, 6, and 9, and the mice drank freely during this period.
   1. Prepare 2% DSS solution by adding 500 mL of sterile distilled water to 10 g DSS (see Table of Materials). Mix with a vortex mixer and store at 4 °C until use.
   2. Each AOM-induced mouse drinks 500 mL of 2% DSS solution freely for 7 d on weeks 3, 6, and 9 after AOM induction.

3. Drug treatment

NOTE: Adult humans need 63 g LJZD per day. According to the conversion formula of experimental mouse and human drug dose, equivalent experimental dose for mice (mg/kg) = human dose (mg/kg)/body weight (60 kg) x 9.1, the daily dose of mice was about 9.6 g/kg.

1. Treat the LJZD and 5-ASA group with 0.1 mL/10 g of the prepared LJZD and 5-ASA solution by gastric gavage at week 7 and week 15, respectively.
   1. For intragastric administration in mice, perform the following procedure. Hold the mouse in the left hand, and in the right hand, hold the stomach perfusion apparatus. Insert the syringe needle in the mouth and slide it down the back wall of the pharynx of the mice. Slide down the pharynx as the mice swallow and continue to move ahead. When there was a sense of resistance, and the syringe could be pushed into the pharynx, pull the needle out and complete the injection.
2. Treat the control and model group with the same amount of saline (see Table of Materials).
3. Treat the mice in each group with appropriate drug once a day at the same time during the administration period.

4. Evaluation of UC-CRC model and efficacy of LJZD

1. Disease activity index score
   NOTE: According to Table 2, the disease activity index (DAI) score was evaluated by combining weight loss, fecal viscosity, and stool bleeding of the mice.
   1. Record the weight of the mice daily from the beginning of adaptive feeding until the end of drug treatment.
   2. Observe the fecal consistency of each experimental mouse carefully and record the bowel movements as one of the three conditions: normal, loose stools, and watery diarrhea.
   3. Record the fecal bleeding of experimental animals as one of the three conditions: no bleeding, little bleeding, and visible blood in stool.
2. Detection of IL-6 level in serum
   1. Treat mice with LJZD or 5-ASA for 9 weeks. Fix the mice by grasping the neck skin of the mouse with the left hand and gently pressing on the experimental table to take the lateral decubitus position. Cut off the whiskers of the mice with scissors. Cut off the whiskers of the mice with scissors carefully(see Table of Materials) to prevent blood contamination.
      NOTE: Mice that could not wiggle freely were considered to be properly fixed. If this does not happen, the mice should be fixed.
   2. Anesthetize the mouse by inhalation of 2% isoflurane. Sterilize the skin around the eyeball with ethanol (see Table of Materials). Press the eye skin on the side gently to make the eyeball congested and protruding.
   3. Clamp the eyeball with an elbow tweezer and remove the eyeballs accurately and quickly. Allow the blood to drip into the centrifugal tube (see Table of Materials). In the process, tap the mouse's heart to speed up blood collection.
   4. Keep the collected blood at room temperature for 30 min and centrifuge it at 3,500 x g for 10 min. Collect the supernatant and detect the IL-6 level according to the IL-6 content detection kit instructions (see Table of Materials).
3. Separating colorectal tissue
   1. Euthanize the mice after blood sampling by inhaling 5% overdosed isoflurane and cervical dislocation (see Table of Materials) in accordance with animal ethics.
   2. Keep the mice in a cryogenic anatomical environment. Immobilize the mice in a supine position. Cut the lower abdomen hair with scissors and sterilize it with ethanol.
   3. Pinch the intersection point between the two thigh roots and the abdominal midline with eyelid forceps (see Table of Materials). Cut a transverse incision about 1-1.5 cm with scissors.
   4. Clip a longitudinal incision along the median line of the abdomen from the midpoint of the transverse incision towards the xiphoid process.
   5. Remove the peri-colorectal tissues in the direction of the anus to separate the colorectum from the surrounding tissue. Be careful not to damage the colorectum.
   6. Push the skin of the abdomen to the sides to fully expose the colorectum. Remove the colorectum from the abdominal cavity with eyelid forceps and cut off the segments from the anus to the cecum (excluding); the total length is about 10 cm. Store the obtained colorectal tissues in saline at 4 °C.
4. Evaluation of the length and weight of the rectum
   1. Extract the saline at 4 °C with a 5 mL needle (see Table of Materials) to flush the inside of the colorectum. Then, place the colorectum on absorbent paper to absorb the moisture of the tissue.
   2. Weigh the colorectal tissues and then place them on A4 paper to measure their length.
5. Measure the number of tumors in the colorectum
   1. Cut the colorectum lengthwise to fully unfold it and observe the number and size of the tumors in the colorectum.
6. Pathological analysis of the colorectum
   1. Fix the colorectum in 4% paraformaldehyde for 24 h. Embed the fixed rectal tissue in melted paraffin and section continuously with a thickness of 5 µm by a tissue freezing microtome.
   2. According to the procedure of Hou et al.²⁴, dewax the sections in xylene and then dehydrate it with serial concentrations of ethanol. After staining with hematoxylin solution for 5 min, rinse the sections with pure water. After that, stain with 0.5% eosin solution for 1 min(see Table of Materials).
   3. Perform gradient dehydration and xylene transparent treatment again. Seal the sections and observe them under the optical microscope (see Table of Materials) and photograph, as described by Xie et al.²⁵.
7. Immunohistochemical analysis of colorectum
   1. Dewax and dehydrate the sections according to the above method. Repair the antigen in the sections with high-pressure thermal repair technique, as described by Gok et al.²⁶.
   2. Soak the sections in endogenous peroxidase blockers at room temperature for 15 min. Seal the sections with goat serum (see Table of Materials).
   3. Add the primary antibody ZO-1 (1:1000), Occludin (1:1000) and KI67 (1:500; see Table of Materials) to the sections and incubate overnight at 4 °C. Rinse the sections with PBS buffer (see Table of Materials), add general purpose secondary antibody (1:5000; see Table of Materials) to it and incubate at 37 °C for 30 min.
   4. Add the DAB solution (see Table of Materials) to the sections for color development. Counterstain the sections with hematoxylin solution.
   5. Dehydrate, transparentize, and seal the sections again. Observe the expression of protein by the optical microscope.

The decoction of LJZD was prepared according to the composition ratio of drugs in Table 1 and the decoction method of TCM in Figure 1A. According to the time point indicated in Figure 1B, mice were intraperitoneally injected with 1 mg/mL AOM on the 7^th day, and mice were given free access to drinking water containing 2% DSS in the 3^rd, 6^th, and 9^th weeks. The UC-CRC mouse model was successfully established in the 15^th week. Meanwhile, mice were treated with LJZD by gavage from week 7 to week 15. The data showed that compared with the control group, the UC-CRC model group had a significant weight loss, which was alleviated by LJZD treatment (Figure 2A, P < 0.01). At the end of the trial, LJZD treatment improved the DAI score compared with the UC-CRC model group (Figure 2B). Compared with the control group, the UC-CRC model group had a shorter colorectal length, which was increased by LJZD treatment (Figure 2C,D, P < 0.01). The ratio of colorectal weight to body weight reflects the development of CRC in mice, and a higher ratio indicates acute tumor development²⁷. Compared with the control group, the colorectal organ index of the model group was significantly increased, and LJZD treatment greatly reduced the colorectal organ index (Figure 2E, P < 0.05). In addition, LJZD treatment also inhibited the formation of colorectal tumors (Figure 2F, P < 0.01) and the level of serum pro-inflammatory factor IL-6²⁸ (Figure 2G, P < 0.05).

Pathological results confirmed that compared with the control group, the mice in the UC-CRC model group had larger colorectal tumors and had formed adenocarcinoma, while LJZD treatment reduced the size and grade of the tumors (Figure 3). Immunohistochemical data demonstrated that LJZD treatment enhanced colorectal barrier function in UC-CRC model mice, as indicated by elevated protein expression of ZO-1 and Occludin²⁹ (Figure 4). In parallel, LJZD treatment suppressed the protein expression levels of the cancer marker KI67³⁰ (Figure 4).

Figure 1: Preparation of LJZD and establishment of a mouse model of coloproctitis cancer.(A) The mixture of Dangshen (12 g), Baizhu (12 g), Gancao (6 g), Chenpi (12 g) and ginger-processed Ban Xia (9 g) was immersed in 1000 mL of distilled water at room temperature for 1 h. The 12 g Fuling powder was soaked in 300 mL of distilled water at room temperature for 1 h. The mixture of the above six herbs was decocted at 100 °C for 40 min. (B) On day 7, C57BL/6J mice were intraperitoneally injected with 1 mg/mL AOM. Mice were given water containing 2% DSS ad libitum in the 3^rd, 6^th, and 9^th weeks. From 7 to 15 weeks, mice were given LJZD by gavage. Abbreviations: AOM, azomethane; DAI, disease activity index; DSS, dextran sulfate sodium; LJZD, Liujunzi Decoction; w, week. Please click here to view a larger version of this figure.

Figure 2: Evaluation of UC-CRC model and efficacy of LJZD. (A) LJZD improved body weight in UC-CRC model mice. (B) LJZD alleviated DAI scores in UC-CRC model mice. (C, D) LJZD increased the length of the colorectum in UC-CRC model mice. (E) LJZD reduced the organ index of the colorectum in UC-CRC model mice. (F) LJZD inhibited colorectal tumorigenesis in coloproctitis cancer model mice. (G) LJZD suppressed serum IL-6 level in UC-CRC model mice. ^#P< 0.05 and ^##P< 0.01, compared with the control group; ^*P < 0.05 and ^**P< 0.01, compared with the model group. Data were expressed as the mean ± standard deviations (n=10) and were analyzed by one-way analysis of variance (ANOVA) followed by Tukey's test. P < 0.05 indicated a statistically significant difference. Please click here to view a larger version of this figure.

Figure 3: Evaluation of the pathological characteristics of colorectal tissues in mice by hematoxylin-eosin staining. There was no pathological damage in the control group. The tumor tissue of the UC-CRC model group was large and had formed adenocarcinoma and high-grade neoplasia, while the LJZD treatment group had reduced tumor tissue accompanied by a small amount of local adenoma and low-grade neoplasia. The black arrow represents the precancerous dysplastic tumor gland. Please click here to view a larger version of this figure.

Figure 4: Effect of LJZD on colorectal barrier function and cancer marker in UC-CRC model mice. (A) Immunohistochemical images of ZO-1, Occludin and KI67. (B) Statistical results of protein expression of ZO-1, Occludin and KI67. LJZD treatment increased the expression of colorectal barrier functional proteins ZO-1 and Occludin, while decreased the level of cancer marker KI67 in UC-CRC model mice. ^## P < 0.01, compared with the control group; ^** P < 0.01, compared with the model group. Data were expressed as the mean ± standard deviations (n=10) and were analyzed by one-way analysis of variance (ANOVA) followed by Tukey's test. P < 0.05 indicated a statistically significant difference. Please click here to view a larger version of this figure.

Table 1: Composition and proportion of drugs in LJZD.

Table 2: Disease activity index score for UC-CRC model in mice.

CRC is one of the most common cancers worldwide, with approximately 1,148,000 new cases and more than 576,000 deaths each year. CRC can be divided into three types according to different causes, including hereditary, sporadic and UC-CRC³¹. The incidence of CRC in patients with inflammatory bowel diseases such as UC is significantly higher than that in the general population. UC stimulates the development of CRC through the inflammatory-cancer pathway, which differs from the typical adenoma-adenocarcinoma pathway⁶. At present, the cause of UC-CRC is unknown, mainly caused by long-term recurrent chronic inflammation, with a mortality rate of up to 60%^32,33. There is no effective treatment for UC-CRC, which has been the hotspot of research in recent years. Understanding the molecular mechanism of UC-CRC is crucial for early detection and precise treatment of UC-CRC.

Currently, there are many methods to construct animal models of UC-CRC, and the formation mechanisms are different. IL-10 or Muc2/4 gene knockout, resulting in epithelial barrier defects, can induce spontaneous colitis in mice, so it is often used to construct animal CRC models based on genetic defects associated with UC^34,35,36. Nevertheless, the animal model of UC-CRC induced by gene knockout cannot simulate the complete pathogenesis of the disease, and its operation method is complex, the experiment is expensive, and it has obvious limitations. Chemical induction is the classical and common method to construct UC-CRC animal models^37,38. DSS, AOM, and dimethylhydrazine are all commonly used chemical inducers for the construction of UC-CRC. However, using these chemical reagents alone to establish animal models takes a long time and has a low success rate³⁹. Studies have shown that the incidence of UC-CRC induced by the combination of AOM and DSS can reach almost 100%⁴⁰. Compared with other methods, the combination of AOM/DSS has the advantages of simple operation, strong controllability, short cycle, and high replication rate, and can better simulate human UC-CRC in terms of pathology and molecular mechanism^40,41,42. Although the use of AOM/DSS cycle stimulation to construct UC-CRC model has many advantages, it also has the disadvantages of long modeling time, expensive modeling reagents, and still cannot fully simulate the pathogenesis of human UC-CRC.

In order to seek a potential therapeutic target for UC-CRC, we used AOM, a chemical carcinogen, in combination with DSS-induced mice to establish a UC-CRC model. In this study, mice exposed to AOM and cyclic stimulation of DSS showed varying degrees of UC-CRC-related symptoms. Compared with the control group, the DAI score of the model group was significantly increased, which was manifested as weight loss, blood stool, loose stool, or watery diarrhea. Whereas LJZD significantly reduced the DAI score. Shortened colorectum length is an important marker of inflammatory response in UC-CRC^43,44. In this study, the colorectum length of the model mice was significantly shorter than that of the control group, and LJZD could improve the shortened colorectum. The colorectal mucosa develops chronic inflammation, which gradually evolves into atypical hyperplasia of colorectal tissue and eventually causes cancer⁴⁵. In this study, the number of tumors in the colorectal tissue of mice in the model group increased significantly compared with the control group, and correspondingly, the colorectal tissue weight also increased. After LJZD treatment, the number of tumors decreased significantly, and the colorectal weight decreased. It is suggested that LJZD has a significant inhibitory effect on the formation of CRC induced by UC. Previous studies have found that LJZD can alleviate inflammation and reduce the expression levels of TNF-α, IL-6, and IL-1β in vivo^16,28,46. In this study, the serum IL-6 level of mice in the model group was increased, but the production of serum IL-6 was significantly inhibited after LJZD treatment.

Microscopic observation showed inflammatory infiltration of rectal tissue, destruction of the mucosal barrier, reduction of mucus secretion, irregular or disappearance of goblet cells, distortion or atrophy of intestinal crypts, and obvious disorder of gland structure in AOM/DSS-treated mice. After LJZD treatment, the morphology of goblet cells in colorectal tissue was normal, the structure of recess and gland was regular, and the damage to the mucosal barrier was improved. Epithelial tight junction (TJ) is an indispensable mechanical barrier to maintain cell integrity and permeability, mainly including ZO-1 and Occludin²⁹. It has been reported that the expression of TJ proteins such as ZO-1 in colorectal tissue of UC-CRC patients is significantly reduced^47,48. This study showed that the colorectal mucosal tissue structure of UC-CRC mice was destroyed, and the expression of ZO-1 and Occludin in the tissue decreased, while LJZD could reverse this decrease, suggesting that LJZD may protect the integrity of the colorectal mucosal barrier by increasing the expression of ZO-1 and Occludin. Nuclear-associated antigen (KI67) is an essential component of cell cycle regulation and is widely used as an indicator of proliferative activity of tumor cells^30,49. In this study, KI67 protein expression was increased in the model group compared with the control group, indicating significant tumor cell proliferation in colorectal tissue. After LJZD treatment, the expression of KI67 protein was decreased, which showed that LJZD had a good effect on UC-CRC.

In summary, AOM intraperitoneal injection and DSS circulation stimulation can effectively establish a UC-CRC model in a short period and are similar to human UC-CRC in terms of histopathology and molecular mechanisms of formation. As a classical prescription, LJZD can reduce the DAI score, colorectal tissue weight, and inflammatory factor levels, effectively inhibit the shortening of colorectal length, and play a role in the stage of UC in UC-CRC mice. Simultaneously, it can increase the expression of TJ proteins, significantly improve the pathological damage of colorectal tissue, reduce the expression of KI67 protein, significantly reduce the incidence of tissue tumors, and inhibit the transformation process of UC to CRC. This study provides a new idea for the treatment of UC-CRC.