Morphofunctional changes in the immune system in colitis-associated colorectal cancer in tolerant and susceptible to hypoxia mice

Experimental animals

The study was made on male C57Bl/6 mice (n = 60) age 1.5–2 months, body weight 20–25 g, taken from the Branch of the Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry of the Russian Academy of Sciences (BIBCh), Russia. Mice were housed ten per 42.5 × 27.6 × 15.3 cm cage with a footprint of 820 cm2 (Tecniplast, Italy), which are designed to hold up to 11 mice weighing 20–25 g or less, at regulated room temperature 25 °C ± 2 °C under 12:12 h light–dark cycle and 40–50% relative humidity with unlimited access to water and food (Char; JSC Range-Agro, Turakovo, Russia). All efforts were made to decrease suffering and possible stress for the animals and performed in accordance with the European Convention for the Protection of Vertebrate Animals Used for Experimental Use (ETS 123, Strasbourg, 1986), and Directives of the European Parliament and the Council of the European Union (2010/63/EU, Strasbourg, 2010). The study was approved by the Bioethics Committee at the Avtsyn Research Institute of Human Morphology (Protocol No. 36/12, March 28, 2022). All procedures were in accordance with the ‘Animal Research Reporting of In Vivo Experiments’ (ARRIVE) guidelines and the AVMA euthanasia guidelines 2020. According to literature, hypoxia tolerance depends on animal’s sex. It was demonstrated that hypoxia tolerant organisms predominantly were detected among females, whereas susceptible and normal organisms predominantly among males, and there were sex differences in the morphofunctional state of the immune system (Kosyreva et al., 2020). To equalize sex differences, this study was performed only on male C57Bl/6 mice. Mice were randomly divided into the following experimental groups with a minimum five animals each. The study was conducted in two biological repeats. The first experiment was conducted in conjunction with a previously published study of gene expression changes in CAC (Dzhalilova et al., 2024a).

Determination of hypoxia tolerance

The animal’s tolerance to hypoxia was determined once in a mercury barometer-coupled decompression chamber, by simulated hypobaric hypoxia equivalent to 10,000 m altitude. Time till the first sign of characteristic hyperventilatory response (‘gasping time’) was recorded using the electronic stopwatch (Kirova, Germanova & Lukyanova, 2013; Jain et al., 2013; Jain et al., 2014; Dzhalilova et al., 2018a; Pavlik et al., 2018; Mironova et al., 2019). Based on the hypoxia tolerance determination results, mice were divided into 2 groups—tolerant (n = 17) with ‘gasping time’ was more than 10 min, and susceptible (n = 18)—gasping time was less than 3 min. Mice that were normally resistant (n = 25) to hypoxia (‘gasping time’ from 3 to 10 min) were not used in the experiments. Test time was between 08.30 am to 12.00 pm and testing order was randomized daily. For each animal, different researchers were involved as follows: a first researcher (DD) determined the hypoxia tolerance and was responsible for the cervical dislocation. This researcher was the only person aware of the group allocation. The second and the third researchers (MS, AM) performed the flow cytometry, morphological and morphometric study.

Colitis-associated colorectal cancer model

For CAC modeling single AOM (A5486-25MG, Sigma Azoxymethan) intraperitoneal injection at a dose of 10 mg/kg (Deng et al., 2022) was performed a month after hypoxia resistance test among tolerant (n = 12) and susceptible (n = 13) to hypoxia mice. A week after the AOM injection, drinking water was replaced with 1% DSS (Dextran sulfate sodium salt, M_r ∼ 40,000, AppliChem, Germany) for 7 days. After 14 days of water drinking, it was again replaced with 0.5% DSS for 7 days, then after 14 days of water drinking—for 5 days with 0.5% DSS (Tanimura et al., 2021). Clinical symptoms were assessed by the presence of diarrhea and blood in the feces of mice. Clinical manifestations occurred after each DSS consumption cycle, and then the animal’s condition returned to normal. If throughout the experiment the mice tended to have a pronounced pain posture or/and continuous diarrhea and blood in the feces, and in cases when the tumor size reaches a maximum dimension of two cm in mice then this was the criteria established for euthanizing animals prior to the planned end. However, during the described investigation, there were no mice that were euthanized prior to the planned end of the experiment. No animal death was observed during the experiment. Cervical dislocation was used for the euthanizing from the experiment by 86 days after the last drinking water replacement with DSS, that is, on the 141st day of the experiment. Physiological solution (0.9% NaCl; Armavirskaya Biofabrika, Russia) was injected intraperitoneally in control groups of tolerant (n = 5) and susceptible (n = 5) to hypoxia mice, then mice consumed drinking water for 140 days, then they were euthanized from the experiment. There were no mice survived after 140 days of the experiment.

Morphological and morphometric study

Data were collected as previously described in Dzhalilova et al. (2024a). Specifically, colon was divided into proximal, medial and distal parts, buffered 10% formalin was used for fixation during 24 h. The tumor size did not reach a maximum dimension of 20 mm in mice. Then colon parts were put in paraffin, stepped-serial histological sections were prepared (from eight to 12 sections) with a 200 μm step, and with hematoxylin and eosin staining. In the distal colon tumors area assessment was made on stepped-serial histological sections, in which tumors were detected. They were photographed at a magnification 50x using an Axioplan 2 Imaging microscope (Carl Zeiss, Oberkochen, Germany). The tumor borders outlining was performed with Adobe Photoshop CS6. Eventually, isolated tumors area was revealed with the Image-Pro Plus 6.0 program, and for each mouse their area was calculated.

A morphological study of the liver and lungs was performed on sections stained with hematoxylin and eosin to assess the presence of metastases. To assess the response of the immune system organs, a morphological study of the thymus, spleen and MLNs was performed. The mice lungs were fixed in Carnoy’s solution (60 ml ethanol, 30 ml chloroform, and 10 ml glacial acetic acid) for 2 h, the liver, thymus and spleen were fixed in Bouin’s solution (75 ml picric acid, 25 ml formalin, and five ml glacial acetic acid) and MLNs were fixed in buffered 10% formalin for 24 h, and organs were embedded in paraffin according to routine procedures. Histological sections of 4–5 μm thickness were produced and stained with hematoxylin and eosin (BioVitrum). In the histological slices of the thymus and spleen, the volume fraction of the functional zones was determined by the point-count method.

Immunohistochemical study

Immunohistochemical detection of Ki-67 was carried out by sandwich method. Colon sections were deparaffinized, citrate buffer pH 6.0 with 0.5% Tween-20 at 100 °C was used for the antigen retrieval and blocked in phosphate-buffered saline with 0.1% bovine serum albumin at room temperature before exposure to antibodies (primary: Rabbit Monoclonal Anti-Ki-67, ab16667, Abcam, Cambridge, UK; secondary: Goat Anti-Rabbit IgG H&L Alexa Fluor 488 ab150077, Abcam). After PBS cleansing, DAB work solution was amplified until color changing visualized. Digital images were photographed using an Axioplan 2 Imaging microscope (Carl Zeiss).

Flow cytometry

Blood was taken from the jugular veins, and the absolute and relative number of lymphocytes was determined using the automatic hematology analyzer Mindray BC-2800Vet (China). Via flow cytometry with a Cytomics FC 500 (Beckman Coulter), the subpopulation composition of peripheral blood cells and MLNs was determined in mice of the control and experimental groups. Peripheral blood from the jugular veins was collected into tubes containing EDTA as an anticoagulant. Isolation of cells from MLNs was performed with Potter homogenizer (Winter et al., 2019). To analyze the cells, antibodies conjugated with FITC (fluorescein isothiocyanate), PE (phycoerythrin), PE-Cy5 and PE-Cy7, to CD19+ (B-lymphocytes), CD3e+CD4+ (T-helper cells), CD3e+CD8+ (cytotoxic T-lymphocytes), CD4+CD25+Foxp3+ (regulatory T-lymphocytes), NK-1.1 (NK cells), CD11b+ (monocytes), F4/80 (macrophages) from eBioscience were used. Lymphocytes and monocytes were selected based on their morphology using forward-versus side-scatter (FSC-SSC) dotplots because the samples were not fixed, erythrocytes were lysed, no more than 4–6 h passed after obtaining the materials and before the study on the flow cytometer, the cells were constantly at room temperature. Combining anti-CD4 and anti-CD8, we identified CD3e+CD8+ and CD3e+CD4+ T cells, regulatory T-lymphocytes were identified on CD4+ T cells by CD25+Foxp3+ expression, monocytes were gated by CD11b+ expression. For flow cytometry data analysis FlowJo software was used (Becton, Dickinson & Company, USA).

Statistical analysis

The data obtained were statistically processed with GraphPad Prism 8.0. Kolmogorov–Smirnov criterion was used for the type of the indicators distribution determination. In case the data were not distributed normally, the differences between indicators significance were determined via the nonparametric Mann–Whitney, Kruskal-Wallis, and Dunn tests. Data were represented as median and interquartile range Me (LQ(25%); UQ(75%)). Fisher’s test was used to assess the statistical significance of clinical symptoms. Differences were considered statistically significant at p < 0.05.

CAC development frequency and tumor morphological features

During clinical observation of mice that received a single dose of AOM and three cycles of DSS, both tolerant and susceptible to hypoxia mice experienced diarrhea and the blood detected in the feces, which was associated with the development of ulcerative colitis and CAC (Fig. 1A). No differences were found when analyzing the statistical significance of the symptoms by Fisher’s test (p = 0.3).

During morphological study in the distal colon tumors were detected in 41.6% of the tolerant (five of 12) and in 84.6% of the susceptible (11 of 13) to hypoxia mice, in the medial colon—in 8.3% of the tolerant (one of 12) and in 7.7% of susceptible (one out of 13), no tumors were detected in the proximal colon of both tolerant and susceptible mice. In the distal colon, tumors were represented by glandular intraepithelial neoplasia (80% of all detected tumors in tolerant, 0% in susceptible), as well as adenocarcinomas (20% of all detected tumors in tolerant, 100% in susceptible), their frequency development was higher (p = 0.003) in susceptible mice (Fig. 1A). Colon histological examination in both tolerant and susceptible to hypoxia mice revealed chronic colitis: single epithelized ulcers were detected in the mucous membrane, the crypts were deformed, their lumens were enlarged, and crypt abscesses were noted. Inflammatory infiltration of lymphocytes, macrophages and plasma cells was observed, mostly pronounced in the basal part of the mucous membrane lamina propria. Glandular intraepithelial neoplasia was characterized by clusters of several crypts lined with hyperchromatic polymorphic epithelial cells with dilated and deformed lumens (Fig. 1B). Exophytic growth of adenocarcinomas was observed; microscopically, adenocarcinomas were represented by many glands lined with proliferating atypical columnar epithelium; crypt abscesses were detected in some crypts - lumens of the crypts were dilated and filled with mucus with a large neutrophils number. The tumor’s stroma was represented by connective tissue with partially ordered thin fibers; diffuse connective tissue infiltration was noted with a small number of lymphocytes, macrophages and focal neutrophils (Fig. 1C).

According to the morphometric study, the tumor area in the distal colon in susceptible to hypoxia mice was statistically significantly higher (p = 0.002) in comparison to tolerant (Fig. 1D). Morphological study of the liver, lungs and MLNs of both tolerant and susceptible to hypoxia mice did not reveal metastases. For results of morphometric immune system organs study and flow cytometry analysis, were used data obtained only from mice with confirmed tumor, i.e., with glandular intraepithelial neoplasia and adenocarcinomas of tolerant (n = 5) and susceptible (n = 11) to hypoxia mice.

Tumor immunohistochemical study

During all cell cycle phases except G0 the Ki-67 antigen is expressed and therefore it is commonly used as proliferation marker (Gerdes et al., 1984). In control groups of tolerant and susceptible to hypoxia mice there were single Ki-67+ cells in the colon mucosa. During CAC, glandular intraepithelial neoplasia and adenocarcinoma both demonstrated an increase in the expression of Ki-67 staining compared with the control groups. More pronounced Ki-67+ staining in tumors were in adenocarcinomas in susceptible to hypoxia mice (Fig. S1).

Thymus, spleen and MLNs morphological and morphometric study during CAC

At morphological examination in the thymus of control group mice, the cortex was represented by densely arranged lymphocytes and epithelial cells (Figs. 2A, 2B). The boundaries between the cortex and medulla were clear. The thymus subcapsular zone was represented by 4–8 rows of lymphoblasts. The thymus medulla was represented by evenly distributed epithelial cells and lymphocytes. Thymic bodies in the form of cyst-like cavities, as well as those consisting of 3–5 epithelial cells and single ones with small keratohyalin deposits were detected in it.

The morphological study in the thymus of tolerant to hypoxia mice during CAC in comparison to the control group revealed the cortex widening, the boundaries between the cortex and medulla were focally unclear. Thymic bodies consisted of 3–5 cells, more than five cells, numerous thymic bodies with keratohyalin and in the form of cyst-like cavities were found (Fig. 2C). In susceptible to hypoxia mice, focal devastation of the cortex and focally unclear boundaries between the cortex and medulla were revealed. Thymic bodies consisted of 3–5 cells, more than five cells, a few thymic bodies with keratohyalin and in the form of cyst-like cavities were detected (Fig. 2D).

The thymus morphometric study in mice tolerant and susceptible to hypoxia in the control groups did not reveal any differences. During CAC, cortex widening and medulla narrowing were observed in tolerant to hypoxia mice in comparison to the control group, while in the susceptible to hypoxia mice no differences were detected. In tolerant to hypoxia mice of the experimental group in comparison to the susceptible, the volume fraction of the cortex was higher, and the medulla was lower (Fig. 3). When studying thymic bodies of various types, an increase in the number of thymic bodies in the form of cyst-like cavities was found only in tolerant to hypoxia mice on the 141st day of the experiment in comparison to the control group (Table 1).

In the spleen morphological study during CAC, lymphoid follicles with and without small germinal centers were detected in both tolerant and susceptible to hypoxia mice; single mitoses were detected in them. The PALS (the periarterial lymphatic sheaths) zone was moderately and weakly expressed. The number of megakaryocytes was increased (Fig. 4).

In the spleen morphometric analysis in tolerant to hypoxia control group mice, the lymphoid follicles volume fraction was larger, and the PALS zones were smaller compared to susceptible. During CAC, in both the tolerant and susceptible to hypoxia mice, an increase in the lymphoid follicles volume fraction and a decrease in the proportion of the PALS zone relative to the white pulp were demonstrated, while the lymphoid follicles germinal centers expansion was observed only in susceptible to hypoxia mice (Fig. 5).

The morphological study of MLNs in the control groups of tolerant and susceptible mice did not reveal any differences, lymphoid follicles with and without germinal centers; in the germinal centers, cells were loosely located, represented by lymphoblasts, lymphocytes, macrophages, and single dying cells. The paracortical zone is focally devastated, the marginal sinuses were not dilated, they contained lymphocytes and macrophages. Intermediate sinuses were not pronounced. In the cerebral sinus the ratio of lymphocytes and macrophages was 1:1. The pulp strands stroma was populated by densely located mature lymphocytes and macrophages (Fig. 6).

In the morphological study of MLNs during CAC, in both tolerant and susceptible to hypoxia mice, the outer cortex and paracortical zone were expanded and densely populated with lymphocytes. Lymphoid follicles without and with germinal centers were revealed. There were loose and densely located lymphoblasts and lymphocytes, fragments of nuclei in the germinal centers. A large number of plasma cells, as well as lymphocytes and macrophages, were detected in the medullary cords. The marginal and medullary sinuses were dilated, there were many lymphocytes and macrophages in them. Macrophages with granular eosinophilic contents, single foreign-body giant cell were detected. Thus, in both tolerant and susceptible to hypoxia adult male C57Bl/6 mice on the 141st day after AOM administration, according to a morphological study of MLNs, B- and T-zones hyperplasia were revealed. A large number of plasma cells were detected in the medullary cords.

Blood and MLNs during CAC flow cytometry

In the control group, the absolute and relative number of lymphocytes was higher in susceptible to hypoxia mice in comparison with tolerant. On the 141st day of the experiment, only in susceptible to hypoxia mice there was a statistically significant decrease in the relative content of lymphocytes (Table 2). It was demonstrated that in the control group that the absolute number of NK cells and CD11b+ monocytes in the blood was higher in susceptible to hypoxia in comparison with tolerant mice. In the experimental groups of both tolerant and susceptible to hypoxia mice, the absolute number of NK cells and T-regulatory lymphocytes and the relative number of CD3e+CD4+ T-lymphocytes in the blood decreased in comparison to the control groups, while the relative number of CD11b+ monocytes increased. However, only in susceptible to hypoxia mice the relative and absolute numbers of CD19+ and relative - CD3e+CD8+ lymphocytes in blood increased. The relative number of CD3e+CD8+ and NK cells in the peripheral blood during CAC was higher in susceptible to hypoxia in comparison to tolerant mice.

In the control group of susceptible to hypoxia mice in the MLNs, the absolute CD3e+CD4+ T-lymphocytes number was lower compared to tolerant. At the same time, on the 141st day of the experiment, in comparison with the control group, no changes were observed in the mice tolerant to hypoxia, and in the susceptible to hypoxia mice the relative and absolute CD19+ and NK cells number, the absolute CD3e+CD8+ lymphocytes number increased, and F4/80+ macrophages relative number decreased (Figs. 7 and 8). The relative and absolute numbers of CD19+ and NK cells in MLNs in CAC were higher in susceptible mice in comparison to tolerant (Figs. 7 and 8).

The summary results are presented in Figs. 9. The identified differences indicated the more pronounced reaction of immune cells to the CAC development in susceptible to hypoxia mice at both the local and systemic levels.