Cholera-toxin suppresses carcinogenesis in a mouse model of inflammation-driven sporadic colon cancer

Abstract

Human studies and clues from animal models have provided important links between gastrointestinal (GI) tract bacteria and colon cancer. Gut microbiota antigenic stimuli play an important role in shaping the intestinal immune responses. Therefore, especially in the case of inflammation-associated colon cancer, gut bacteria antigens may affect tumorigenesis. The present study aimed to investigate the effects of the oral administration of a bacterial product with known immunomodulatory properties on inflammation-driven colorectal neoplasmatogenesis. For that, we used cholera-toxin and a well-established mouse model of colon cancer in which neoplasia is initiated by a single dose of the genotoxic agent azoxymethane (AOM) and subsequently promoted by inflammation caused by the colitogenic substance dextran sodium sulfate (DSS). We found that a single, low, non-pathogenic dose of CT, given orally at the beginning of each DSS treatment cycle downregulated neutrophils and upregulated regulatory T-cells and IL-10 in the colonic mucosa. The CT-induced disruption of the tumor-promoting character of DSS-induced inflammation led to the reduction of the AOM-initiated colonic polypoidogenesis. This result adds value to the emerging notion that certain GI tract bacteria or their products affect the immune system and render the microenvironment of preneoplastic lesions less favorable for promoting their evolution to cancer.

Figure 1.

CT reduces AOM/DSS-induced colorectal adenomatous polyp formation. (a) Gross pathology of the descending colon and rectum of selected mice (five animals per group). Grossly visible polyps are indicated with arrow-heads. Treatment with CT reduced both the multiplicity and occurrence of polyps. (b) Close-up macroscopical view of multiple polyps in the distal part of descending colon and rectum of an AOM-DSS-treated mouse. The normal colorectal mucosa of an AOM-DSS-CT-treated mouse is shown on the side for comparison. (c) The beneficial effect of CT is statistically significant. (d) Low power magnification microscopical overview of a typical AOM/DSS broad-based, non-pedunculated adenomatous polyp with closely packed tubular abnormal glands (tubular adenoma) and occasional cystic glands filled with mucin. (e) Histopathological grades of dysplasia/adenoma lesions contained in polyps. The progression from low grade dysplasia/adenoma to high grade dysplasia/adenoma and CIS is characterized by progressively increased morphological abnormalities such as glandular shape and size irregularities, epithelial pseudostratification, cellular atypia, nuclear pleomorphism and mitotic figures. (c) Numbers on the y-axis of bar graph correspond to the mean ± SEM of histological parameters assessed; *P < 0.05. (d) Hematoxylin and Eosin. Scale bars: 1500 μm (c) and 50 μm (d).

Figure 2.

Dysplasia and inflammatory cells at 3.5 months after DSS treatment. (a) Focal dysplasia in the non-neoplastic colonic mucosa of AOM-DSS-treated mice. A mouse that received no further treatment has HGD lesions (black arrows), while its non-treated counterpart on the right has mild (white arrow-head) and LGD (white arrow). The effect of CT on reducing the total number and severity of these dysplastic lesions is statistically significant as shown in the dysplasia score bar graph. (b) While their colon architecture is by large restored and colitis absent, the mice which sustained colitogenic insults show more resident immune cells in the colonic mucosa lamina propria compared to their treatment-matched controls. The total population of these cells, however, was comparable between CT-treated and non-treated mice. (c) Morphometric counts of immunohistochemically labeled neutrophils (MPO+) provide evidence that CT contributed significantly toward a more effective long-term regression of this inflammatory cell type to the normal colon mucosa base-line level. (d) At the same time, Treg (Foxp3+) are significantly upregulated due to CT treatment in both MLN (compare side-by-side images) and the colonic lamina propria. (a) and (b) Hematoxylin and Eosin. (c) and (d) IHC; diaminobenzidine chromogen, hematoxylin counterstain. Scale bars: 50 μm (a and b); 25 μm (c and d). Numbers on the y-axis of bar graphs correspond to the mean ± SEM of the histological parameters assessed; *P < 0.05, **P < 0.001, ***P < 0.0001. MPO+ cells were counted in ×20 and Foxp3+ cells in ×40 high power magnification images.

Figure 3.

Colitis and dysplasia at 3 days after DSS treatment. (a) DSS-induced focal mucosal ulcerative lesions with neutrophil-rich granulation tissue formation were evident in both AOM-DSS and AOM-DSS-CT mice. Both groups were given similar histopathological scores for colitis. (b) Dysplastic glands showed typical IHC features including highly increased cellular proliferation (ki-67), increased epithelial cell apoptosis (caspase-3) and cytoplasmic stabilization of β-catenin. (c) Focal dysplasia (outlined by arrows) was found in comparable extent and severity grade in both groups of AOM/DSS-treated mice. (a) and (c) Hematoxylin and Eosin. (b) IHC; diaminobenzidine chromogen, hematoxylin counterstain. Scale bars: 100 μm (a, Ki-67 and β-catenin of b; AOM-DSS-CT of c); 50 μm (caspase-3 of b), 250 μm (AOM-DSS of c). The y-axis depicts the mean ± SEM of histological scores given in each experimental group; NS: P > 0.05.

Figure 4.

CT downregulates neutrophils and CD3+ lymphocytes at 3 days after DSS treatment. (a) MPO+ cells (neutrophils) in the lamina propria of the colon. The acute inflammation cell component is significantly reduced with CT administration. (b) Likewise, CD3+ cells are significantly decreased in the ulcerative lesions (upper left panel), the mucosal lymphoid follicles (upper right panel) and, also, in the colonic mucosa lamina propria. (a) and (b) IHC; diaminobenzidine chromogen, hematoxylin counterstain. Scale bars: 50 μm. The y-axis of bar graphs depicts the mean ± SEM of IHC-labeled cell counts in ×20 high power magnification images. *P < 0.05, **P < 0.001, ***P < 0.0001.

Figure 5.

CT upregulates Treg at 3 days after DSS treatment. In both (a) granulation tissue of focal typical DSS-induced colonic mucosa lesions and (b) MLN, Foxp3+ cell numbers are higher in AOM/DSS-treated mice that also received CT compared with the mice that received no further treatment. (c) Foxp3+ cell morphometric counts in these two sites as well as in colonic mucosa lamina propria and lymphoid follicles yield a consistent statistical significance for this effect of CT. (a) and (b) IHC; diaminobenzidine chromogen, hematoxylin counterstain. Scale bars: 25 μm. (c) Numbers on the y-axis of bar graphs correspond to the mean ± SEM of Foxp3+ cells counted in ×40 high power magnification images. *P < 0.05, **P < 0.001, ***P < 0.0001.

Figure 6.

CT upregulates the expression of cytokines at 3 days after DSS treatment. (a) Quantitative cytokine gene expression analysis of mouse colon shows that IL-10, TNF-a, IL-6 and Tgf-β are in higher levels after treatment with CT. Statistical significant differences between experimental groups are reached for IL-10 and TNF-a. In situ detection of (b) IL-6+ cells in ulcerative lesions (primarily neutrophils) and (c) IL-17 in MLN. Morphometric counts reveal a CT-induced increase for both cytokines. Statistical significance, however, is not reached. (b) and (b) IHC; diaminobenzidine chromogen, hematoxylin counterstain. Scale bars: 25 μm. (b); 50 μm (c). Numbers on the y-axis of bar graphs correspond to the mean ± SEM of IL6+ cells counted in ×40 high power or IL-17+ image pixels counted in ×20 high power magnification images. *P < 0.05, **P < 0.001, ***P < 0.0001.