Effects of metronidazole on colorectal cancer occurrence and colorectal cancer liver metastases by regulating Fusobacterium nucleatum in mice

Abstract

Objective: Colorectal cancer (CRC) represents a leading cause of cancer-related deaths. Metronidazole (MNZ) is exceedingly implicated in CRC. This study explored the roles of MNZ in mouse CRC occurrence and liver metastasis (CRLM).

Methods: Male BALB/c nude mice were subjected to CRC and CRLM modeling, orally administration with MNZ (1 g/L) 1 week before modeling, and disease activity index (DAI) evaluation. Fresh stool and anal swab samples were collected on the morning of the 28th day after modeling. The relative expression of Fusobacterium nucleatum (F. nucleatum) DNA was assessed by quantitative polymerase chain reaction. After euthanasia, tumor tissues and liver tissues were separated and the tumor volume and weight change were measured. The liver tissues were stained with hematoxylin-eosin to quantitatively analyze the metastatic liver nodules. Malignant tumor biomarker Ki67 protein levels in liver tissues/DNA from stool samples were detected by immunohistochemistry/high-throughput 16S rRNA gene sequencing. Bioinformatics analysis was performed on the raw sequence data to analyze microbial community richness (Chao1 index, ACE index) and microbial community diversity (Shannon index).

Results: The DAI and F. nucleatum DNA relative expression in feces and anal swabs of the CRC and CRLM groups were raised and repressed after MNZ intervention. MNZ repressed tumor occurrence and growth in mice to a certain extent, alleviated CRLM malignant degree (reduced liver metastases and Ki67-positive cell density/number), and suppressed CRC liver metastasis by regulating intestinal flora structure, which affected the intestinal characteristic flora of CRC and CRLM mice.

Conclusion: MNZ suppressed CRC occurrence and CRLM in mice by regulating intestinal F. nucleatum.

Keywords: CT26 cells; DNA relative expression; F. nucleatum; colorectal cancer; colorectal cancer liver metastases; gut microbiota; metronidazole.

Figure 1

Effects of MNZ on the abundance of F. nucleatum during liver metastasis in CRC mice. (A) DAI of mice of each group was appraised. (B) The relative expression of F. nucleatum DNA in feces and anal swab samples was assessed by qPCR. N = 6. The data were expressed as mean ± standard deviation. One‐way ANOVA was adopted for data comparisons among multiple groups, followed by Tukey's multiple comparisons test. **p < .01, ***p < .001. CRC, colorectal cancer; DAI, disease activity index; MNZ, metronidazole; qPCR, quantitative polymerase chain reaction.

Figure 2

MNZ inhibited tumor occurrence and growth in mice to a certain extent. (A) Tumor body diagram of mice in each group; (B) The changes in tumor volume in nude mice were detected; (C) Tumor weight was measured. N = 6. The data were expressed as mean ± standard deviation. One‐way repeated measures ANOVA (B) or independent sample t‐test (C) were employed for data comparisons between two groups. MNZ, metronidazole.

Figure 3

MNZ alleviated the malignant degree of CRC liver metastasis. (A) The liver tissues were stained with H&E and the liver metastatic nodules were quantitatively analyzed, and metastatic nodules were indicated by black arrows; (B) The expression of Ki67‐positive cells in liver tissues was assessed by IHC. N = 6. The data were expressed as mean ± standard deviation. Independent sample t‐test was employed for data comparison between two groups. *p < .05. CRC, colorectal cancer; H&E, hematoxylin–eosin; IHC, immunohistochemistry; MNZ, metronidazole.

Figure 4

MNZ repressed CRC liver metastasis by regulating intestinal flora structure in mice. (A) Shannon index curve; (B) sample dilution curve; (C) species accumulation curve at genus level; (D) Chao1 index; (E) ACE index; (F) Shannon index; (G) principal component analysis (PCA): each point represented a sample, and different colors represented different samples/groups. The abscissa represented the first principal component, the percentage represented the contribution value of the first principal component to the sample difference, the ordinate represented the second principal component, and the percentage represented the contribution value of the second principal component to the sample difference. The two samples were closer when the composition of the two samples were more similar; (H) principal coordinates analysis (PCoA) showed that the samples were closer on the coordinate map when they had a greater similarity; (I) for nonmetric multidimensional scaling (NMDS) analysis, the points in the figure represented each sample, different colors represented different groups, and the distance between points represented the degree of difference. When the stress was less than 0.2, it indicated certain reliability of the NMDS analysis. The samples were closer in the coordinate map when they had the higher similarity. (A) represented the Sham group, (B) represented the CRC group, (C) represented the CRC + MNZ group, (D) represented the CRLM group, and (E) represented the CRLM + MNZ group. CRC, colorectal cancer; CRLM, CRC liver metastases; MNZ, metronidazole.

Figure 5

Effect of MNZ on characteristic intestinal microflora of CRC and CRLM mice. Relative abundance changes of intestinal characteristic flora of mice in each group at the level of (A) phylum; (B) family; (C) genus. The abscissa was the sample name (A–E represented the Sham group, CRC group, CRC + MNZ group, CRLM group, and CRLM + MNZ group, respectively); the ordinate was the relative abundance percentage. Different colors represented different species; the stacking column was a taxon with relative abundance of top 10 at each taxonomic level. MNZ, metronidazole; CRC, colorectal cancer; CRLM, CRC liver metastases.