Background: The development and progression of colorectal cancer (CRC) are influenced by the gut environment, much of which is modulated by microbial-derived metabolites. Although some research has been conducted on the gut microbiota, there have been limited empirical investigations on the role of the microbial-derived metabolites in CRC. Methods: In this study, we used LC-MS and 16S rRNA sequencing to identify gut microbiome-associated fecal metabolites in patients with CRC and healthy controls. Moreover, we examined the effects of Faecalibacterium prausnitzii and tyrosol on CRC by establishing orthotopic and subcutaneous tumor mouse models. Additionally, we conducted in vitro experiments to investigate the mechanism through which tyrosol inhibits tumor cell growth. Results: Our study revealed changes in the gut microbiome and metabolome that are linked to CRC. We observed that Faecalibacterium prausnitzii, a bacterium known for its multiple anti-CRC properties, is significantly more abundant in the intestines of healthy individuals than in those of individuals with CRC. In mouse tumor models, our study illustrated that Faecalibacterium prausnitzii has the ability to inhibit tumor growth by reducing inflammatory responses and enhancing tumor immunity. Additionally, research investigating the relationship between CRC-associated features and microbe–metabolite interactions revealed a correlation between Faecalibacterium prausnitzii and tyrosol, both of which are less abundant in the intestines of tumor patients. Tyrosol demonstrated antitumor activity in vivo and specifically targeted CRC cells without affecting intestinal epithelial cells in cell experiments. Moreover, tyrosol treatment effectively reduced the levels of reactive oxygen species (ROS) and inflammatory cytokines in MC38 cells. Western blot analysis further revealed that tyrosol inhibited the activation of the NF-κB and HIF-1 signaling pathways. Conclusions: This study investigated the relationship between CRC development and changes in the gut microbiota and microbial-derived metabolites. Specifically, the intestinal metabolite tyrosol exhibits antitumor effects by inhibiting HIF-1α/NF-κB signaling pathway activation, leading to a reduction in the levels of ROS and inflammatory factors. These findings indicate that manipulating the gut microbiota and its metabolites could be a promising approach for preventing and treating CRC and could provide insights for the development of anticancer drugs.

Introduction: Accurate classification of colorectal histopathology images is challenging because diagnostically relevant patterns are distributed across both fine-grained cellular textures and broader tissue architecture. Conventional handcrafted descriptors often lack sufficient discriminative capacity, while purely spatial deep models do not explicitly exploit frequency-domain information. Methods: We propose FGA-Net, a Fourier-Guided Attention Network for feature enhancement in colorectal cancer digital pathology images. The model combines patch embedding, transformer-style encoding, and a Fourier-guided Attention for feature Refinement (FAR) module integrating FFT/IFFT transformation, squeeze-and-excitation recalibration, and self-attention. Experiments were conducted on the public EBHI dataset under the binary benign-versus-malignant setting at 200× magnification. Results: FGA-Net consistently outperformed five handcrafted feature families across seven downstream classifiers. The best configuration, FGA-Net with ANN, achieved an accuracy of 87.54%, surpassing the strongest handcrafted baseline by 11.52 percentage points. Ablation studies further showed that both the spectral branch and the channel recalibration branch contributed positively to performance. Discussion: The results indicate that Fourier-guided attention effectively improves pathology feature representation by jointly modeling spectral cues and long-range spatial dependencies. FGA-Net provides a more discriminative and transferable representation for colorectal histopathology classification and offers a promising direction for future computational pathology research.