Article Text
PDF
Abstract
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.
- Colorectal Cancer
- Colon Cancer
- T cell
Data availability statement
The sequencing data have been deposited in the China National GeneBank DataBase, CNGBdb (CNP0005087) and (CNP0005098). These data are publicly accessible at the following links: http://db.cngb.org/cnsa/project/CNP0005087_52f53eec/reviewlink/ and http://db.cngb.org/cnsa/project/CNP0005098_ef986974/reviewlink/. Additional data can be obtained from the corresponding authors upon reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Previous studies have demonstrated a strong correlation between the development and progression of colorectal cancer and the composition of the intestinal microbiota. Imbalances in the microbiota and their metabolism can greatly influence tumor progression by triggering intestinal inflammatory responses and oxidative stress. Although some research has been conducted on the intestinal microbiota, there is still a lack of understanding regarding the effects of specific metabolites and their underlying mechanisms.
WHAT THIS STUDY ADDS
Based on data analysis of clinical samples, this study revealed significant differences in the overall composition and function of the gut microbiota and intestinal metabolites between individuals with colorectal cancer and healthy individuals. Notably, the study revealed that tyrosol levels were significantly lower in patients with colorectal cancer. Subsequent in vivo and in vitro experiments further demonstrated the effectiveness of tyrosol in inhibiting tumor progression.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
The relationship between the development of colorectal cancer and changes in the gut microbiota and metabolites was examined in this study. These results suggest that modifying the gut microbiota and its metabolites could be a potential strategy for preventing and treating colorectal cancer. Furthermore, these findings may offer valuable insights for the development of anticancer drugs.
Background
Colorectal cancer (CRC) is a leading cause of cancer-related death, accounting for 9.2% of global mortality.1 2 There is increasing evidence that CRC can developed as a result of the adoption of a Westernized diet, chronic stress, and alterations in the gut microbiota.3 The role of the gut microbiome and metabolome in CRC biology has been increasingly recognized.4 5 Previous research reports have highlighted that the tumor formation rate in germ-free mice is significantly reduced when using chemically induced mouse CRC models.6 Further research has revealed that bacteria such as Streptococcus bovis (S. bovis), Fusobacterium nucleatum, and enterotoxigenic Bacteroides fragilis (ETBF) in the intestine can contribute to the progression of CRC.7 Specifically, S. bovis proteins induce inflammatory responses and increase the risk of cancer, while F. nucleatum promotes early intestinal cancer by increasing bacterial adhesion to mucosal surfaces.8 Additionally, ETBF promotes tumor development by inducing signal transducer and activator of transcription 3 (STAT3) and Th17 responses. These findings emphasize the importance of conducting in-depth studies on the relationship between the intestinal flora and CRC to prevent and treat this disease.
The gut microbiota can use undigested dietary fibers, proteins, and peptides to produce various bioactive metabolites, which play crucial roles in human health as signaling molecules. These metabolites have dual effects on cancer and can either promote or inhibit cancer development. Among the protein fermentation products, certain compounds, such as N-nitroso compounds (NOCs), polyamines, and ammonia, exhibit carcinogenic toxicity.9 NOCs induce carcinogenesis by causing mutations through DNA alkylation. Moreover, high levels of polyamines are toxic and cause oxidative stress in the body.10 In addition to these metabolites, hydrogen sulfide, bile acid metabolism, and ethanol can also play a role in the development of CRC through various mechanisms. Although there have been numerous studies on the roles of intestinal metabolites in CRC, our understanding of the specific mechanisms by which these metabolites operate within the tumor microenvironment remains limited.
The tumor microenvironment is the ecosystem surrounding a tumor inside the body that promotes tumor development and progression. Oxidative stress is a typical feature of the tumor microenvironment.11 Reactive oxygen species (ROSs) are produced by mitochondria under oxidative stress and play an important role in tumorigenesis. They are involved in various biological processes, including cellular proliferation, evasion of apoptosis, tissue invasion, and angiogenesis, which promote tumor development and progression.12 Additionally, ROS activate nuclear factor κB (NF-κB), MAPK, and hypoxia-inducible factor 1 (HIF-1) to regulate the cellular redox status.13 HIF-1, which plays a crucial role in tumorigenesis, is overexpressed in a wide range of human cancers, such as breast cancer, ovarian cancer, oligodendroglioma, CRC, and esophageal cancer. The HIF family of transcription factors has long been acknowledged as potential cancer treatment targets. PX-478, a HIF-1α inhibitor, prevents the development of pancreatic ductal adenocarcinoma and esophageal squamous cell carcinoma,14 thereby representing a new treatment strategy.
In this study, we investigated the composition and functions of microbes and metabolites in healthy controls and patients with CRC. We observed significant differences in the overall composition and function of the gut microbiota and intestinal metabolites in individuals with CRC. Through metabolomics and correlation analysis using 16S rRNA gene sequencing, we discovered a notable decrease in tyrosol concentrations in patients with CRC. Furthermore, tyrosol inhibited the proliferation of CRC cells (CT26, HCT116, MC38) but not normal intestinal epithelial cells (NCM460). Our mechanistic analysis revealed that tyrosol suppresses colorectal tumor growth by targeting the NF-κB and HIF-1 signaling pathways. Overall, this study revealed that tyrosol, a gut microbiota-associated metabolite, can inhibit the growth of CRC, and its mechanism of action was elucidated. These findings may provide new insights into potential cancer treatment strategies.
Methods
Human samples
A set of 19 stool samples from patients with CRC and healthy donors were included in this work. All patients with CRC had no prior history of radiochemotherapy. Additionally, none of the participants had received antibiotics before enrolment. Fecal samples were immediately flash frozen in liquid nitrogen following collection and stored at −80°C before sequencing.
Animals and cell lines
BALB/c mice (male, 18–20 g; 4–6 weeks) and C57BL/6 mice (male, 18–20 g; 4–6 weeks) were purchased from Liaoning Changsheng Technology Industrial (Liaoning, China). The experimental procedures were conducted according to the guidelines of the Animal Welfare and Research Ethics Committee. Mammalian cell lines, including FHC, NCM460, CT26 and CT26-Luc, MC38, HT29, HCT-116, HCT-8, and SW480, were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The cells were cultured in DMEM (BI, Israel), RPMI 1640 (BI, Israel), or DMEM/F12 (Sigma‒Aldrich) supplemented with 10% fetal bovine serum (FBS, Bioind, Israel) and incubated at 37°C with 5% CO2.
16S rRNA gene sequencing and bacterial community analysis
Fecal samples from humans (n=38, 19 samples from each group of the healthy and CRC groups) were extracted from the bacterial total genomic DNA by the CTAB method. 16S rRNA genes in distinct V4 region was amplified with specific primer 16S V4: 515F-806R and barcodes. All PCR reactions contained Phusion High-Fidelity PCR Master Mix (New England Biolabs). The PCR products were mixed in equal proportions, and then the Qiagen Gel Extraction Kit (Qiagen, Germany) was used to purify the mixed PCR products. The constructed library was quantified by Qubit and Q-PCR. After qualified library, the NovaSeq6000 was used for PE250 on-machine sequencing. To enhance the accuracy and reliability of the information analysis results, the original data were initially spliced and filtered to yield valid data (Clean Data). Subsequently, DADA2 was employed for noise reduction based on the valid data,15 resulting in the final amplicon sequence variant (ASV).16 Species annotation was conducted on the representative sequences of each ASV to obtain corresponding species information and species-based abundance distribution. Concurrently, ASV abundance, alpha diversity, and Venn diagrams were analyzed to extract information regarding species richness and evenness within samples, as well as data on common and unique ASVs among different samples or populations. Furthermore, multiple sequence alignments were performed on ASVs, and a phylogenetic tree was constructed to investigate the differences in community structure among various samples or groups through dimensionality reduction analysis. To further examine the differences in community structure between grouped samples, statistical methods such as the T-test, ALDEx2, and ANCOM were used to evaluate the significance of differences in species composition and the variations between grouped samples.
Untargeted metabolomics
Metabolites extracted from the discovery cohort were injected into the Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS) system analysis. Ultra High Performance Liquid Chromatography-Tandem Mass Spectrometry (UHPLC-MS/MS) analyses were performed using a Vanquish UHPLC system (ThermoFisher, Germany) coupled with an Orbitrap Q Exactive HF mass spectrometer (Thermo Fisher, Germany) in Novogene (Beijing, China). The raw data files generated by UHPLC-MS/MS were processed using Compound Discoverer 3.1 (CD3.1, ThermoFisher) to perform peak alignment, peak picking, and quantitation for each metabolite. After that, peak intensities were normalized to the total spectral intensity. The normalized data were used to predict the molecular formula based on additive ions, molecular ion peaks, and fragment ions. And then peaks were matched with the mzCloud (https://www.mzcloud.org/), mzVaultand MassList database to obtain accurate qualitative and relative quantitative results. Statistical analyses were performed using the statistical software R (R V.R-3.4.3), Python (Python V.2.7.6 version), and CentOS (CentOS release V.6.6). Metabolites were annotated using the KEGG database (https://www.genome.jp/kegg/pathway.html), HMDB database (https://hmdb.ca/metabolites), and LIPIDMaps database (http://www.lipidmaps.org/).
Tumor models
Healthy male C57BL/6 and BALB/c mice were used to establish subcutaneous and orthotopic tumor models, respectively. To induce subcutaneous tumor models, 5×106 MC38 cells were injected subcutaneously into the right flanks of the mice. The tumor length (L) and width (W) were recorded every other day, and the tumor volume was calculated using the following formula: tumor volume (cm3)= (L×W2)/2. For orthotopic tumor models, 1×106 CT26-Luc cells were surgically injected into the cecum. Tumor size was detected through in vivo imaging technology using the Tanon Imaging System (Tanon 4600, Tanon, China).
CCK-8 assay
To assess cell viability, we used a Cell Counting Kit-8 (Meilun, Dalian, China) following the manufacturer’s instructions. In brief, 5000 cells in 100 µL of medium were seeded into each well of a 96-well plate. After 24 hours, 10 µL of CCK8 solution was added to each well, and the plates were incubated for 1 hour at 37°C. Finally, the absorbance at 450 nm was measured to calculate the cell viability.
Cytokine assays
The expression levels of the inflammatory factors TNF-α, IL-1β, and IL-6 in cell culture supernatant samples were quantified with specific ELISA kits (BioLegend, California, USA) according to the manufacturer’s instructions.
Tyrosol assays
Tyrosol content in feces and serum was quantified using a specific ELISA kit (Jingmei, Jiangsu, China), following the manufacturer’s instructions.
Western blot analysis
MC38 CRC cells were harvested and treated with RIPA lysis buffer containing fluoride (Beyotime, Jiangsu, China). Then, the total protein concentration was measured by a BCA assay (Fisher Scientific, Pittsburgh, Pennsylvania, USA). Next, the proteins were separated by SDS‒PAGE and transferred to polyvinylidene fluoride (PVDF) membranes. The PVDF membranes were incubated with specific primary antibodies, including HIF-1α, HIF-1β (Affinity Biosciences, Ohio, USA), p-p65, p65, p-IκB, IκB, and β-actin (Immnoway Biotechnology, USA), followed by incubation with secondary anti-IgG antibodies conjugated with HRP (Immnoway Biotechnology, USA). The proteins were detected using an enhanced chemiluminescence plus western blotting detection system (Tanon, China).
Immunohistochemistry
The tumor tissues were processed into paraffin-embedded sections and subjected to immunohistochemical staining. In brief, the paraffin-embedded sections were deparaffinized, subjected to antigen retrieval, blocked, incubated with a primary antibody against the HIF-1α protein, incubated with a secondary antibody, and finally subjected to DAB staining. The sections were then scanned using a slice scanner and panoramic scanner scanning software, with a focus on collecting composite images. Optical density values were analyzed using ImageJ software.
ROS assay
ROSs were measured using an ROS assay kit (JianCheng, Nanjing, China) following the manufacturer’s instructions. Briefly, CRC cells were cultured in a 96-well plate for 24 hours, and then 10 µM DCFH-DA was added for 30 min. Finally, the fluorescence was measured using a fluorescence microplate reader with an excitation wavelength of 488 nm and an emission wavelength of 525 nm.
Superoxide dismutase activity assay
To measure superoxide dismutase (SOD) enzyme activity in tumor cells, we used the Total Superoxide Dismutase Assay Kit with WST-8 (Beyotime, Jiangsu, China) following the manufacturer’s protocol.
Flow cytometry
Quantitative analysis of immune cells infiltrating the blood and tumors was performed using flow cytometry. Fresh tumors were cut into small pieces and homogenized to form single-cell suspensions. The cells were collected, washed, and then incubated with various fluorescein-conjugated antibodies, such as CD45, CD3, CD4, CD8, CD25, and Foxp3, which were purchased from BioLegend (San Diego, California, USA). After two washes, the cell suspension was filtered through a 400-mesh filter and analyzed with a flow cytometer. The data were obtained using a FACSAria III cytometer (BD Bioscience, San Jose, California, USA) and analyzed with FlowJo V.10 software.
Statistical analysis
The statistical analysis was performed using GraphPad Prism V.6.0. The error bars represent the SEM. Statistical significance was determined using Student’s t-test or one-way analysis of variance. Values of p<0.05 were considered to indicate statistical significance. Other specific statistical analyses are described in the “Methods” section.
Results
The human gut microbiome is altered in patients with CRC
We evaluated fecal samples collected from patients with CRC, the TF group, and healthy individuals, the CF group. Alterations in the overall composition of the gut microbiome were identified in patients with CRC. The diversity analysis (Alpha diversity) of a single sample reflects the richness and diversity of the microbial community present. The Chao index, which employs the Chao1 algorithm, estimates the number of species in the sample, thereby indicating community richness. The results demonstrated that the Chao1 value for patients with CRC was significantly higher than that of healthy individuals, suggesting a greater presence of low-abundance species in the fecal community of patients with CRC and an overall increase in richness. Pielou’s evenness index reflects the uniformity of the community. Results indicated that the index for the CF group was significantly higher than that for the TF group, suggesting that the bacterial community in the CF group exhibits greater uniformity. The Shannon index assesses both the total number of taxa in the sample and their relative proportions, while the Simpson index evaluates diversity and evenness in species distribution by calculating the probability that two randomly selected individuals belong to different species. Both indices provide insights into community diversity. The findings revealed that the Simpson index for the CF group was significantly higher than that for the TF group, whereas the difference in the Shannon index was not statistically significant, indicating an increase in community diversity in the feces of patients with CRC (figure 1A). To demonstrate that higher species richness is not attributable to increased sampling depth, this experiment constructs rarefaction curves with the amount of extracted data plotted on the X-axis and the Alpha diversity index value on the Y-axis. The results indicate that the curve approaches a plateau, signifying that the amount of sequencing data is becoming adequate. Consequently, additional data will not significantly influence the alpha diversity index or affect the experimental outcomes (figure 1B). The β diversity as determined by principal components analysis (PCoA), significantly differed between the CF and TF groups according to the weighted UniFrac distance (figure 1C). The classify-sklearn algorithm of QIIME2 was employed for species annotation using the Naive Bayes classifier for each ASV. From the ASV annotation results and the characteristic table of each sample, a species abundance table was generated (online supplemental table S1). Based on the species annotation results at various classification levels, a relative abundance column chart was created. T-tests were conducted to identify species exhibiting significant differences between groups at the phylum, genus, and species taxonomic levels (online supplemental figure S1–3). Following this analysis, a volcano plot was generated. Additionally, the ALDEx2 (online supplemental tables S2–4) and ANCOM-BC (online supplemental tables S5–7) methods were employed to further validate the composition of the microbiome data at the phylum, genus, and species levels, as well as to construct an ANCOM scatter plot. Furthermore, detection of the bacterial phyla revealed that the relative abundance of Bacteroidota was significantly decreased, while the relative abundance of Firmicutes C was markedly increased in the feces of healthy individuals (figure 1D). At the genus level, the microbiota composition differed significantly between the TF and CF groups. The TF group had an increased relative abundance of dominant bacteria from Porphyromonas _A_859424, and Fusobacterium_C, whereas the relative abundance of bacteria from Dialister, Oliverpabstia, and Fimenecus was increased in the CF group (figure 1D). We found that Bacteroides_H fragilis were overrepresented in the TF group. However, Faecalibacterium prausnitzii_C_71358 was more abundant in the CF group (figure 1F). These findings indicate that patients with CRC exhibit distinct fecal microbiota compositions.
Supplemental material
Supplemental material
Supplemental material
Supplemental material
Supplemental material
Supplemental material
Supplemental material
Supplemental material
Patients with colorectal cancer have distinct gut microbiota compositions. (A) The alpha diversity in terms of chao 1, pielou_e, Simpson index, and Shannon index. (B) The rarefaction Curve in terms of chao 1, pielou_e, Simpson index, and Shannon index. (C) Principal co-ordinates analysis based on the weighted UniFrac distance. (D) Changes in the gut bacterial community at the phylum level. (E) Changes in the gut bacterial community at the genus level. (F) Changes in the gut bacterial community at the species level.Data are presented as the mean±SEM. *p<0.05; ***p<0.001.
Faecalibacterium prausnitzii inhibits tumor growth by suppressing inflammation and increasing the proportion of CD8+ T cells in the blood and tumor tissue
This study focused on the potential benefits of Faecalibacterium prausnitzii in the prevention of CRC. MC-38 mouse CRC cells were injected subcutaneously into the right flanks of the mice to establish an ectopic model (figure 2A). Compared with the control, supplementation with Faecalibacterium prausnitzii resulted in a significant reduction in tumor growth (figure 2B). Subsequently, the impact of microbial depletion prior to Faecalibacterium prausnitzii treatment was investigated. Mice were treated with antibiotics 3 days before Faecalibacterium prausnitzii administration (figure 2C). The results showed that antibiotic treatment effectively inhibited tumor growth, with a more pronounced effect observed in the presence of Faecalibacterium prausnitzii (figure 2D). To investigate how Faecalibacterium prausnitzii inhibits CRC growth in mice, we initially assessed the levels of the inflammatory factors TNF-α, IL-1β, and IL-6 in the serum of mice. Our results showed that the administration of Faecalibacterium prausnitzii led to a significant reduction in the levels of these inflammatory factors (figure 2E). To further elucidate the impact of Faecalibacterium prausnitzii on tumor immunity in mice, we conducted a flow cytometry analysis of CD8+T cells in the blood of mice and CD8+T cells in tumor tissues. Our findings revealed that treatment with Faecalibacterium prausnitzii resulted in an increase in the proportion of CD8+T cells in the blood (figure 2F). Moreover, there was a notable increase in the proportion of CD8+T cells within tumor tissues following treatment (figure 2G).
Faecalibacterium prausnitzii inhibits tumor growth by regulating tumor immunity. (A) Experimental setting of the approach of Faecalibacterium prausnitzii in the CRC model. (B) The antitumor activity of Faecalibacterium prausnitzii in the subcutaneous tumor model. Tumor volume and tumor weight represent excised tumor at day 16. (C) Experimental setting of the Faecium prevotella approach in a CRC model after antibiotic treatment. (D) The antitumor activity of Faecalibacterium prausnitzii in the subcutaneous tumor model after antibiotic treatment. (E) Levels of proinflammatory cytokines, including IL-1β, IL-6, and TNF-α. (F) Dot plots showing the CD8+ T cells in the blood by flow cytometry. (G) The CD8+ T cells within tumor tissues by flow cytometry. CRC, colorectal cancer. FP, Faecalibacterium prausnitzii group. Data are presented as the mean±SEM. *p<0.05; **p<0.01;****p<0.0001.
Integrated microbiome-metabolome analysis reveals the link between the microbiome-derived metabolite tyrosol and CRC
To identify gut microbial metabolites that may play a role in CRC, untargeted metabolomic profiling was performed in the discovery cohort. The results demonstrated distinct metabolite profiles in patients with CRC (TMs) and controls (CMs). The Pearson correlation coefficient was computed to evaluate the correlation between QC samples, indicating a strong correlation among the QC samples and high accuracy of the entire detection process (online supplemental figure S4). The calculation of the quantitative value of non-targeted metabolism involved integrating the mass spectrum peak corresponding to each metabolite to determine the area under the peak curve. The ordinate (AU) of the mass spectrum represents the relative intensity, while the abscissa (min) indicates the retention time. Each metabolite’s mass spectrum peak can be modeled as a Gaussian function. For non-targeted metabolism, CD software is employed to calculate the area under the Gaussian function curve, yielding the peak area value. In the multivariate statistical analysis, the metaX software is used to perform principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) on the data. PCA is used to observe the distribution within and between sample groups, whereas PLS-DA is applied to obtain the variable importance in projection (VIP) values for the metabolites. In the univariate analysis, the data are log2 transformed, followed by a two-sided Student’s t-test to assess the statistical significance (p value) of each metabolite between the two groups. The fold difference of each metabolite between the two groups is calculated as the fold change (FC) value. Based on the criteria of VIP>1, FC>1.5 or FC<0.667, and p<0.05, the abundance of 422 metabolites was found to increase, while the abundance of 414 metabolites decreased. Correlation heatmaps were generated to identify clusters of metabolites and to further investigate changes in metabolite abundance (online supplemental table S8). The metabolites were used to generate a correlation heatmap to identify metabolites that clustered together and further investigate changes in metabolite abundance (online supplemental figure S5). PLS-DA is a supervised discriminant analysis statistical method that employs partial least squares regression to establish a relationship model between metabolite expression and sample categories for the purpose of predicting these categories. The results indicated a significant separation between TMs and CMs (figure 3A). To assess the quality of the model, the grouping labels of each sample were randomly shuffled prior to modeling and prediction, allowing for sorting and validation of the model. Each modeling iteration is associated with a set of R² and Q² values, which were used to evaluate whether the model exhibits signs of overfitting. The findings revealed that the R² values exceed the Q² values, and the intercept of the Q² regression line with the Y-axis was less than zero. Consequently, the model was determined not to be overfitted and was deemed effective in accurately describing the samples (online supplemental figure S6). A volcano plot revealed a significant increase in the abundance of adrenic acid and arachidonic acid, while the abundance of neohesperidin and naringenin decreased significantly in TMs group compared with CMs (figure 3B). Moreover, functional enrichment analysis revealed significantly enriched KEGG pathways related to the biosynthesis of unsaturated fatty acids (figure 3C). Alterations in microbiome composition in patients with abnormal colorectal conditions contribute to the rewiring of local metabolome profiles. To further determine the association between the gut microbiome and abnormal colorectal-associated fecal metabolites, a microbiome-metabolomics analysis was used in this study. Given that the metabolomics and 16S data approximate a normal distribution, we can preliminarily assess the correlation between bacterial flora and intestinal metabolites by calculating the Pearson correlation coefficient. We conducted Pearson correlation analysis between the significantly differentially abundant bacterial groups identified through 16S rRNA sequencing and the significantly differentially abundant metabolites obtained from metabolomic analysis. The Pearson correlation coefficient (r) and the p value (|r|>0.8 and p<0.05) were calculated to generate a heat map (figure 3D). The results revealed a positive correlation between adrenic acid and pathogenic bacteria such as Mediterraneibacter_A_155507 torques and Bacteroides_H fragilis, whereas a negative correlation was observed with the beneficial bacteria Faecalibacterium prausnitzii_C_71358. Moreover, tyrosol was positively correlated with Faecalibacterium prausnitzii_C_71358, and tyrosol levels were significantly elevated in the healthy group. Concurrently, ELISA was used to measure the tyrosol levels in the serum and feces. Following the administration of Faecalibacterium prausnitzii, a notable increase in tyrosol levels was observed in both the serum and feces of CRC model mice compared with control mice (figure 3E). Tyrosol, the primary component of olive oil, is predominantly sourced from food, which suggests that there is a correlation between diet and tyrosol levels. The findings of a subsequent experiment revealed a significant increase in tyrosol level in the feces and serum of mice following the oral administration of olive oil. Moreover, coadministration of Faecalibacterium prausnitzii with olive oil further elevated tyrosol levels in the feces and serum of mice (figure 3F).
Supplemental material
Metabolic profiles of health and colorectal cancer patients. (A) Partial least squares discrimination analysis, PLS-DA score plots for gut samples. (B) Volcano plot for differential metabolites in comparison between CM and TM. (C) Bubble map of KEGG enrichment. Abscissa is a p value, and smaller p values are statistically more significant. The ordinate is the KEGG pathway. The size of the bubbles in the figure represents enriched metabolic concentrations. (D) Metabome-16s association analysis Heatmap of Correlation Analysis. The relative abundance of each differential microbiota and the correlation coefficient r and p values between different differential metabolites were calculated using the Pearson statistical method. The heatmap of the results indicates the significance level with * marking for p values ≤0.05. (E) Tyrosol content in serum and feces after stimulation with Faecalibacterium prausnitzii. (F) Tyrosol content in serum and feces of mice in EVOO tumor model. PLS-DA, partial least squares discriminant analysis.Data are presented as the mean±SEM. **p<0.01; ***p<0.001; ****p<0.0001. FP, Faecalibacterium prausnitzii group. EVOO, extra virgin olive oil group. CM, control group fecal metabolomics. TM, tumor group fecal metabolomics.
Tyrosol reduces inflammation and oxidative stress by inhibiting the NF-κB and HIF-1 signaling pathways
The CCK-8 assay revealed that tyrosol greatly decreased the viability of various human CRC cells (HCT116, HCT-8, SW480, and HT29) as well as mouse CRC cells (MC38 and CT26). However, no significant effect on the viability of normal intestinal epithelial cells, NCM460 cells, or FHC cells was observed (figure 4A–C). We evaluated the cytotoxicity of tyrosol on normal liver cells (LO2) and kidney cells (HK2) and observed that tyrosol had no significant effect on either HK-2 or LO-2 cells following stimulation with tyrosol (online supplemental figure S7). Colony formation experiments revealed that tyrosol decreased the clonogenic capacity of CRC cells (figure 4D). Moreover, tyrosol strongly inhibited the colony formation of both human tumor cells (HCT116) and mouse tumor cells (MC38). ELISA was used to measure the levels of inflammatory cytokines in cell culture supernatant samples, and the tyrosol group exhibited significantly lower expression levels of TNF-α, IL-1β, and IL-6 than the control group (figure 4E). Additionally, compared with the control group, the tyrosol group displayed a notable decrease in ROS levels and a significant increase in SOD levels (figure 4F). To further elucidate the mechanisms underlying the anti-inflammatory and antioxidant effects of tyrosol, we investigated its impact on the NF-κB and HIF-1 signaling pathways. Our findings demonstrated that tyrosol effectively decreased the protein expression of p-p65 and p-IκB in MC38 cells, leading to significant suppression of the NF-κB pathway. Additionally, tyrosol reduced the protein levels of HIF-1α and HIF-1β (figure 4G), thereby suppressing the HIF-1 signaling pathway. These results suggest that tyrosol exerts anti-inflammatory and antioxidant effects on tumor cells by targeting the NF-κB and HIF-1 signaling pathways.
Tyrosol inhibits CRC cell viability and reduces colony formation. (A) No effect of tyrosol on normal intestinal epithelium. (B and C) Inhibitors of murine and human colorectal cancer growth by tyrosol. (D) Effect of the tyrosol in the colony formation of MC38 and HCT116. (E) Levels of proinflammatory cytokines, including IL-1β, IL-6, and TNF-α in in cell culture supernatant samples. (F) ROS levels and SOD levels of MC38 and HCT116 were measured. (G) Representative protein levels of p-p65, p-IκB, HIF-1α, and HIF-1β and intensity analysis of HIF-1α, HIF-1β, p-p65, and p-IκB. The values presented are the means±SEM of three independent experiments. An asterisk is placed above differences with a significance exceeding *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001. CRC, colorectal cancer.
Tyrosol reduces tumor growth in both subcutaneous and orthotopic CRC models
To evaluate the impact of tyrosol on antitumor activity, we established a subcutaneous CRC xenograft model in vivo. Briefly, 3 days following MC38 cell inoculation, C57BL/6 mice were intraperitoneally injected with 10 mg/kg tyrosol every day. As the treatment duration increased, the tumor volume in the tyrosol group decreased compared with that in the control group (figure 5A), with the most significant difference observed on the final day of the experiment (figure 5B). After the mice were sacrificed, the mean tumor weight of the control group was 206 mg, while that of the control group was 162 mg, indicating that tyrosol inhibited tumor growth (figure 5C). Next, a subcutaneous tumor model was established in immunodeficient BALB/c nude mice by injecting the human CRC cell line HCT116. Compared with those in the control group, the tumors in the tyrosol-treated group were significantly smaller (online supplemental figure S8). To further recapitulate the tumor microenvironment, we generated an orthotopic CRC model. Tumor growth was observed using the Tanon Imaging System in vivo imaging system with firefly D-luciferin substrate injection until sacrifice (figure 5D). The results showed that the tumor volume and tumor weight of the mice in the tyrosol group were significantly greater than those in the control group (figure 5E). Tyrosol is the predominant compound found in extra virgin olive oil. An olive oil mouse animal model was created to investigate its effects (figure 5F). The results showed that intragastric administration of extra virgin olive oil can significantly inhibit tumor growth. Furthermore, when Faecalibacterium prausnitzii was concurrently administered, the tumor inhibition effect was further enhanced (figure 5G).
Effect of tyrosol on tumor growth in both the subcutaneous tumor model and the orthotopic CRC model. (A) The antitumor activity of tyrosol in the subcutaneous tumor model. (B) Tumor volume and tumor weight (C) represent excised tumor at day 18. (D) In vivo bioluminescent images in the orthotopic CRC model. (E) Representative photographs and tumor weight of mice in different groups. (F) Experimental setting of the approach of EVOO in the CRC model. (G) The antitumor activity of Faecalibacterium prausnitzii in the subcutaneous tumor model after EVOO treatment. In this figure, the data shown are average±SEM. An asterisk is placed above differences with a significance exceeding *p<0.05; ∗∗p<0.01; ∗∗∗p<0.001. CRC, colorectal cancer. FP, Faecalibacterium prausnitzii group. EVOO, extra virgin olive oil group.
Tyrosol enhances tumor immunity and inhibits NF-κB/HIF-1 pathways
In a subcutaneous tumor mouse model, tyrosol treatment led to an increase in the proportion of peripheral blood CD8+T cells (figure 6A). Flow cytometry analysis of the tumor tissue revealed an increase in the proportion of intratumoral CD8+T cells in the tyrosol-treated group (figure 6B). These results suggest that tyrosol may have systemic antitumor effects by increasing the infiltration of CD8+T cells into tumors. The progression of CRC is intimately linked to the NF-κB and HIF-1 signaling pathways. Western blotting was used to detect protein expression levels. The results showed that the levels of p-p65, p-IκB, HIF-1α, and HIF-1β were significantly lower in the tyrosol group than in the control group (figure 6C–E). To further investigate the impact of tyrosol on the HIF-1 signaling pathway in tumor tissues, we conducted immunohistochemistry experiments. The findings demonstrated that tyrosol effectively decreased the protein expression of HIF-1α in tumor tissues (figure 6F), thereby impeding the progression of CRC in mice.
Tyrosol enhances tumor immunity and inhibits NF-κB/HIF-1 pathways. (A) Dot plots showing the CD8+ T cells in the blood by flow cytometry. (B) The CD8+ T cells within tumor tissues by flow cytometry. (C) Representative protein levels of p-p65, p-IκB, HIF-1α, and HIF-1β. (D) The intensity analysis of HIF-1α, HIF-1β, p-p65, and p-IκB. (F) Immunohistochemistry staining assessed the HIF-1α level in tumor tissues (scale bar, 50 μm). The positive-stained areas are shown in brown. In this figure, the data shown are average±SEM. An asterisk is placed above differences with a significance exceeding ∗∗p<0.01; ∗∗∗p< 0.001; ****p<0.0001.
Discussion
CRC, a prevalent malignant tumor affecting the digestive tract, is the second-leading cause of cancer-related deaths worldwide.17 Researchers have discovered a correlation between the composition of the intestinal microbiota and the risk of developing CRC. The microbiota consists of specific bacterial species that play critical roles in the formation of colon tumors.18 The results of our research support the findings of previous studies. Based on an analysis of fecal samples, we observed a significant difference in the gut microbiota diversity between patients with CRC and healthy individuals. There was a notable increase in the relative abundance of Bacteroides_H fragilis in the intestinal microbiota of patients with CRC. Moreover, the abundance of Porphyromonas _A_859424 in the gut was found to be significantly greater in patients with CRC, suggesting that this genus could be used to diagnose CRC.19 In addition to the previously mentioned cancer-causing microorganisms, we also found that the abundance of the beneficial microorganism Bacteroides_H fragilis was significantly reduced in the guts of patients with CRC. In previous studies, Faecalibacterium prausnitzii has been suggested as a probiotic supplement that may help prevent and manage CRC.20 We discovered that intragastric administration of Faecalibacterium prausnitzii effectively suppressed the progression of CRC in tumor model mice. This finding underscores the important role that Faecalibacterium prausnitzii plays in the development of CRC.
In addition to the direct effects of micro-organisms, metabolic modification of the intestinal tract has also been noted as a contributor to the formation of tumors. These metabolites have both carcinogenic and anticancer effects.21 Through KEGG pathway enrichment analysis, we found that the biosynthesis of unsaturated fatty acids was primarily enriched in the differentially metabolized products. Numerous studies have demonstrated that fatty acids can directly bind to tumor cell membranes, significantly altering the cellular fatty acid composition. Polyunsaturated fatty acids can modulate gene expression, regulate cell proliferation and differentiation, induce apoptosis, and promote ROS production and lipid peroxidation, thereby impacting tumor growth.22 In our study, we observed a significant increase in the abundance of harmful fatty acid metabolites, such as adrenic acid, arachidonic acid, and myristic acid, in the guts of patients with CRC, while the abundance of the beneficial metabolite sinapinic acid significantly decreased. Adrenic acid affects the sensitivity of gastric cancer cells to iron-induced cell death and can induce oxidative stress in liver cells.23 24 Moreover, adrenic acid and its metabolites regulate inflammatory responses and various cellular processes, including proliferation, survival, angiogenesis, invasion, and metastasis, thereby promoting carcinogenesis. Similarly, myristic acid also increases the risk of cancer occurrence. Conversely, research has demonstrated that sinapinic acid effectively inhibits tumor growth. According to previous reports, sinapinic acid inhibits the proliferation of colon cancer cells in a dose-dependent manner.25 Therefore, we speculate that in the healthy population, the gut metabolite sinapinic acid may play a preventive role in tumor occurrence, while high abundance of the gut metabolites adrenic acid, arachidonic acid, and myristic acid increases the risk of cancer by acting on oxidative stress-related pathways. Our study reveals significant differences in polyphenol metabolites between healthy individuals and patients with CRC. While there are variations in metabolites among patients with CRC that may be attributed to personal dietary habits, age, and other factors, the differences observed between the two groups are more pronounced than those within the healthy population. Notably, we observed significantly elevated levels of neohesperidin and naringenin in the gut of healthy individuals. Both metabolites demonstrate potent anti-inflammatory, antiproliferative, and tumor-suppressive properties in the colon. Neohesperidin, in particular, exhibits antioxidant and anti-inflammatory effects, positioning it as a promising therapeutic option for various cancers.26 27 Dysregulation of the gut microbiota and alterations in metabolic profiles have been implicated in the pathogenesis of CRC. The results of our correlation analysis indicated a positive correlation between adrenergic acid and the Mediterranean pathogenic bacteria Bacteroides a_155507 torques and Bacteroides_H fragilis. Previous studies have shown that the abundance of Mediterraneibacter torques increases in melanoma patients, which impacts the therapeutic efficacy of immune checkpoint inhibition.28 Our findings also revealed an elevation of this bacterium in patients with CRC, suggesting its potential role in promoting CRC progression through the modulation of adrenic acid metabolism. Furthermore, another study reported significant increases in both Bacteroides fragilis and adrenic acid in the intestines of patients with CRC,29 which aligns with our observations. Notably, the abundance of the metabolite tyrosol, positively associated with these beneficial bacteria, is significantly increased in a healthy gut and may play a crucial role in tumor growth.
Diet is considered one of the factors that contribute to the development of CRC, as it affects the composition and metabolism of the intestinal flora.30 Numerous studies have demonstrated a clear inverse relationship between the Mediterranean diet and the risk of CRC.31 A key component of the Mediterranean diet is extra virgin olive oil, which plays an important role in preventing CRC. In mouse tumor models, intragastric administration of EVOO effectively suppressed tumor growth. Furthermore, co-supplementation with Faecalibacterium prausnitzii was found to enhance the inhibitory effect of EVOO on tumors. The level of tyrosol, a metabolite of olive oil, was increased in the feces and serum following EVOO consumption and a more robust increase was observed in the presence of Faecalibacterium prausnitzii. Remarkably, the antitumor effects of tyrosol were evident when tyrosol was administered intraperitoneally. Experimental findings suggest that the presence of tyrosol is influenced by diet and the gut microbiota and is crucial for the inhibition of CRC growth.
Tyrosol has various biological activities, including antioxidant, anti-inflammatory, and anticancer effects. Research has demonstrated that most tumor cells have an imbalanced intracellular redox system, and long-term high-level oxidative stress promotes tumor growth.32 33 ROSs are important oxidative products that can cause DNA damage and genomic instability, contributing to the accumulation of carcinogens.34 In vitro, the addition of tyrosol significantly reduced ROS levels in mouse CRC cells. This finding suggested that tyrosol may inhibit factors that promote tumor cell proliferation. Moreover, studies have indicated that chronic intestinal inflammation can lead CRC.35 In our experiments, tyrosol reduced the expression of TNF-α, IL-6, and IL-1β. As a result, tyrosol may suppress tumor development by reducing ROS levels and the expression of inflammatory cytokines.
Modulating the immune response is a critical factor in determining the course of cancer.36 The balance between regulatory and effector cells in the immune system is crucial for an immune response against tumors. This balance is maintained through the interplay between effector cells and regulatory T cells. Disruption of this balance can impact tumor progression, as studies have shown that low CD8+T cell numbers and high Treg cell numbers are linked to a poor prognosis.37 Our research demonstrated that both intragastric administration of Faecalibacterium prausnitzii and intraperitoneal injection of tyrosol can increase CD8+T cell numbers in the blood, thereby enhancing the antitumor immune response. Furthermore, examination of tumor tissues revealed increased CD8+T cell infiltration following treatment with Faecalibacterium prausnitzii or tyrosol, leading to an enhanced immune response within the tumor microenvironment. Consequently, Faecalibacterium prausnitzii and tyrosol may play a role in preventing CRC by modulating tumor immunity.
Hypoxia-inducible factor 1 is significantly upregulated in various human cancer types. It induces the expression of several genes associated with cancer cell growth and survival, angiogenesis, metastasis, cancer metabolism, maintenance of cancer stem cells, and resistance to various cancer treatment methods. Moreover, HIF-1α can promote cancer cell proliferation and survival by interacting with P53 and P21cip1.38 Our experiments demonstrated a significant decrease in the expression of HIF-1α and HIF-1β in MC38 cells treated with tyrosol. This reduction could be one factor contributing to the inhibitory effect of tyrosol on CRC. In addition, HIF can also trigger the activation of NF-κB, which is a nuclear transcription factor believed to be involved in multiple protumorigenic processes in cancer.39 Furthermore, it increases the expression of vascular endothelial growth factor and its receptors, thus stimulating tumor vascularization.40 Our research indicated that tyrosol significantly inhibited the expression of NF-κB signaling pathway components in mouse tumors. This finding implies that the anti-CRC effects of tyrosol can be attributed to its ability to inhibit HIF/NF-κB signaling pathway activation.
In conclusion, the occurrence of CRC is closely linked to the composition of the intestinal flora and its metabolites. In patients with CRC, the intestinal flora is imbalanced and characterized by an increased abundance of harmful bacteria and a decreased abundance of beneficial bacteria. Faecalibacterium prausnitzii has been shown to inhibit tumor growth by reducing inflammatory responses and increasing the number of effector cells, ultimately improving tumor immunity. Furthermore, there are significant alterations in the metabolites present in the intestine. These metabolic changes primarily affect the biosynthesis of unsaturated fatty acids. Additionally, correlation analysis revealed that tyrosol plays a crucial role in this process. Tyrosol can interfere with the HIF-1/NF-κB signaling pathway, thereby reducing the release of inflammatory factors, alleviating oxidative stress, and ultimately inhibiting the occurrence and progression of CRC.
Data availability statement
The sequencing data have been deposited in the China National GeneBank DataBase, CNGBdb (CNP0005087) and (CNP0005098). These data are publicly accessible at the following links: http://db.cngb.org/cnsa/project/CNP0005087_52f53eec/reviewlink/ and http://db.cngb.org/cnsa/project/CNP0005098_ef986974/reviewlink/. Additional data can be obtained from the corresponding authors upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
Clinical samples obtained at The First Bethune Hospital of Jilin University were approved by the Ethical Committee, and informed consent was obtained from all subjects (2022-KS-037). All mice were approved by the Animal Ethics Committee of the First Hospital of Jilin University (SYXK (JI) 2019-0012).
Acknowledgments
This work was supported by a grant from the Interdisciplinary Integration and Innovation Project of JLU (JLUXKJC2021ZY08) and Jilin Provincial Healthcare Talent Special Program (JLSWSRCZX2023). We are grateful to Department of Colorectal&anal surgery, General Surgery Center, First Hospital of Jilin University for partial support of this work. We appreciate the Beijing Novogene for their service of the assessment of metabolites.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
JG, FM and RH contributed equally.
Contributors JG contributed to article writing, literature search, and results evaluation. FM and RH performed histological analysis and article revision. HR, JC, LC and KZ performed the final revision of the article and expert opinions. PH and ZL contributed to the literature search and results evaluation. JT conducted the final revision of the article, evaluated the results, and served as the guarantor of the work. GW contributed to the study design.
Funding Interdisciplinary Integration and Innovation Project of JLU (JLUXKJC2021ZY08) and Jilin Provincial Healthcare Talent Special Program (JLSWSRCZX2023).
Competing interests No, there are no competing interests.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.