Characteristics of the gut microbiome profile in obese patients with colorectal cancer

Abstract

Background and aim: Obesity affects the gut microbiome, which in turn increases the risk for colorectal cancer. Several studies have shown the mechanisms by which some bacteria may influence the development of colorectal cancer; however, gut microbiome characteristics in obese patients with colorectal cancer remain unclear. Therefore, this study evaluated their gut microbiome profile and its relationship with metabolic markers.

Methods: The study assessed fecal samples from 36 consecutive patients with colorectal cancer and 38 controls without colorectal cancer. To identify microbiotic variations between patients with colorectal cancer and controls, as well as between nonobese and obese individuals, 16S rRNA gene amplicon sequencing was performed.

Results: Principal coordinate analysis showed significant differences in the overall structure of the microbiome among the study groups. The α-diversity, assessed by the Chao1 index or Shannon index, was higher in patients with colorectal cancer versus controls. The relative abundance of the genera Enterococcus, Capnocytophaga, and Polaribacter was significantly altered in obese patients with colorectal cancer, whose serum low-density lipoprotein concentrations were positively correlated with the abundance of the genus Enterococcus; among the most abundant species was Enterococcus faecalis, observed at lower levels in obese versus nonobese patients.

Conclusions: This study demonstrated several compositional alterations of the gut microbiome in patients with colorectal cancer and showed that a reduced presence of E. faecalis may be associated with obesity-related colorectal cancer development. The gut microbiome may provide novel insights into the potential mechanisms in obesity-related colorectal carcinogenesis.

Keywords: Capnocytophaga; colorectal neoplasms; enterococcus; gastrointestinal microbiome; obesity.

Figure 1

Principal coordinate analysis to assess the beta diversity based on OTU abundance at the genus levels. Unweighted UniFrac distances were calculated between the patients with CRC (cross) and the controls (circle) (a), or between obese (pink) and nonobese participants (blue) in controls (b) or patients with CRC (c), or between obese patients with CRC and obese controls (d). Microbial structural differences were observed in each of the comparisons. Controls, individuals without colorectal cancer; CRC, colorectal cancer; OTU, operational taxonomic unit; PC, principal component. , Non‐obese control; , obese control; , non‐obese CRC; , obese CRC.

Figure 2

Alpha diversity in the patients with CRC and the controls with or without obesity. The observed Shannon and Chao1 phylogenetic diversity indices were calculated by the “vegan” version 2.5‐6 package of R version 3.6.1. The line inside the box represents the median, while the whiskers represent the lowest and highest values within the 1.5 IQR (Table S3). No significant differences were observed in the median Shannon index among the groups (a, b). The median Chao1 index was significantly higher in patients with CRC versus controls (c) in either case of nonobese or obese participants (d; Kruskal–Wallis test, P = 0.02; Steel–Dwass test showed no significant differences among the groups). No significant intergroup differences were noted in the Choa1 index between nonobese and obese controls or patients with CRC. **P < 0.01. CRC, colorectal cancer; IQR, internal quartile range.

Figure 3

Phylum‐level microbial composition. The phylum‐level taxonomic composition of all participants (a). Comparison between the patients with CRC and the controls; the median relative abundances of the phyla Actinobacteria (P = 0.03), Fusobacteria (P < 0.001), and Tenericutes (P < 0.001) were significantly higher, whereas no significant differences were observed in the median relative abundance of the phyla Firmicutes (P = 0.42), Bacteroidetes (P = 0.16), Proteobacteria (P = 0.77), and Cyanobacteria (P = 0.21). *P < 0.05, **P < 0.01. CRC, colorectal cancer. , Firmicutes; , Bacteroidetes; , Actinobacteria; , Proteobacteria; , Fusobacteria; , Tenericutes; , Cyanobacteria; , others.

Figure 4

Heatmap visualizing the Z‐score distribution of the genera in patients with colorectal cancer (CRC) and the controls. The Z‐scores were used to assess the observation's deviation of microbiome components from the mean values. The proportion of 38 genera (violet‐red) was significantly higher in the patients with CRC than in the controls: Lactobacillus, Fusobacterium, Lysinibacillus, Bulleidia, Collinsella, Enterococcus, Streptococcus, Anaerotruncus, Tepidimonas, Staphylococcus, Dermatophilus, Gemella, Polaribacter, Bacillus, Corynebacterium, Campylobacter, Capnocytophaga, Sedimentibacter, Polynucleobacter, Propionibacterium, Saccharomonospora, Paludibacter, Ruminococcus, Gallibacterium, Candidatus Regiella, Acholeplasma, Tetragenococcus, Alteromonas, Nitrosovibrio, Alistipes, Oscillospira, Peptoniphilus, Micrococcus, Desemzia, Porphyromonas, Nitrosococcus, Natronincola, and Odoribacter. In contrast, the proportion of nine genera (blue) was significantly lower in patients with CRC: Lactococcus, Roseburia, Curvibacter, Lachnospira, Haemophilus, Alishewanella, Ewingella, Mitsuokella, and Diaphorobacter.

Figure 5

Comparison of the genera between obese and nonobese patients with colorectal cancer (CRC). In the obese patients with CRC, the genus Enterococcus (P = 0.04) was significantly lower, and the genera Polaribacter (P = 0.03) and Capnocytophaga (P = 0.03) were significantly higher than in the nonobese patients with CRC. In the controls, no significant differences were observed in the genera Enterococcus (P = 0.65) and Capnocytophaga (P = 0.74) between obese and nonobese participants. None of the controls has Polaribacter. *P < 0.05.

Figure 6

Association with metabolic markers. The Spearman correlation coefficient was used for metabolic markers to assess the correlations (a). The genus Polaribacter was observed in none of the controls. In the patients with CRC, particularly in the obese patients, the LDL concentrations were associated with the relative abundance of the genus Enterococcus (b). No strong correlation between the relative abundance of the genus Capnocytophaga or Polaribacter and metabolic markers, including FPG, HbA1c, FPI, TG, TC, HDL, and LDL concentrations and HOMA‐IR, was observed in patients with CRC. In controls, no significant correlation between the genus Enterococcus or Capnocytophaga and metabolic markers was noted. CRC, colorectal cancer; FPG, fasting plasma glucose; HOMA‐IR, homeostasis model assessment of insulin resistance; HDL, high‐density lipoprotein; LDL, low‐density lipoprotein; NA, not applicable; TC, total cholesterol; TG, triglyceride.

Figure 7

Analysis of Enterococcus species. E. faecalis was the most abundant species of the genus Enterococcus in the participants. In patients with colorectal cancer (CRC), the relative abundance of E. faecalis against all Enterococcus species (P < 0.01, a) or all microbiome components (P = 0.01, b) was significantly lower in obese versus nonobese patients (Kruskal–Wallis test, P = 0.003; Steel–Dwass test, P = 0.01). *P < 0.05. , E. faecalis; , E. saccharolyticus; , E. sulfureus; , E. lactis; , E. casseliflavus; , E. hirae; , E. gallinarum; , E. avium; , E. hermanniensis; , E. columbae; , E. cecorum; , E. haemoperoxidus; , unclassified.