Extended Data Fig. 8: Metabolomic changes in the tricarboxylic acid (TCA) pathways and metagenomic changes in amino acid metabolism and other representative pathways.

a, Quantified levels of metabolites involved in the tricarboxylic acid (TCA) pathway. The levels of the three TCA metabolites, succinate, fumarate and malate, were significantly higher P < 0.005; one-sized Mann–Whitney U test) in S0 samples (and SIII/IV samples for fumarate) (+++, P < 0.005; ++, P < 0.01; +, P < 0.05) compared to healthy control samples. It is uncertain what is the cause of accumulation of succinate, fumarate and malate in the feces of patients with early colorectal cancer, despite extremely low concentrations of other TCA intermediates such as 2-oxoglutarate. It is known that some bacteria synthesize ATP using a reverse reaction of succinate dehydrogenase and produce succinate as a byproduct, as part of fumarate respiration, in which fumarate rather than molecular oxygen is used as electron acceptor. The boxes represent 25th–75th percentiles, black lines indicate the median, whiskers extend to the maximum and minimum values within 1.5× the interquartile range and dots indicate outliers. The concentration is shown on the y axis (nmol g⁻¹). Healthy (n = 127), MP (n = 45), S0 (n = 30), SI/II (n = 80), SIII/IV (n = 68). N.D., not detected and/or not determined. b, Pathway modules for metabolism types omitted from Fig. 3b. The pathway modules are modified from KEGG pathway maps ‘Alanine, aspartate and glutamate metabolism’, ‘Cysteine and methionine metabolism’, ‘Methane metabolism’ and ‘Arginine and proline metabolism’. ‘Leucine degradation’ is constructed based on leucine metabolism of Clostridium difficile, as the bacterial map is not available in KEGG. For each KO gene, bar plots show KO gene abundances averaged over samples within each of the five groups, healthy (n = 251), MP (n = 67), S0 (n = 73), SI/II (n = 111) and SIII/IV (n = 74) in order of left to right, and colored according to the order of the values. Each KO gene is composed of organism genes represented by circles. The sizes and colors of the circles are proportional to the relative abundances of the organism genes. Organism genes are grouped into one row and indicated by the organism name. The three most abundant organisms in the healthy controls are shown using three letter codes (for example, ova for Oscillibacter valericigenes, kpe for Klebsiella pneumoniae 342). Abbreviations for other organism names can be found in Supplementary Table 4. Gene numbers linked to each of the genes are listed in Supplementary Table 5. Dots in each pathway represent intermediate metabolites. The colors of the boxes of pathway components are marked in red for significant elevation (P < 0.005; one-sided Mann–Whitney U test) for any of the stages (MP, S0, SI/II and SIII/IV) compared to the healthy controls.