Participants were recruited from patients diagnosed with colorectal cancer through endoscopic examination and pathological assessment at Zhengzhou Central Hospital Affiliated to Zhengzhou University in 2024. Exclusion criteria included patients with recurrent gastric cancer or secondary malignancies, as well as those undergoing chemotherapy, immunotherapy, or radiotherapy. Ultimately, patients with complete information were selected for the study. Participant clinical data are provided in Supplementary Table 1 . The procedures used in this study adhere to the tenets of the Declaration of Helsinki and have been approved by the Institutional Review Board (Ethical Batch Number: ZXYY2024278). Informed consent was obtained from all the participants. Tumor and paired normal tissues were collected and promptly frozen in a -80°C refrigerator.

The CRC cell lines HCT116 and SW480 were purchased from Procell Life Science Technology. They were routinely monitored by PCR to ensure they were mycoplasma-free and authenticated by short tandem repeat profiling. HCT116 cells in McCoy’s 5A medium, SW480 cells in Iscove’s Modified Dulbecco’s Medium. All the cells were cultured under standard incubator conditions (37°C, 5% CO₂) and passaged at approximately 90% confluence. We obtained Roseburia intestinalis from Ningbo Mingzhou Biotechnology Co., Ltd., with a concentration of 10⁹ colony-forming units (CFU)/mL (63, 64). The Roseburia intestinalis and HCT116 cells were co-cultured for 24 hours and the cells were collected for follow-up experiments.

Samples were retrieved from -80°C and each group of samples was mixed with four volumes of lysis buffer (8 M urea, 1% protease inhibitor, 3 μM TSA, and 50 mM NAM) and subjected to ultrasonic lysis. The lysate was then centrifuged at 4°C and 12,000 g for 10 minutes to remove cell debris. The supernatant was transferred to a new centrifuge tube, and the protein concentration was determined using a BCA assay kit.

Equal amounts of protein from each sample were subjected to enzymatic digestion. The volume was adjusted to uniformity using lysis buffer, and 20% TCA was slowly added to achieve a final concentration of 20%. The mixture was vortexed and allowed to precipitate at 4°C for 2 hours. The samples were then centrifuged at 4,500 g for 5 minutes to discard the supernatant, and the precipitate was washed 2–3 times with pre-cooled acetone. After air-drying the precipitate, 200 mM TEAB was added, and the precipitate was sonicated to disperse it. Trypsin was added at a 1:50 ratio (w/w, enzyme:protein), and digestion was allowed to proceed overnight. Dithiothreitol (DTT) was then added to achieve a final concentration of 5 mM, followed by reduction at 56°C for 30 minutes. Subsequently, iodoacetamide (IAA) was added to achieve a final concentration of 11 mM, and the mixture was incubated in the dark at room temperature for 15 minutes.

The peptide segments were dissolved in immunoprecipitation (IP) buffer (100 mM NaCl, 1 mM EDTA, 50 mM Tris-HCl, 0.5% NP-40, pH 8.0) and the supernatant was transferred to pre-washed resin. The mixture was placed on a rotating shaker at 4°C and gently shaken for overnight incubation. After incubation, the resin was washed four times with IP buffer and twice with deionized water. Finally, the elution was performed using 0.1% trifluoroacetic acid (TFA) elution buffer, and the peptides bound to the resin were eluted three times. The elution fractions were collected and vacuum freeze-dried. After drying, the samples were desalted according to the instructions provided with the C18 ZipTips and subsequently vacuum freeze-dried before analysis by liquid chromatography-mass spectrometry (LC-MS).

The peptides were dissolved in mobile phase A and separated using the NanoElute ultra-high-performance liquid chromatography (UHPLC) system. Mobile phase A consisted of an aqueous solution containing 0.1% formic acid and 2% acetonitrile, while mobile phase B was an acetonitrile-water solution containing 0.1% formic acid. The liquid chromatography gradient was set as follows: from 0 to 18 minutes, 6%-22% B; from 18 to 22 minutes, 22%-30% B; from 22 to 26 minutes, 30%-80% B; and from 26 to 30 minutes, 80% B, with a flow rate maintained at 450 nl/min. After separation via the ultra-high-performance liquid chromatography system, the peptides were injected into a capillary ion source for ionization and subsequently analyzed by the timsTOF Pro 2 mass spectrometer for data acquisition. The ion source voltage was set to 1.65 kV, and both the peptide precursor ions and their fragment ions were detected and analyzed using TOF. The data acquisition mode utilized was data-independent parallel accumulation serial fragmentation (dia-PASEF), with the first mass spectrum scan range set from 100 to 1700 m/z. Following the acquisition of a single first mass spectrum, eight acquisitions in PASEF mode were performed, with the second mass spectrum scan covering a range of 425-1025, using a 25 m/z window.

Using the MoMo tool based on the Motif-X algorithm to analyze the motif characteristics of modification sites. The analysis was based on peptide sequences composed of 10 amino acids upstream and downstream of all potential modification sites within the species. A feature sequence was considered a motif of modified peptides if the quantity of peptides exhibiting this sequence exceeded 20 and the statistical significance (P-value) was less than 0.000001. The relative quantification ratio of modification sites between two groups of samples was used as the fold change, with a threshold of greater than 1.5 for significant up-regulation and less than 1/1.5 for significant down-regulation.

The peptides were dissolved in mobile phase A and separated using the NanoElute ultra-high-performance liquid chromatography (UHPLC) system. Mobile phase A comprised an aqueous solution containing 0.1% formic acid and 2% acetonitrile, while mobile phase B consisted of an acetonitrile-water solution containing 0.1% formic acid. The liquid chromatography gradient was configured as follows: from 0 to 14 minutes, 6%-24% B; from 14 to 16 minutes, 24%-35% B; from 16 to 18 minutes, 35%-80% B; and from 18 to 20 minutes, 80% B, with a flow rate maintained at 500 nl/min. After separation via the ultra-high-performance liquid chromatography system, the peptides were injected into a capillary ion source for ionization and subsequently analyzed by the timsTOF Pro 2 mass spectrometer for data acquisition. The ion source voltage was set to 1.75 kV, and both the peptide precursor ions and their fragment ions were detected and analyzed using TOF. The data acquisition mode utilized was data-independent parallel accumulation serial fragmentation (dia-PASEF), with the first mass spectrum scan range set from 300 to 1500 m/z. After acquiring a single first mass spectrum, twenty acquisitions in PASEF mode were performed, with the second mass spectrum scan covering a range of 400-850, using a 7 m/z window.

For proteomic analysis, the raw LC-MS datasets were first searched against database and converted into matrices containing normalized intensity (the raw intensity after correcting the sample/batch effect) of proteins. The normalized intensity (I) was transformed to the relative quantitative value (R) after centralization. The formula is listed as follow: Rij= Iij/Mean(Ij), where i represents sample and j represents protein. For post-translational modification omics, firstly, the intensities of modified peptides (I) were centralized and transformed into relative quantitative values (R) of modified peptides in each sample. Then, the relative quantitative value of the modified peptide was divided by the relative quantitative value of corresponding protein to remove the influence from protein expression of modifications (65).

Differentially expressed proteins (DEPs) with a fold change ≥1.5 were mapped to the STRING database to extract high-confidence interactions (confidence score >0.7). The PPI network was visualized using the R package “visNetwork”. In the network, nodes represent DEPs, with colors indicating expression changes (blue: downregulated; red: upregulated) and intensity reflecting fold-change magnitude. Node size corresponds to the number of interacting partners. The proteomics data generated by mass spectrometry have been deposited in the ProteomeXchange Consortium (http://proteomecentral.proteomexchange.org) through the iProX partner repository with the accession number PXD066351.

An aliquot of 20 μg of protein sample was mixed with 4× sample buffer and diluted to 1×. An appropriate volume of protein lysis buffer was then added to achieve a final protein concentration of 1–2 mg/mL. The mixture was heated at 95°C for 10 minutes. Equal volumes of the sample and 20% pre-stained protein marker were loaded. Electrophoresis was conducted at a constant voltage of 80 V for 30 minutes, followed by an increase to 120 V until the bromophenol blue dye exited the separating gel. The NC membrane was equilibrated by immersing it in pre-chilled transfer buffer for 30 minutes. The transfer apparatus was assembled with the correct electrode orientation and placed in a 4°C environment, applying a constant current of 200 mA for 1 hour for the transfer process. The membrane was blocked with 5% non-fat dry milk prepared in 1× TBST at room temperature for 1 hour. The membrane was rinsed three times with TBST, each rinse lasting 10 minutes. The primary antibody was diluted in TBST containing 2.5% BSA and incubated overnight on a roller at 4°C. Afterward, a secondary antibody was applied and incubated at room temperature for 1 hour. A chemiluminescent HRP substrate was added and incubated for 2 minutes, followed by signal capture according to the operating instructions of the chemiluminescent imaging system.

The viability of cells was determined with Cell Counting Kit-8 (CCK-8). At a density of 2000 cells per well, cells were seeded into 96-well plates. Six replicate wells were set up for each group. The medium was replaced with 100μl fresh complete medium containing 10μl CCK8 solution following a culture of 0, 24, 48 and 72h, respectively. Then, the plates were incubated at 37°C and 5% CO₂ for 2h, and the absorbance was measured at a wavelength of 450nm. Three independent experiments were performed.

1000 cells/well were plated in 6-well plates, and pools of cells were used to assess growth and clonogenic ability. Ten days after plating the cells, clonogenic progenitors were determined and cells in each group were replated for 6 times. Colonies were rinsed twice with ice-cold PBS, fixed with 4% paraformaldehyde for 20 min on ice and washed twice with PBS. After fixed with methanol, cells were stained with 0.1% crystal violet solution 30 min and then the colonies were imaged and counted. The clone formation rate was calculated as follows: number of cell clone formation/number of inoculated cells ×100%. Three independent experiments were performed.

The data were presented as mean ± standard deviation. Statistical analysis between two groups of normally distributed data was performed using the Student’s t-test. Non-normally distributed data were analyzed using the nonparametric Mann-Whitney U test. For comparisons among multiple groups, homogeneity of variance was first verified using Levene’s test, followed by standard one-way analysis of variance (ANOVA). The P <0 .05 was considered significant. Statistical analysis was performed on SPSS 24.0.

In colorectal cancer (CRC) tumor tissues, there is a marked increase in global lysine lactylation levels, indicating a metabolic reprogramming characteristic of tumor cells ( Supplementary Figures 1A, B ). Upon intervention with R.i, we observed a significant increase in acetylation levels alongside a decrease in lactylation levels in CRC cells ( Figure 1A ; Supplementary Figure 1C ). This shift suggests a potential interplay between these two PTMs that may influence cancer cell behavior and metabolism. To systematically investigate the regulatory patterns of Ruminococcus intestinalis on lactylation and acetylation in CRC, we performed a quantitative proteomics study focusing on both lactylation and acetylation. Our experimental workflow, depicted in Figure 1B , illustrates the steps taken to investigate these modifications systematically, including protein extraction, trypsin digestion, and enrichment of lactylated peptides using anti-Pan Kla antibodies, followed by identification through liquid chromatography/tandem mass spectrometry (LC-MS/MS). Overall, this study provides insights into the dynamic changes in acetylation and lactylation in CRC and highlights the potential role of R.i in metabolic regulation within this cancer context.

In a total of 8123 peptide segments, 5791 modified peptides were identified, leading to the detection of 6134 distinct acetylation modification sites across 3037 acetylated proteins. Among these, 5443 acetylation modification sites were identified in 2710 quantifiable acetylated proteins ( Supplementary Figure 2A ). A total of 23564 peptide segments were analyzed, resulting in the identification of 7594 modified peptides, which revealed 7882 distinct lactylation modification sites across 2386 lactylated proteins. Among these, 7095 lactylation modification sites were identified in 2156 quantifiable lactylated proteins ( Supplementary Figure 2B ).

Acetylation analysis revealed a total of 493 up-regulated proteins and 794 down-regulated proteins. The acetylation sites demonstrated a distribution of 595 up-regulated sites and 1037 down-regulated sites ( Figure 2A ). In addition, lactylation profiling identified 326 up-regulated proteins alongside 1401 down-regulated proteins, with the corresponding lactylation sites showing 473 upregulated and 3727 down-regulated sites ( Figure 2B ). The differentially expressed proteins are presented in Supplementary Tables 2, 3 .

The functional roles of proteins are intricately linked to their subcellular localization. To characterize the distribution of each acetylated (Kac) protein, we conducted a subcellular distribution analysis, as shown in Figure 2C . Our results revealed that, in HCT116 cells induced by R.i, acetylated proteins were predominantly localized in the nucleus (39.4%), followed by the cytoplasm (31.18%), mitochondria (11.74%), plasma membrane (5.7%), and extracellular (4.36%). To achieve a thorough understanding of the functional characteristics of various proteins, we performed a comprehensive functional annotation of the identified proteins. The Supplementary Figure 2C illustrates the number of acetylated proteins and modification sites annotated to COG/KOG functional classification, protein domains, KEGG pathways, Gene Ontology, WikiPathways, Transcription Factor, Reactome and HallMark. Using the MoMo analysis tool based on the motif-x algorithm, we examined the motif characteristics of the acetylation modification sites. The analysis focused on peptide sequences composed of 10 amino acids upstream and downstream of all identified modification sites. Figure 2E displays the variation in frequency of amino acid occurrence near the modification sites. According to COG/KOG category analysis, proteins involved in Posttranslational modification, transport, and metabolism are more likely to be acetylated ( Supplementary Figure 3A ). Most differentially expressed proteins are enriched in Posttranslational modification, Translation, Lipid transport and metabolism ( Supplementary Figure 3B ).

Lactylated proteins exhibited a different distribution ( Figure 2D ), with the majority found in the nucleus (61.52%), followed by the cytoplasm (23.23%), mitochondria (5.02%), and plasma membrane (3.39%). Supplementary Figure 2D illustrates the distribution of lactylated proteins and their corresponding modification sites across various functional categories, including COG/KOG classifications, protein domains, KEGG pathways, Gene Ontology terms, WikiPathways, transcription factors, Reactome annotations, and Hallmark signatures. To further elucidate the motif characteristics of the lactylation modification sites, we employed the MoMo analysis tool based on the motif-x algorithm. This analysis focused on the peptide sequences, consisting of 10 amino acids, flanking all identified modification sites. As depicted in Figure 2F , we observed notable variations in the frequency of amino acid occurrences adjacent to the modification sites.

According to the COG/KOG classification analysis, proteins involved in post-translational modifications, signaling mechanisms, transcription, lipid transport, and metabolism are more susceptible to lactylation ( Supplementary Figure 4A ). The majority of differentially expressed proteins were found to be enriched in pathways associated with translation, infectious diseases, and cancer ( Supplementary Figure 4B ). This enrichment suggests that lactylation may play a crucial role in modulating key biological processes and pathways implicated in tumorigenesis and disease progression. The propensity for lactylation among these functionally significant proteins underscores its potential impact on cellular signaling networks and highlights lactylation as an essential post-translational modification warranting further investigation in the context of cancer biology.

The quantitative proteomic analysis of CRC cells treated with R.i was conducted using the 4D-Fast DIA. An overview of protein identification is presented in Supplementary Figure 5A , where a total of 60,089 identified peptide segments were recorded, with 57,492 unique peptide sequences; a total of 7,987 proteins were identified, and 7,857 proteins were suitable for quantitative comparison. To gain an in-depth understanding of the functional characteristics of the identified proteins, we performed comprehensive functional annotations on them. Among these, 6,591 proteins were classified into COG/KOG functional categories, 4,803 proteins were associated with protein domains, 3,850 proteins were enriched in KEGG pathways, 7,635 proteins were enriched in Gene Ontology (GO) terms, 3,923 proteins were enriched in WikiPathways, 795 proteins were identified as transcription factors (TFs), 5,293 proteins were enriched in Reactome pathways, and 2,497 proteins were associated with Hallmark pathways ( Supplementary Figure 5B ).

Compared to the control group, there were 1,872 upregulated proteins and 2,280 downregulated proteins ( Supplementary Figure 5C ). The functional classification of the differentially expressed proteins is presented in Supplementary Figure 5D . Based on the findings presented in Supplementary Figure 5E , the differentially expressed proteins were significantly enriched in several key metabolic pathways, including glutathione metabolism, N-glycan biosynthesis, sphingolipid metabolism, the interconversion of pentose and glucuronate, and mucin-type O-glycan biosynthesis. These pathways are critical for various cellular functions, and their alteration may have significant implications for cellular behavior and the overall metabolic landscape within the context of the treatment being investigated. Further exploration of these enriched pathways could provide valuable insights into the molecular mechanisms underlying the observed changes in protein expression.

First, a comparison of the protein groups and the acetylation modified groups was conducted at the quantitative level, and the expression abundance of the proteins was statistically analyzed. Subsequently, differentially expressed proteins and differential modification sites were identified based on a specified threshold, followed by significance distribution analysis.

The Venn diagram was used to illustrate the intersection of identified proteins between the protein group and the modified group, with a total of 2,832 proteins ( Figure 3A ). The distribution of intensity values for the proteins identified in the protein group and those that underwent modifications was illustrated using a histogram ( Figure 3B ). Based on the criteria for differential selection from proteomics and modificationomics, differentially expressed proteins and differential modification sites were identified separately. The significantly upregulated modification sites in the experimental group are RPL30_K42, GTF2F1_K407, WDR43_K564, RPS29_K48 and DYNC1LI1_K303, while the significantly downregulated proteins are EP300_K1554, CREBBP_K1597, EP300_K1555, IMPDH2_K511 and RALY_K165 ( Figure 3C ). The acetylome profiling identified significant enrichment of differentially expressed proteins in several critical metabolic pathways, including: Glycolysis, Lipid metabolism, Pyruvate metabolism, Glycerophospholipid metabolism. The pathways enriched with modified proteins are presented in Figure 4 .

The Venn diagram was used to illustrate the intersection of identified proteins between the protein group and the lactylation modified group, with a total of 2,239 proteins ( Figure 3D ). The distribution of intensity values for the proteins identified in the protein group and those that underwent modifications was illustrated using a histogram ( Figure 3E ). The significantly upregulated proteins and their modified sites are SMC2_K749, HSP90AB1_K186, TXNRD1_K129, PRDX1_K178, and ANXA2_K49, while the significantly downregulated modified proteins and their sites are RALY_K165, MATR3_K487, TRIM25_K392, CHD8_K246, and UTP3_K12 ( Figure 3F ). Lactylome analysis demonstrated distinct protein enrichment in: Glycolysis, Galactose metabolism, Pentose phosphate pathway, Non-homologous end-joining. The pathways enriched with modified proteins are illustrated in Figure 5 . Notably, glycolysis emerged as the principal convergent pathway between acetylation and lactylation modifications, suggesting its central regulatory role in metabolic reprogramming under these PTMs.

Functional enrichment analysis of proteins with concurrent lactylation and acetylation modifications clustered into pathways such as ATP-dependent chromatin remodeling, RNA degradation, glycolysis, and the cell cycle ( Supplementary Figure 6A ). Similarly, functional enrichment analysis of the modification sites exhibiting both lactylation and acetylation revealed clustering into pathways including the spliceosome, ATP-dependent chromatin remodeling, glycolysis, lysine degradation, and central carbon metabolism ( Supplementary Figure 6B ). This analysis highlights the significant biological pathways associated with proteins undergoing simultaneous lactylation and acetylation, indicating their potential roles in critical cellular processes such as chromatin remodeling, energy metabolism, and RNA processing. The overlapping pathways suggest a complex interplay between these PTMs in regulating fundamental cellular functions.

This study further analyzed the co-expression network between differentially expressed proteins (DEPs) and lactylation modification sites, revealing potential functional interactions. Functional enrichment analysis demonstrated significant associations with ribosome biogenesis (20 upregulated and 14 downregulated proteins) and spliceosome (8 upregulated and 8 downregulated proteins) ( Figure 6 ).

The proliferation of cells is a major feature of tumors. Using CCK-8 and colony formation assays, we evaluated the effects of R.i and R.i-conditioned medium (R.i-CM) on CRC cell proliferation. CCK8 kit was used to determine cell viability. The results demonstrated that R.i-CM significantly inhibited CRC cell viability compared to the control group, whereas live R.i showed no significant inhibitory effect ( Supplementary Figure 7A ). The colony formation assay was performed to confirm the clonogenic potential. The results showed that R.i-CM triggered a significant decrease of colony numbers in CRC cells ( Supplementary Figures 7B, C ).