Skip to main content
NCBI home page
Clinical Epigenetics logoLink to Clinical Epigenetics
. 2025 Dec 5;18:11. doi: 10.1186/s13148-025-02012-w

Bacterial lipopolysaccharide alters DNA methylation in colorectal cancer cells

Jessica Permain 1, Arielle Sulit 1, Timothy Eglinton 1, Rachel Purcell 1,✉
PMCID: PMC12797708  PMID: 41351128

Abstract

Colorectal cancer (CRC) is commonly associated with epigenetic modifications, including altered DNA methylation. Recent studies suggest that tumour-resident bacteria may influence CRC development, yet the impact of bacteria on epigenetic regulation is not understood. This study investigates the effect of lipopolysaccharide (LPS) from Fusobacterium periodonticum and Bacteroides fragilis, bacteria that are abundant in CRC tumours with high CpG island methylator phenotype (CIMP), on DNA methylation in HT29 colorectal cancer cells. HT29 cells were treated with LPS from F. periodonticum, B. fragilis, or a combination of both. DNA methylation was assessed using reduced representation bisulfite sequencing (RRBS), followed by bioinformatic analysis to identify differentially methylated CpG sites. RT-qPCR was used to analyse the expression of selected genes with altered CpG promoter methylation. F. periodonticum LPS treatment induced both hypermethylation and hypomethylation in HT29 cells, with significant hypermethylation observed near specific promoter regions, including PEPD and VAV3, with associated decrease in gene expression of these genes. B. fragilis LPS treatment predominantly induced hypomethylation. Co-treatment with both LPS molecules resulted in distinct methylation patterns, with B. fragilis LPS attenuating F. periodonticum LPS-induced hypermethylation. Bacterial LPS can induce dynamic alterations in DNA methylation profiles in HT29 colorectal cancer cells, leading to changes in gene expression. These findings suggest a novel link between tumour-resident bacteria and DNA methylation in colorectal cancer, highlighting, for the first time, a potential mechanism by which bacteria may influence colorectal carcinogenesis.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13148-025-02012-w.

Introduction

Colorectal cancer (CRC) is a common malignancy, with an incidence rate of 40/100,000 in certain North American, European and Oceanian populations [1, 2]. Recent studies have explored the influence of tumour-resident bacteria on CRC development, with a focus on immune modulation [3]. However, the impact that bacteria and their components have on genetic and epigenetic regulation is less well studied.

Epigenetic modifications, such as altered DNA methylation, are hallmarks of various cancers, including CRC, with both hyper- and hypomethylation playing crucial roles in colorectal carcinogenesis. Global hypomethylation, observed in 70% of sporadic colorectal tumours, can induce genetic instability [4]. LINE-1 hypomethylation shows a linear correlation with advancing stage of CRC [5] and liver metastases have been shown to have LINE-1 hypomethylation compared to primary CRC tissues, also supporting the association between the degree of DNA methylation and stage of CRC development [6]. Liver metastases are also associated with hypomethylation of the proto-oncogenes MET, CHRM3 and RAV3IP, which influence cell proliferation, survival and invasion [6]. Conversely, DNA hypermethylation of CpG islands near genetic promoters can inactivate tumour suppressor genes, resulting in tumour progression [7, 8].

The serrated pathway of CRC development is characterised by DNA hypermethylation and is frequently associated with the CpG island methylator phenotype (CIMP). Hypermethylation of mismatch repair genes, such as MLH1, is common in CIMP-positive tumours, and can result in mismatch repair deficiency (dMMR) [9] Interestingly, tumours arising from the serrated pathway exhibit distinct bacterial profiles; analysis of dMMR tumours revealed an abundance of Fusobacterium nucleatum, Fusobacterium periodonticum and Bacteroides fragilis, which was not observed in mismatch repair proficient (pMMR) tumours [10]. This suggests a potential association between tumour-resident bacteria and hypermethylation.

Studies of the effects of specific bacterial components on methylation have shown that E. coli lipopolysaccharide (LPS) can alter DNA methylation of COX-2 in epithelial cell lines [11] and induce CpG island hypermethylation of SOCS-1 in macrophage cell lines [12]; SOCS-1 methylation has also been associated with CRC and linked to tumour stages [13].

We have previously shown that F. periodonticum and B. fragilis are differentially abundant in CRC tumours characterised by high CIMP, and these tumours are enriched for LPS biosynthetic processes [14]. Given these findings, this study aims to investigate the effect of F. periodonticum and B. fragilis LPS on DNA methylation in colorectal cancer cells, to explore the link between tumour-resident bacteria and altered DNA methylation in CRC.

Methods

Lipopolysaccharide preparation and treatment

LPS was extracted from Fusobacterium periodonticum 2/1/31 and Bacteroides fragilis 2/1/16 using Bacterial Lipopolysaccharides (LPS) Extraction Kit (Alpha Diagnostic International) as per the manufacturer’s instructions. The concentration of LPS treatments for all experiments was 300 ng/mL.

Cell culture and treatment

HT29 cells (American Type Culture Collection, USA) were cultured in McCoy’s medium (ThermoFisher), with 10% fetal bovine serum and 1% penicillin–streptomycin and grown at 5% CO2. For all experiments, cells were seeded at a density of 1 × 105 in 24-well plates and incubated for 48 h at 37 °C. LPS treatments were added and incubated for an additional 24 h at 37 °C, prior to DNA extraction.

DNA extraction

Genomic DNA was extracted from HT29 cells using the Qiagen DNeasy® Blood and Tissue kit according to the manufacturer’s instructions. Briefly, proteinase K (20 μL) was added to each sample, followed by 200 μL of AL buffer, then the samples were vortexed and incubated at 56 °C for 10 min. Ethanol (200 μL) was added to the samples and the sample solution was pipetted onto the DNeasy® mini spin columns, centrifuged through the column, washed with AW1 and AW2 wash buffers (500 μL) before drying with centrifugation of 6000 × g for 1 min between washes and 20,000 × g for 3 min to dry the column. DNA was eluted with AE buffer (200 μL) and centrifugation at 6000 × g for 1 min.

Reduced representation bisulfite sequencing

Reduced representation bisulfite sequencing (RRBS) library preparation was performed using 4 μL of previously extracted DNA and the Zymo-Seq RRBS Library Kit (Zymo) following the manufacturer’s instructions. Sequencing was then outsourced to Quick Biology, California, USA using NovaSeq, read length was 2 × 150 bp and sequencing depth was 90 million reads (45 million each direction) per sample.

Computational data analysis

Trim-Galore was used to perform adapter and quality trimming on the reads [15], with the options –rrbs and –non_directional set. Sequences that matched to phiX were removed using bowtie2 [16]. The reads were then mapped to the human genome and methylated sites were celled using Bismark [17]. The results were input into MethylKit for downstream analysis. Minimum coverage for a site to be considered in the analysis was 5x. Sites considered to be differentially methylated between groups had a q-value < 0.05 and differential methylation > 15%. Differentially methylated sites were annotated using bed files from https://genome.ucsc.edu/cgi-bin/hgTable. Analysis focused on sites within the promoter regions.

Quantitative gene expression analysis

RNA was extracted from LPS-treated cells using RNeasy Mini Kit (Qiagen), as described in the manufacturer’s instructions. Reverse transcription was performed using qScript cDNA SuperMix (Quantabio, USA), with 2 μL of sample RNA and 18 μL of mater mix. Thermocycling conditions were 5 min at 25 °C, 60 min at 42 °C, 5 min at 82 °C, hold at 4 °C. TaqMan real-time quantitative polymerase chain reaction (RT-qPCR) was carried out to quantify expression of PEPD and VAV3 using HPRT and ACTB as reference genes (Table 1). RT-qPCR was carried out using the LightCycler® 480 instrument (Roche), with an annealing temperature of 60 °C and 50 amplification cycles. All reactions were carried out in triplicate.

Table 1.

TaqMan™ (ThermoFisher) probes used for gene expression assays

Target gene TaqMan™ probe Amplicon length
HPRT Hs02800695_m1 82
ACTB Hs01060665_g1 63
PEPD Hs00944655_m1 74
VAV3 Hs00944655_m1 75
Open in a new tab

Statistical analysis

Relative gene expression was quantified using the ΔΔCT method [18]. Paired t-tests were performed to determine statistically significant differences in relative gene expression using GraphPad Prism 10.2.3 (GraphPad Software, USA). Statistical significance was considered p < 0.05.

Results

Quality control was performed for each sample, including phiX removal, before input into bismark. Read counts from each step of the process is shown in Supplementary Table S1.1, as well as mapping efficiency of reads to the human genome. Counts of methylated and unmethylated C’s, as identified by bismark, are described in Supplementary Table S1.2.

F. periodonticum LPS treatment alters DNA methylation in HT29 colorectal cancer cells

The effects of F. periodonticum LPS on DNA methylation were determined using RRBS and bioinformatic analysis. HT29 cells were treated with ± 300 ng/mL of F. periodonticum LPS and incubated for 24 h at 37 °C. RRBS was used to determine methylation differences between treated and untreated cells.

The top 20 hyper- and hypomethylated promoter regions were analyzed (Fig. 1A). The most prominent hypermethylation was observed on chromosome 2 at site 201,116,552 (q = 1.397 × 10⁻5) where there was an approximate 25% increase in methylated sites in LPS-treated cells compared to untreated (Fig. 1B). Chromosome 9 was the second most hypermethylated at site 41,074,474 (q = 2.176 × 10⁻⁷) (Fig. 1A), with LPS treated cells showing near 100% methylation, an approximate 23% increase in methylated sites compared to untreated cells (Fig. 1B). The most significant hypomethylation occurred on chromosome 17 at site 82,236,799 (q = 2.126 × 10⁻⁷) (Fig. 1A), where untreated cells showed 83–90% methylated sites, compared to F. periodonticum LPS-treated cells with 18–40% methylation.

Fig. 1.

Fig. 1

Open in a new tab

Differential methylation in Fusobacterium periodonticum treated HT29 cells. A shows methylation differences between F. periodonticum LPS-treated and untreated HT29 cells with q-values stated. Y- axis indicates specific chromosomal sites of methylation. B shows the percentage of methylated sites in each samples within specific chromosomal sites. C shows each specific hyper- and hypomethylated site and the distance in number of base pairs to the closest genetic feature on the right handed Y-axis. Fpe (F. periodonticum LPS), meth.diff (differential methylation), %mC (percentage of methylated cytosine)

The nearest genetic features to hyper- or hypomethylated promoter regions were identified using bed files (Fig. 1C). Hyper-methylated sites were located within 250 bases of genes including LOC124902161, LOC105373836, CYP11A1, CD200, SERP2, GNA12, FAM234B, and DLEU7. Hypomethylated sites were closest to YEAT2, RASL10A, LOC105371688, NCOA1, IKZF3, and MIR6787, all within 250 bases of their respective promoter regions.

F. periodonticum LPS-induced methylation influences gene expression in HT29 cells

To better understand the effect of site-specific differential methylation on gene transcription, HT29 cells were treated with F. periodonticum LPS (300 ng/mL) and relative gene expression was determined using RT-qPCR. We observed an inverse relationship between gene expression and percentage methylation in both PEPD (Fig. 2A, B) and VAV3 (Fig. 2C, D). F. periodonticum LPS induced a significant decrease (p = 0.0391) in PEPD gene expression compared to untreated cells (Fig. 2A). This decrease in gene expression corresponded with a significant (q = 0.01) increase in percentage methylation of the PEPD promoter in cells treated with F. periodonticum LPS compared to untreated controls (Fig. 2B). Furthermore, F. periodonticum LPS treatment resulted in a significant (p = 0.0074) decrease of VAV3 gene expression (Fig. 2C) compared to untreated cells. The decrease in gene expression corresponded to a significant (q = 0.006) increase in percentage methylation of the VAV3 promoter in F. periodonticum LPS treated cells (Fig. 2D). This data indicates an inverse relationship between LPS induced DNA methylation and gene expression in HT29 cell lines.

Fig. 2.

Fig. 2

Open in a new tab

Relationship between gene expression and LPS induced DNA methylation. A Relative gene expression of PEPD in HT29 cells treated with F. periodonticum LPS compared to untreated cells. B Percentage of methylation of PEPD promoter region treated with F. periodonticum LPS. C Relative gene expression of VAV3 treated with F. periodonticum LPS compared to untreated cells. D Percentage of methylation of VAV3 promoter region compared to untreated cells. Three replicates were used for gene expression and data was analysed with a paired t-test. %mC (percentage of methylated cytosine)

Bacteroides fragilis LPS treatment alters DNA methylation in HT29 colorectal cancer cells

Differential methylation was also observed in HT29 cells treated with B. fragilis LPS (300 ng/mL) compared untreated cells. Notably, only 10 specific sites exhibited hypermethylation, with the most significant being on chromosome 7 at site 26,398,715 (q- 4.466 × 10−3) (Fig. S1.3) and the majority of altered sites being hypomethylated (Fig.  S1.3). Chromosome 17 displayed two sites of differential hypermethylation (site 39,607,040 q = 1.805 × 10−2 and site 30,292,106 q = 3.361 × 10−2) and one site of hypomethylation (site 12,790,189 q = 6.432 × 10–3) (Fig. S1.3).

The majority of hyper- and hypomethylated sites were located within 300 base pairs of promoter regions (Fig. S1.3). For the hypermethylated sites, RBM14-RBM4 was the closest genetic feature, with methylated sites situated approximately 10–20 base pairs from the promoter region (Fig. S1.3). The next hypermethylated site closest to a promoter was associated with the ZMYM2 gene, with hypermethylation occurring approximately 150 base pairs from the promoter region NEUROD2 had the most distant hypermethylated site, approximately 900 base pairs from its promoter region.

Among the top 20 hypomethylated sites shown in Fig. S1.3, CPT1A was the closest genetic feature, hypomethylation occurring within 10–20 base pairs of its promoter region. BNAT1 and ONECUT2 followed, with hypomethylated sites approximately 100 base pairs from their respective promoter regions (Fig. S1.3).

These data suggests that B. fragilis LPS induces hypomethylation more frequently than hypermethylation.

F. periodonticum and B. fragilis LPS co-treatment alters DNA methylation in HT29 colorectal cancer cells

Treatment with LPS from F. periodonticum and B. fragilis LPS in this study has shown distinct modulation of methylation, with B. fragilis LPS demonstrating more frequent hypomethylation compared to F. periodonticum LPS. However, it was not known how co-treatment with these LPS molecules may affect methylation patterns. Co-treatment with both LPS molecules induced both hyper- and hypomethylation in HT29 cells. Chromosome 17 at site 76,384,993 showed the most substantial increase in methylated sites by 37% (q = 4.999 × 10–4) in LPS co-treated cells compared to untreated (Fig. S1.4), whereas, chromosome 20 at site 63,253,861 displayed the largest difference of hypomethylation in LPS treated cells compared to untreated, a 55% decrease in methylated sites (Fig. S1.4). Conversely, chromosome 20 at site 63,253,861 showed the most significant hypomethylation (q = 6.284 × 10–5), a 55% decrease compared to untreated cells.

The majority (55%) of hypermethylated sites were located within 250 base pairs of the promoter regions, with the SYNE3 gene promoter being closest, approximately < 50 base pairs from the hypermethylated site. Forty-five percent of the hypomethylated sites were located within 250 base pairs of promoter regions, with the PRKG2 gene promoter being the closest (0–20 base pairs) to the hypomethylated site (Fig. S1.4). CCDC82, CETN1 and ZBTB42 genes were also identified as having differentially hypomethylated sites close to their promoter regions, all of which were located approximately 10–70 base pairs from their respective promoters (Fig. S1.4).

B. fragilis LPS attenuates F. periodonticum LPS induced hypermethylation

Two genes were selected for further investigation due to their potential links to colorectal cancer. The percentage of differential promoter hypermethylation and corresponding q-values were determined for each gene and treatment. Both VAV3 (q = 0.006) and PEPD (q = 0.01) were found to be hypermethylated when HT29 cells were treated with F. periodonticum LPS, compared to untreated controls. The addition of B. fragilis LPS resulted in a decrease in methylation in the promotor region of PEPD, compared to F. periodonticum LPS alone (Fig. 2). The co-treatment with B. fragilis + F. periodontium LPS also resulted in a decrease in methylation of the promoter region of VAV3; this resulted in methylation levels below both F. periodonticum LPS treatment (q = 0.0787) and untreated cells (q = 0.7896) (Fig. 3).

Fig. 3.

Fig. 3

Open in a new tab

LPS treatment alters VAV3 and PEPD percentage methylation. Percentage of methylation of the VAV3 and PEPD promoter region, in response to F. periodonticum LPS and B. fragilis + F. periodonticum LPS treatment. Fpe (F. periodonticum), Bfr (B. fragilis), %mC (percentage of methylated cytosine)

Discussion

Epigenetic modifications, particularly aberrant DNA methylation patterns, are a hallmark of cancer development. However, precise mechanisms by which the microbiome influences these epigenetic alterations remain elusive.

Recently, tumour-resident bacteria have been found to be associated with host methylation. A study, published in 2024, investigated CpG methylation and bacterial associations in tumour tissues from the cancer genome atlas (TCGA) and their own cohort [19]. Both Fusobacteria and Bacteroides species, as well as Stenotrophomanas, Ralstonia and Clostrodium species were associated with increased promoter CpG methylation in tumour tissues [19]. However, a mechanistic link between tumour resident-bacteria is yet to be established.

This study has demonstrated for the first time that bacterial lipopolysaccharide (LPS) treatment of CRC cell lines results in dynamic alterations of the DNA methylation profiles, encompassing both hypermethylation and hypomethylation alterations. This is the first report of LPS altering methylation in the context of cancer, not just colorectal. These epigenetic alterations can subsequently alter gene expression patterns, which may potentially lead to the dysregulation of cellular processes implicated in colorectal carcinogenesis. This study provides evidence that bacterial LPS may be capable of independently inducing significant changes to the epigenetic landscape, thus, introducing a novel microbial mediated pathway of oncogenesis.

Several pathogenic bacteria have been implicated in the hypermethylation of tumour suppressor genes in CRC. Both F. nucleatum and Hungatella hathewayi are associated with hypermethylation of the tumour suppressor genes, CDX2 and MLH1, in CRC tissue compared to adjacent healthy mucosa [20]. Additionally, in vitro experiments have demonstrated that F. nucleatum and H. hathewayi can cause an increase in global hypermethylation in CRC cell lines, relative to untreated cells [20]. Our study has demonstrated that additional bacterial species can induce both hyper- and hypomethylation. Specifically, LPS molecules derived from F. periodonticum and B. fragilis were sufficient to elicit differential hyper- and hypomethylation patterns in HT29 CRC cells.

F. periodonticum LPS induced widespread promoter hypermethylation at a range of genomic features. Identification of the top twenty hypermethylated sites and their associated genetic features included tripartite motif-containing 67 (TRIM67). TRIM67 functions to stabilize and activate p53, thereby inhibiting CRC proliferation and suppressing cell invasion [21, 22]. The observed hypermethylation induced by F. periodonticum LPS may result in the inhibition of gene transcription, consequently leading to p53 inactivation and subsequent promotion of colorectal carcinogenesis. F. periodonticum LPS treatment also resulted in hypermethylation of peptidase D (PEPD) as well as downregulation of PEPD gene expression compared to untreated cells. PEPD has been shown to attenuate the oncogenic potential of epidermal growth factor receptor (EGFR) and human epidermal growth factor receptor 2 (HER2) in colorectal cells, resulting in diminished cellular proliferation [23]. F. periodonticum LPS-induced hypermethylation of the PEPD promoter and subsequent reduction in gene expression observed in our study may lead to enhanced oncogenic signaling through EGFR and HER2 in colorectal cancer microenvironments with elevated F. periodonticum LPS, leading to increased cancer cell proliferation.

However, hypermethylation may not always impact tumour suppressor genes to promote colorectal cancer. F. periodonticum LPS treatment of the HT29 cells resulted in promoter hypermethylation of the vav guanine nucleotide exchange factor 3 (VAV3) oncogene. VAV3 protein overexpression has been correlated with enhanced cellular proliferation in colorectal cancer [24]. Conversely, VAV3 knockdown studies have demonstrated cell cycle arrest and attenuated cellular growth compared to control cells, suggesting a pro-oncogenic role for VAV3 [24]. The epigenetic silencing of VAV3 through DNA promoter hypermethylation may therefore contribute to decreased cancer cell proliferation, indicating a potential protective mechanism of LPS against colorectal cancer. This finding highlights the complex, context-dependent nature of epigenetic regulation in carcinogenic processes.

The presence of Bacteroides fragilis within colorectal neoplasms has been observed in many studies, and this study has demonstrated that its LPS molecule can modulate DNA methylation patterns in colorectal cancer cell lines. In comparison to F. periodontium LPS, B. fragilis LPS induced a greater degree of hypomethylation relative to hypermethylation. Treatment of HT29 CRC cells with B. fragilis resulted in the hypermethylation of the ZMYM2 promoter, which encodes for a zinc finger protein. Recently studies have shown that ZMYM2 is involved in methylation of embryonic stem cells, and ZMYM2-/- mice resulted in failed DNA methylation and upregulation of germline genes [25]. Our data showing methylation and, thus, potential epigenetic silencing of ZMYM2 may be why the B. fragilis LPS treatment group showed lower overall hypermethylation compared to the other treatment groups.

Very few studies have looked into this potential mechanism in cancer; however, it has been noted that ZMYM2 knockdown xenographs in mice show decreased ovarian cancer cell growth compared to ZMYM2 + / + models, suggesting oncogenic potential [26]. The implications of hypermethylation of the ZMYM2 promoter in colorectal cancer needs to be further investigated, however, it suggests a potential mechanism linking tumour-resident bacteria to the altered epigenetic landscape observed in tumours.

As previously mentioned, B. fragilis LPS-treated cells displayed a significant differential hypomethylation of genetic promoters compared to untreated cells. Several known oncogenes were included in the hypomethylated sites. The one cut domain family member 2 (ONECUT2) gene, associated with upregulation of the ONECUT protein in various malignancies, including colorectal cancer, exhibited promoter hypomethylation following B. fragilis LPS treatment [27]. ONECUT2 overexpression has been linked to enhanced cell migration, invasion and metastatic potential [28]. Our findings suggest that B. fragilis LPS-induced hypomethylation of the ONECUT2 promoter may lead to increased ONECUT protein expression, potentially influencing the metastatic capacity of colorectal cancer cells.

The tumour microenvironment is complex, with a wide range of different tumour-resident bacteria present. Therefore, it is critical to understand how different bacterial LPS affect DNA methylation when there is more than one type of LPS molecule present in an environment. Co-treatment of HT29 colorectal cancer cells with both F. periodonticum LPS and B. fragilis LPS elicited unique hyper- and hypomethylated sites compared to the individual LPS exposures. However the co-treatment of B. fragilis LPS with F. periodonticum LPS resulted in a decrease in percentage methylation in the promoter region associated with both VAV3 and PEPD compared to F. periodonticum LPS alone, suggesting that B. fragilis LPS attenuates the hypermethylation induced by F. periodonticum LPS. This phenomenon has previously been observed in the context of cytokine production, where F. periodonticum LPS-induced cytokine release was attenuated by co-treatment with B. fragilis LPS [14]. Therefore interactions between LPS molecules may be critical not just for the methylation profiles of CRC cells but also in the wider context of the tumour microenvironment. While we have shown only one interaction, it is an important illustration of the nuanced effects of the unique tumour-resident microbes, which may also affect the heterogeneity in methylation phenotypes observed in CRC patients.

The precise mechanism of action remains unknown and was not investigated in this study. It has previously been reported that LPS incubation induced a 2–4 fold increase in DNA methyltransferase 1 (DNMT1) concentration compared to controls in PBMCs. This suggests that LPS incubation may upregulate DNMT1 and therefore increase methylation levels through DNTM1 activity [29]. Furthermore, a study using human pulmonary microvascular endothelial cells (HPMECs) found that E.coli incubation for 12 h resulted in tet methylcytosine dioxygenase 2 (TET2) decrease, suggesting a possible decrease in de-methylation. [30]. E.coli LPS is functionally and structurally similar to Fusobacterium species LPS, as shown in the study by Garcia-Vello et al. 2021 [31]. This suggests that the F. periodonticum LPS used in this study may act in a similar manner with regards to DNMT1 and TET2, resulting in an increase in methylation activity. However, future research is required to determine how F. periodonticum LPS influences DNMT1 and TET2 activity specifically.

Furthermore, the differences in LPS structure may possibly influence the different methylation patterns observed from F. periodonticum and B. fragilis LPS treatments. Jacobson et al. proposed a penta-acylated LPS structure for B. fragilis [32], whereas a hexa-acylated LPS structure has been proposed for Fusobacterium species [31], and as previous research suggests, LPS structure has been shown to influence the magnitude of biological effects [33]. This would suggest that structurally different LPS molecules could exert different effects on methylation, both on the type and magnitude of methylation. Further research is necessary into how each LPS molecule may influence the different components involved in DNA methylation, such as DNMT1 and TET2.

This study is limited by the use of HT29 cells alone; however, the aim of this work was to establish a proof-of-concept that tumour-resident bacteria and their LPS molecules are a mechanistic link to altered DNA CpG methylation, commonly observed in a subset of colorectal tumours. We have demonstrated that treatment of HT29 CRC cells with F. periodonticum LPS and B. fragilis LPS, resulted in significant alterations to DNA methylation profiles, subsequently influencing gene expression patterns. Specifically, this study demonstrates that bacterial LPS treatment can result in both hypermethylation and hypomethylation of gene promoter regions, potentially leading to transcriptional repression or activation of key genes involved in colorectal carcinogenesis. These findings provide initial evidence that bacterial LPS may induce epigenetic alterations in tumour cells, a novel mechanism by which the microbiome may contribute to CRC. Although further validation in additional models is warranted in future studies, this work establishes a foundation for understanding how specific bacterial products can modulate the epigenetic landscape of CRC.

Supplementary Information

Supplementary file 1. (570.3KB, docx)

Author contributions

JP carried out laboratory and bioinformatics analysis and wrote the initial draft of the manuscript; AS carried out bioinformatics analysis and contributed towards writing of the manuscript; TE contributed to study design and interpretation of the data; RP designed the study, contributed to manuscript revision and interpretation of data. All authors reviewed the final manuscript.

Funding

Rachel Purcell and Arielle Sulit received funding from Maurice Wilkins Centre for Molecular Biodiscovery and Bowel Cancer Research, Aotearoa and Rachel Purcell received funding from the Health Research Council of New Zealand.

Data availability

The raw bisulfite-sequenced data was deposited to SRA database under Bioproject ID: PRJNA1237886, and are available at the following URL: http:/www.ncbi.nlm.nih.gov/bioproject/1237886.

Declarations

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  • 1.Arnold M, Sierra MS, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global patterns and trends in colorectal cancer incidence and mortality. Gut. 2017;66(4):683–91. [DOI] [PubMed] [Google Scholar]
  • 2.Xi Y, Xu P. Global colorectal cancer burden in 2020 and projections to 2040. Transl Oncol. 2021;14(10):101174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Permain J, Hock B, Eglinton T, Purcell R. Functional links between the microbiome and the molecular pathways of colorectal carcinogenesis. Cancer Metastasis Rev. 2024;1–12. [DOI] [PMC free article] [PubMed]
  • 4.Belkhiri A, El-Rifai W. 5-methylcytosine hydroxylation- mediated LINE-1 hypomethylation: a novel mechanism of proto-oncogenes activation in colorectal cancer? Gut. 2014;63(4):538–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Sunami E, de Maat M, Vu A, Turner RR, Hoon DS. LINE-1 hypomethylation during primary colon cancer progression. PLoS ONE. 2011;6(4):e18884. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Hur K, Cejas P, Feliu J, Moreno-Rubio J, Burgos E, Boland CR, et al. Hypomethylation of long interspersed nuclear element-1 (LINE-1) leads to activation of proto-oncogenes in human colorectal cancer metastasis. Gut. 2014;63(4):635–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Chen CY, Wu JJ, Lin YJ, Hsu CH, Hu JM, Chang PK, et al. Significance of hypermethylation of tumor-suppressor genes PTGER4 and ZNF43 at CpG sites in the prognosis of colorectal cancer. Int J Mol Sci. 2022;23(18):10225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Yegnasubramanian S, Kowalski J, Gonzalgo ML, Zahurak M, Piantadosi S, Walsh PC, et al. Hypermethylation of CpG islands in primary and metastatic human prostate cancer. Cancer Res. 2004;64(6):1975–86. [DOI] [PubMed] [Google Scholar]
  • 9.De Palma FDE, D’argenio V, Pol J, Kroemer G, Maiuri MC, Salvatore F. The molecular hallmarks of the serrated pathway in colorectal cancer. Cancers. 2019;11(7):1017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Hale VL, Jeraldo P, Chen J, Mundy M, Yao J, Priya S, et al. Distinct microbes, metabolites, and ecologies define the microbiome in deficient and proficient mismatch repair colorectal cancers. Genome Med. 2018;10:1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Angrisano T, Pero R, Brancaccio M, Coretti L, Florio E, Pezone A, et al. Cyclical DNA methylation and histone changes are induced by lps to activate cox-2 in human intestinal epithelial cells. PLoS ONE. 2016;11(6):e0156671. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Cheng C, Huang C, Ma TT, Bian EB, He Y, Zhang L, et al. SOCS1 hypermethylation mediated by DNMT1 is associated with lipopolysaccharide-induced inflammatory cytokines in macrophages. Toxicol Lett. 2014;225(3):488–97. [DOI] [PubMed] [Google Scholar]
  • 13.Kang XC, Chen ML, Yang F, Gao BQ, Yang QH, Zheng WW, et al. Promoter methylation and expression of SOCS-1 affect clinical outcome and epithelial-mesenchymal transition in colorectal cancer. Biomed Pharmacother. 2016;80:23–9. [DOI] [PubMed] [Google Scholar]
  • 14.Sulit A, Daigneault M, Allen-Vercoe E, Silander O, Hock B, McKenzie J, et al. Bacterial lipopolysaccharide modulates immune response in the colorectal tumor microenvironment. NPJ Biofilms Microbiomes. 2023;9(1):59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Krueger F (n.d.). Babraham bioinformatics - trim Galore! www.bioinformatics.babraham.ac.uk. https://www.bioinformatics.babraham.ac.uk/projects/trim_galore/
  • 16.Bowtie 2: fast and sensitive read alignment. (n.d.). Bowtie-Bio.sourceforge.net. https://bowtie-bio.sourceforge.net/bowtie2/index.shtml
  • 17.Krueger F, Andrews SR. Bismark: a flexible aligner and methylation caller for Bisulfite-Seq applications. Bioinformatics. 2011;27(11):1571–2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2− ΔΔCT method. Methods. 2001;25(4):402–8. [DOI] [PubMed] [Google Scholar]
  • 19.Liu Z, Zhang Q, Zhang H, Yi Z, Ma H, Wang X, et al. Colorectal cancer microbiome programs DNA methylation of host cells by affecting methyl donor metabolism. Genome Med. 2024;16(1):77. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Xia X, Wu WKK, Wong SH, Liu D, Kwong TNY, Nakatsu G, et al. Bacteria pathogens drive host colonic epithelial cell promoter hypermethylation of tumor suppressor genes in colorectal cancer. Microbiome. 2020;8:1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Wang S, Zhang Y, Huang J, Wong CC, Zhai J, Li C, et al. TRIM67 activates p53 to suppress colorectal cancer initiation and progression. Cancer Res. 2019;79(16):4086–98. [DOI] [PubMed] [Google Scholar]
  • 22.Liu Y, Wang G, Jiang X, Li W, Zhai C, Shang F, et al. TRIM67 inhibits tumor proliferation and metastasis by mediating MAPK11 in colorectal cancer. J Cancer. 2020;11(20):6025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Yang L, Bhattacharya A, Li Y, Sexton S, Ling X, Li F, et al. Depleting receptor tyrosine kinases EGFR and HER2 overcomes resistance to EGFR inhibitors in colorectal cancer. J Exp Clin Cancer Res. 2022;41(1):184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Uen YH, Fang CL, Hseu YC, Shen PC, Yang HL, Wen KS, et al. VAV3 oncogene expression in colorectal cancer: clinical aspects and functional characterization. Sci Rep. 2015;5(1):9360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Graham-Paquin AL, Saini D, Sirois J, Hossain I, Katz MS, Zhuang QKW, et al. Zmym2 is essential for methylation of germline genes and active transposons in embryonic development. Nucleic Acids Res. 2023;51(14):7314–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Zeng X, Zheng W, Sheng Y, Ma H. UBE2B promotes ovarian cancer growth via promoting RAD18 mediated ZMYM2 monoubiquitination and stabilization. Bioengineered. 2022;13(4):8000–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Yu J, Li D, Jiang H. Emerging role of ONECUT2 in tumors. Oncol Lett. 2020;20(6):1–1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Sun Y, Shen S, Liu X, Tang H, Wang Z, Yu Z, et al. MiR-429 inhibits cells growth and invasion and regulates EMT-related marker genes by targeting Onecut2 in colorectal carcinoma. Mol Cell Biochem. 2014;390:19–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Koos B, Moderegger EL, Rump K, Nowak H, Willemsen K, Holtkamp C, et al. LPS-induced endotoxemia evokes epigenetic alterations in mitochondrial DNA that impacts inflammatory response. Cells. 2020;9(10):2282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Qin W, Brands X, Matsumoto H, Butler JM, van’t Veer C, de Vos AF, et al. Role of myeloid tet methylcytosine dioxygenase 2 in pulmonary and peritoneal inflammation induced by lipopolysaccharide and peritonitis induced by Escherichia coli. Cells. 2021;11(1):82. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Garcia-Vello P, Di Lorenzo F, Lamprinaki D, Notaro A, Speciale I, Molinaro A, et al. Structure of the O-antigen and the lipid A from the lipopolysaccharide of Fusobacterium nucleatum ATCC 51191. ChemBioChem. 2021;22(7):1252–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Jacobson AN, Choudhury BP, Fischbach MA. The biosynthesis of lipooligosaccharide from Bacteroides thetaiotaomicron. MBio. 2018;9(2):10–1128. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Kuzmich NN, Sivak KV, Chubarev VN, Porozov YB, Savateeva-Lyubimova TN, Peri F. TLR4 signaling pathway modulators as potential therapeutics in inflammation and sepsis. Vaccines. 2017;5(4):34. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary file 1. (570.3KB, docx)

Data Availability Statement

The raw bisulfite-sequenced data was deposited to SRA database under Bioproject ID: PRJNA1237886, and are available at the following URL: http:/www.ncbi.nlm.nih.gov/bioproject/1237886.

Articles from Clinical Epigenetics are provided here courtesy of BMC

RESOURCES