Abstract
DNA methylation has been implied to play a role in the immune dysfunction associated with inflammatory bowel disease (IBD) and the disease development of ulcerative colitis (UC). Changes of the DNA methylation and correlated gene expression in patient samples with inactive UC might reveal possible regulatory features important for further treatment options for UC. Targeted bisulfite sequencing and whole transcriptome sequencing were performed on mucosal biopsies from patients with active UC (UC, n = 14), inactive UC (RM, n = 20), and non-IBD patients which served as controls (NN, n = 11). The differentially methylated regions (DMRs) were identified by DMRseq. Correlation analysis was performed between DMRs and their nearest differentially expressed genes (DEGs). Principal component analysis (PCA) was performed based on correlated DMR regulated genes. DMR regulated genes then were functional annotated. Cell-type deconvolutions were performed based on methylation levels. The comparisons revealed a total of 38 methylation-regulated genes in inactive UC that are potentially regulated by DMRs (correlation p value < 0.1). Several methylation-regulated genes could be identified in inactive UC participating in IL-10 and cytokine signalling pathways such as IL1B and STAT3. DNA methylation events in inactive UC seem to be fine-tuned by the balancing pro- and anti- inflammatory pathways to maintain a prevailed healing process to restore dynamic epithelium homeostasis.
Similar content being viewed by others
Introduction
Ulcerative colitis (UC) is a chronic inflammatory disorder of the colon with a relapsing course1. To achieve remission, long-term treatment is often required2. Multiple factors can cause the disease, and research has focused on investigating genetic susceptibility, microbiome communities, environmental factors, and immune responses in UC patients3,4,5 Current research denotes the importance of interactions between inflammation and environmental factors6,7,8. Therefore, dynamic processes such as DNA methylation, have been implied to play a role in the immune dysfunction associated with inflammatory bowel disease (IBD)9,10,11 and disease development of UC12,13,14,15. DNA methylation in active UC has been reported recently9,10,12,13. Variations in DNA methylation patterns have been associated with homeostasis and defence, immune responses, and progression and development of colorectal cancer (CRC)12,13.
However, DNA methylation during UC in remission has not been explored in detail. Long-term treatment of UC patients is often necessary to achieve induction and maintenance of clinical remission16,17,18. Previous work on transcriptional signatures in UC has revealed that UC in remission is a permanently altered state of UC with a still ongoing quiescent inflammation19,20. Therefore, induced epigenetic changes due to long-term treatment can be expected. In this study, the genome-wide DNA methylome in a UC remission patient cohort was investigated to see if methylation contributes to the expression regulation of specific molecular signatures recently found in a patient cohort with different remission duration21. It is believed that changes in the DNA methylation status in remission patients, when correlated to differentially expressed genes (DEGs), may reveal possible regulatory features important for further treatment options for UC patients.
Materials and methods
Patient material
Mucosal biopsies (n = 43) were collected with a standardised sampling method of 12 newly diagnosed, treatment-naïve UC patients with mild to moderate disease activity, 20 patients with inactive UC (remission) and 11 control subjects from a former study21. Controls were derived from subjects performing a cancer screening, with normal colonoscopy and normal colonic histological examination. UC was diagnosed based upon established clinical endoscopic and histological criteria as defined by the ECCO guidelines22. The grade of inflammation was assessed during colonoscopy using the UC disease activity index (UCDAI) endoscopic sub-score with 3 to 10 for mild to moderate disease23. TNF-α mRNA expression levels were measured by real-time PCR to determine disease activity. All UC patients were initially treated with 5-aminosalicylic acid (5-ASA), in some cases also supplemented with immunosuppressive drugs (Imurel, methotrexate (MTX), and Infliximab) until disease remission. Patients with achieved clinical and endoscopic remission and normalised TNF-α levels were included in this study, with a defined UCDAI score ≤ 2, an endoscopic sub-score of 0 or 1, and TNFα-levels < 7500 copies/μg protein. All patient characteristics are depicted in Table 1. The samples were taken from an established Biobank approved by the Norwegian Board of Health (952/2006). The study participants signed informed and written consent forms. Approvals were granted by the Regional Committee of Medical Ethics of Northern Norway, Ref no: 14/2004, 1349/2012 and 29895/2020.
DNA and RNA isolation
Genomic DNA and total RNA was isolated using the Allprep DNA/ RNA Mini Kit from Qiagen (catalogue number 80204) and the QIAcube instrument (Qiagen), according to the manufacturer's protocol. The quantity and purity of both DNA and RNA were assessed by using the NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Wilmington, DE). The Experion Automated Electrophoresis System (Bio-Rad, Hercules, CA.) and the RNA StdSens Analysis Kit (Bio-Rad, catalogue number 700–7103) were used to evaluate RNA integrity. All RNA samples used for analyses showed an RNA integrity number (RIN) value between 8.0 and 10.0. DNA and RNA samples were kept at − 70°C until further use.
Institutional review board statement
The study was conducted according to the guidelines of the Declaration of Helsinki. Approvals were granted by the Regional Committee of Medical Ethics of Northern Norway, Ref no: 14/2004, 1349/2012 and 29,895/2020.
Informed consent statement
Written informed consent has been obtained from the study participants to publish this paper.
Library preparation and next-generation sequencing
The libraries were prepared using the SeqCap Epi CpGiant Enrichment Kit (Roche, Switzerland) which enables the targeting of selected genomic regions from bisulfite-treated genomic DNA to identify specific regions in the genome for methylation variation assessment and as previously described13. DNA was bisulfite converted using the EZ DNA Methylation-lightning Kit (Zymo Research, USA, cat no: D5030) prior to the hybridisation step and according to the manufacturer’s instructions. The amount of input material was 1060 ng of genomic DNA per sample. DNA library quality was assessed using the Bioanalyzer 2100 and the Agilent DNA 1000 kit (cat no: 5067–1504, Agilent Technologies, Santa Clara, USA), according to the manufacturer's instructions. DNA libraries generated fragments with an average size of 322 bp. DNA libraries were diluted to 4nM before sequencing. Whole transcriptome libraries were prepared with the TruSeq Stranded Total RNA LT Sample Prep Kit from Illumina (cat no: RS-122–2203). The amount of input material was 1μg of total RNA. The Bioanalyzer 2100 and the Agilent DNA 1000 kit (cat no: 5067–1504, Agilent Technologies, Santa Clara, USA) were used to assess the quality of the RNA libraries. RNA libraries generated fragments with an average size of 301 bp. Libraries were normalised to 10 nM and diluted to 4 nM prior to sequencing. DNA and RNA libraries were sequenced on the NextSeq 550 instrument, using a high output flow cell 150 cycles (cat no: FC-404–2002, Illumina, USA) and according to the manufacturer's instruction. The libraries were sequenced using paired-end mode.
Data analysis
A flow chart illustrating the downstream analysis process is shown in Fig. 1.
RNA sequencing
Quality scoring, base calling, and adapter removal were performed on the Illumina NextSeq 550 sequencing instrument. The output FastQ file was aligned with reference GENCODE Human Release 33 (Human Genome Assembly GRCh38.p13) ((https://www.ncbi.nlm.nih.gov/grc/human/data) by Kallisto24. DEseq2 was used to generate a list of differential expressed gene transcripts (DEGs)25. DEGs with only p.adj < 0.05 were kept.
Bisulfite converted DNA and DMRSeq
Quality scoring, base calling, and adapter removal were performed on the Illumina NextSeq 550 sequencing instrument. The output FastQ file was aligned with reference GENCODE Human Release 38 by Bismark with Bowtie226,27. The output BAM files were then used to generate coverage data and methylation data by using Bismark methylation extractor27. The results of Bismark were further processed with DMRseq28 to find differentially methylated regions. Only differentially methylated regions (DMRs) with a q-value < 0.05 were kept. In this context, the q-value served as an adjustment for multiple comparison to control the false discovery rate.
Identification of possible specific DMRs of inactive UC (RM)
DMRseq was performed on three individual comparisons: inactive UC (RM) vs active UC (UC), inactive UC (RM) vs controls (NN), and active UC (UC) vs controls (NN) resulting in three sets of DMRs with q < 0.05. The resulting DMRs were merged by overlapping genomic locations, and non-overlapping DMRs were discarded. To identify inactive UC specific DMRs additional t-tests were performed between the average DMR methylation levels of inactive UC samples against both control and active UC samples. DMRs whose p-value was less than 0.05 in both t-test comparisons were considered as inactive UC specific. The additional t-test ensured that the inactive UC group was in fact different from both the active UC and control group.
Identifying DEGs correlated to specific DMRs of inactive UC (RM)
For each DMR located within the 2000 bp upstream region of the transcription start site (TSS), sample DMR and DEG were correlated using average DMR methylation and DEG expression levels by Kendall correlation29. UCSC known gene (GRCh38) with R package TxDb.Hsapiens.UCSC.hg38.knownGene were used for TSS sites and functional region reference30. Correlated genes with p < 0.1 were kept based on the correlation coefficient τ < 0.
Annotations and pathway enrichment
Genes associated with transitional methylation patterns, where inactive UC (RM) methylation levels are between active UC (UC) and controls (NN) went through enrichment analysis for pathways enrichment using the Panther/Reactome overrepresentation analysis (Reactome version 77 released 2021–10-01) using the Fisher’s exact test31. For GO annotations of all 38 genes, clusterProfiler was used32.
Cell deconvolutions
Cell deconvolutions were performed on DMRs that overlapped Illumina Epic array coordinates using the EpiDISH R package33. Average sample relative methylation values for these DMRs were used as input to EpiDISH in Robust Partial Correlation (RPC) mode. Differences between groups were calculated using ANOVA and Tukey’s range test34. Cell deconvolutions for the expression data were performed with CIBERSORTx (https://cibersortx.stanford.edu/). The LM22 (22 immune cell types) was selected as a signature matrix. The normalized expression matrix was chosen as the input matrix file. The remaining parameters were left at default values35.
Results
Characterisation of DNA methylation in inactive UC
By combining genome-wide methylation data and whole transcriptome data, insight into the molecular mechanisms of inactive UC was established. Bisulfite sequencing provided DNA methylation levels in patient biopsy samples from inactive UC (n = 20), treatment-naïve active UC (n = 12) and non-IBD controls (n = 11). (Table 1). The different methylation patterns between inactive UC, active UC and controls were identified by DMRseq28. DMRs were detected in the following comparisons: 313 DMRs were detected in inactive UC vs. controls, 5,316 DMRs were detected in inactive UC vs. active UC, and 8,262 DMRs were detected in active UC vs. controls. By considering all three comparisons, the methylation levels of a total of 3098 DMRs were negatively correlated with neighbouring transcript expression. Analysis of the combined DMRs revealed 52 DMRs (38 genes) with specific and transitional patterns as depicted in Supplementary Data 1, Fig. 2. Principal component analysis (PCA) with the correlated DMRs could discriminate samples of inactive UC, active UC, and controls at the transcriptomic level (Fig. 3), with 51.0% and 13.3% variance. An example for detailed comparisons between expression and methylation levels are visualised for annexin A11 (ANXA11) (Fig. 4). Here, the presence of DMRs provides valuable information regarding the potential regulation positions and overlapping on the promoter region (shown as “Prom” in Supplementary data 2). ANXA11 is specifically hypo-methylated in DMRs on chr10.470 compared to controls and active UC, thus the expression of ANXA11 in inactive UC is uniquely over-expressed compared to controls and UC (upper left panel, Fig. 4). It is hereby noted that the DMRs labels are specific to this study and do not represent universal IDs.
Methylation-regulated gene profiling in inactive UC.
Among the 38 genes, two methylation patterns were found, one which is specific for inactive UC compared to controls and active UC, and one where methylation patterns of inactive UC are in transitional state between controls and active UC (Fig. 2). Two specific profiles were identified where inactive UC can be further subdivided into hyper- and hypomethylated compared to UC and controls. Two different transitional profiles were identified where inactive UC shows an intermediate state of methylation in relation to controls and active UC (Supplementary Data 1).
Genes with specific methylation patterns that are hyper-methylated in inactive UC include competing endogenous lncRNA 1 for mir-4707-5p and mir-4767 (CERNA2), citron rho-interacting serine/threonine kinase (CIT), integrin subunit alpha 4 (ITGA4), and zinc finger protein 626 (ZNF626). Genes uniquely hypo-methylated in inactive UC compared to active UC includes actin filament associated protein 1 antisense RNA 1 (AFAP1-AS1), annexin A11 (ANXA11), hyaluronidase 1 (HYAL1), macrophage stimulating 1 receptor (MST1R), and S100 calcium binding protein P (S100P).
The transitional genes patterns represent genes that have an intermediate methylation status between UC and controls (Fig. 2). 28 genes were found to be hyper-methylated in controls compared to active UC, including ADAM metallopeptidase domain 8 (ADAM8), coiled-coil domain containing 88B (CCDC88B), CD80, DENN domain containing 2D (DENND2D), dual specificity phosphatase 10 (DUSP10), signal transducer and activator of transcription 3 (STAT3), and interleukin 1B (IL1B), interferon stimulated exonuclease gene 20 (ISG20) and LIM domain kinase 1 (LIMK1). DIP2C antisense RNA 1 (PRR26) is the only gene found to be hyper-methylated in active UC compared to controls. A comprehensive list of figures of all methylation-regulated genes can be found in Supplementary Data 1.
Methylation-regulated genes are related to inflammation
The 28 methylation regulated genes in the intermediate state were functionally annotated with Gene Ontology (GO) terms with the Panther/Reactome overrepresentation test (Reactome v.77, released 2021-10-01) (Fig. 5). Genes like TRIM22, PSMB8, CD80, IL1B, ISG20, HIF1A, STAT3 were all hyper-methylated and downregulated to a lesser extent in inactive UC compared to UC. These genes were annotated to the IL-10 pathway and cytokine signalling in immune system (Fig. 5).
Functional enrichment revealed 101 immunological and inflammation related signalling pathways (p.adj < 0.05), which include interleukin-6 production, CD4-positive, alpha–beta T cell activation and lymphocyte differentiation represented by genes like ADAM8, IL1B, ISG20, CD80, STAT3 and ZC3H12A (Fig. 5; Supplementary Data 3). Genes in specific patterns including CIT, MST1R, HYAL1, ITGA4 were annotated as hyaluronan metabolic process, epithelium migration and phagocytosis. The total of 103 functional GO terms are listed in Supplementary Data 3.
Cell deconvolutions based on DNA methylations discriminate cell fractions
Cell deconvolution was performed by mapping DMR genomic coordinates to Illumina EPIC array identifiers33. The deconvolution results revealed differences in cell fractions for inactive UC (RM), active UC (UC) and controls (NN). Epithelial cell fractions were higher in normal and inactive UC than active UC (padj < 0.01). The proportion of immune cells is significantly higher in active UC compared to inactive UC and controls (padj < 0.01). Similarly, the fibroblast cell fractions in inactive UC and controls are slightly higher than in active UC (Fig. 6, Supplementary Data 4). Using CIBERSORTx on the normalized gene expression matrix showed an increase in most immuno-derived cells, especially of neutrophils in active UC as compared to inactive UC and controls (Supplementary Data 5)35.
Discussion
In a previous work, DNA methylation patterns have been identified for active, treatment-naïve UC12,13,36. In this study, the methylation status in inactive UC (RM) has been explored by determining gene expression regulated by global DNA methylation overlapping the promoter region of genes. By correlating DNA methylation data to expression levels of genes several regulatory DNA methylation features of relevance for inactive UC could be identified and are discussed below.
Long-term treatment of UC patients is often necessary to achieve induction and maintenance of clinical remission16,17,18. UC medications such as immunosuppressive drugs have been shown to have side effects on immune response and can change the DNA methylations status37,38. Induced epigenetic changes due to long-term treatment can be expected for 5-ASA which is commonly used as a first-line treatment for UC and might therefore have the potential to change the methylation status. This notion is supported by in vitro studies that have shown that 5-ASA treatment increases the expression of DNA methyltransferase 1 (DMNT1) which is responsible for most of the methylation events occurring on the human genome39 Therefore, it can be anticipated that the observed methylation changes in inactive UC patients might be a result of 5-ASA treatment.
Four genes have been found to be specifically hyper-methylated in inactive UC compared to active UC and controls (NN) (Fig. 2), CERNA2, CIT, IGTA4 and ZNF 626 (Supplementary Data 1). The observed specific hyper-methylation of CERNA2 in inactive UC compared to UC indicates anti-inflammatory characteristics. CERNA2 has been recently reported to play a role in inflammation40 and has been identified as an independent predictor for clinical prognosis of gastric cancer41. It is notable that the expression of CERNA2 has been shown to correlate with poor clinical parameters and an unfavourable prognosis of different cancer patient groups while silencing of CERNA2 expression inhibits cancer cell growth and promotes cell apoptosis42,43. In this context, CIT was specifically hyper-methylated in inactive UC. Silencing of CIT has been shown to reduce tumour growth in multiple myeloma and breast cancer cells by promoting apoptosis44,45. Therefore, specific hypermethylation of CIT and CERNA2 might play a role in the fine-tuning of the regulation of apoptotic events during inactive UC21. ITGA4 is another specifically hyper-methylated gene in inactive UC compared to UC and is a well-known therapeutic target for the treatment of IBD (Fig. 2 & Supplementary Data 2). The observed downregulation of ITGA4 expression implies reduced leukocyte infiltration into the GI tract through the interaction with MAdCAM‐1 which is expressed on high endothelial venules (HEV) within vessels of mucosal tissue46,47,48,49,50.
Quiescent inflammation present in inactive UC has previously been reported20,51,52 and many of the observed methylation-regulated expressed genes identified in inactive UC are involved in inflammation. Surprisingly, increased levels of a well-known marker for inflammation S100P were observed to a greater extent in inactive UC compared to active UC and controls. The observed specific hypo-methylation and upregulation of S100P in inactive UC might, in addition to S100P’s inflammatory responses, contribute to the regulation of tissue development and regeneration or repair as previously reported53 (Fig. 2 & Supplementary Data 2).
Chronic inflammation like UC results in dramatic deposition of hyaluronic acid (HA) within affected tissues which both precedes and promotes immune cell infiltration, tissue destruction, and coagulation54. The observed increased expression of HYAL1 in inactive UC might lead to decreased levels of HA in active UC, thereby modulating the promotion and resolution of the disease by controlling recruitment of immune cells, by release of inflammatory cytokines, and by balancing haemostasis55 (Supplementary Table 2,Fig. 2). Hypo-methylation and increased expression of MSTR-1 in inactive UC compared to UC might directly be involved in the wound healing process by promoting epithelial cell migration and proliferation, as this innate immune response regulates the migration of macrophages increasing the phagocytic activity56. Increased fractions of epithelial cells and decreased fractions of immune cells in inactive UC are in concordance with the results obtained by cell deconvolutions (Fig. 6).
Two other specifically hypo-methylated genes in inactive UC of relevance for IBD include ANXA11 and lncRNA AFAP1-AS1. ANXA11 has been shown to be a sarcoidosis susceptibility gene57. An association between sarcoidosis and ulcerative colitis has been reported58. The expression of lncRNA AFAP1-AS1 has been shown to promote the progression of CRC59 and has been acknowledged as a biomarker for diagnosis and prognosis estimation of CRC patients60. However, the potential role of the here observed hypo-methylation of AFAP1-AS in inactive UC is still unclear.
Annotations of genes with intermediate methylation patterns for inactive UC revealed two pathways, IL10 signalling and cytokine signalling in immune system pathways (Fig. 5, Supplementary Data 3). These genes were all hyper-methylated and downregulated in inactive UC compared to active UC but to a lesser extent then in controls, meaning that normal levels of expression are not completely achieved by hyper-methylation. The downregulation of the proinflammatory cytokine IL1B, might reduce T cell immune response by downregulation of co-stimulatory molecules such as CD8061,62,63. In addition, IL1B production is diminished by the observed downregulation of STAT3 expression and implies reduced phosphorylation of IL1B. Hyper-methylation of STAT3 in inactive UC compared to active UC might be involved in the regulation of adaptive immune responses by reducing survival of pathogenic T cells and TNF-α64. The transcriptional coregulator SBNO2 (strawberry notch homolog 2) is hyper-methylated in inactive UC compared to UC and contributes to the downstream anti-inflammatory effects of IL-10 which is dependent on STAT3 activation65.
Compared to controls, ZC3H12A is hypo-methylated in inactive UC which might indicate a modulation of the inflammatory response by promoting the degradation of a set of translationally active cytokine-induced inflammation-related mRNAs, such as IL6 and IL12B66. ZC3H12A induces the deubiquitylation of the transcription factor HIF1A which is also hyper-methylated in inactive UC compared to active UC, thereby positively regulating the expression of proangiogenic HIF1A-targeted genes67. The decrease of HIF1A expression in inactive UC may function as a transcriptional regulator of the adaptive response to hypoxia maintaining biological homeostasis68. In this context, cell deconvolutions revealed epithelial cell fractions in inactive UC were comparable to epithelial fractions in control samples (Fig. 6). It is notable that somatic mutations of ZC3H12A have been found in UC patients’ epithelium which might have an unknown influence on the DNA methylation regulated expression69. The hypo-methylation of LIMK1 in inactive UC compared to controls could lead to reduced T-cell regulation in inactive UC through Rho/Rac pathways. A single nucleotide polymorphism (SNP) rs6460071 in LIMK1 has been reported to be most significantly associated with proximal endoscopic extension in CRC and is a predictor of outcome in UC70.
DNA methylation has been found to influence the regulation of interferon’s antiviral processes mediated by TRIM22, IGS20, and DENND2D. The hypermethylation of the interferon-induced antiviral protein TRIM22 compared to UC might contribute to a decrease in disease development through the NF‑κB signalling pathway71. The increased expression of antiviral ISG20 in UC still needs to be confirmed. However, the hypo-methylation and increased expression of ISG20 in inactive UC compared to controls may be a potential susceptibility biomarker or pharmacological target as has been shown for other inflammatory conditions72. It is hereby noted that a prognostic impact of expression and methylation status of DENN/MADD domain-containing protein 2D in gastric cancer has been proposed73,74,75. GALNT2 catalyses the initial reaction in O-linked oligosaccharide biosynthesis and has a broad spectrum of substrates for peptides such as MUC5AC, MUC1A, MUC1B. An increase of GALNT2 expression has been recently reported for UC patients in the active stage compared to patients in the remission76. This result might be in part also be explained and confirmed by the observed hyper-methylation of GALNT2 in inactive UC compared to UC (Supplementary Data 1, Fig. 2).
A limitation of this work is the small sample size and the heterogeneity of tissue samples as it was not possible to discriminate inactive UC in terms of remission duration and DNA methylation as has been recently seen for gene expression21. It is important to note that epigenome-wide association studies do not always precede changes in transcription as has been recently reported77.
Conclusions
Several differentially expressed genes involved in IL-10/cytokine signalling pathways may be under the control of DNA methylation events which might indicate fine-tuned processes regulating the balance between quiescent inflammation and mucosal healing in inactive UC.
Data availability
Processed RNA-seq data are deposited in NCBI’s Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/) and are accessible through GEO series accession numbers GSE128682 and GSE169360. Regarding the availability of DNA data, it is hereby noted that, according to the Norwegian Health Research Act §34, the processing of health information can only take place in accordance with the consent given. In this case, the availability of unprocessed DNA information would not be in accordance with the participants’ consent. All data generated or analysed during this study are included in this published article and supplementary data files.
References
Khor, B., Gardet, A. & Xavier, R. J. Genetics and pathogenesis of inflammatory bowel disease. Nature 474, 307–317 (2011).
Peyrin-Biroulet, L., Bressenot, A. & Kampman, W. Histologic remission: the ultimate therapeutic goal in ulcerative colitis?. Clin Gastroenterol Hepatol 12, 929–34 e2 (2014).
Jostins, L. et al. Host–microbe interactions have shaped the genetic architecture of inflammatory bowel disease. Nature 491, 119–124 (2012).
Lee, J. C. et al. Genome-wide association study identifies distinct genetic contributions to prognosis and susceptibility in Crohn’s disease. Nat. Genetics 49, 262–268 (2017).
Moustafa, A. et al. Genetic risk, dysbiosis, and treatment stratification using host genome and gut microbiome in inflammatory bowel disease. Clin. Trans. Gastroenterol. 9, e132 (2018).
Fakhoury, M., Al-Salami, H., Negrulj, R. & Mooranian, A. Inflammatory bowel disease: clinical aspects and treatments. J. Inflamm. Res. 7, 113 (2014).
Aleksandrova, K., Romero-Mosquera, B. & Hernandez, V. Diet, gut microbiome and epigenetics: emerging links with inflammatory bowel diseases and prospects for management and prevention. Nutrients 9, 962 (2017).
Zhang, H., Zeng, Z., Mukherjee, A. & Shen, B. Molecular diagnosis and classification of inflammatory bowel disease. Expert Rev. Mol. Diagn. 18, 867–886 (2018).
Karatzas, P. S., Mantzaris, G. J., Safioleas, M. & Gazouli, M. DNA methylation profile of genes involved in inflammation and autoimmunity in inflammatory bowel disease. Medicine 93, e309 (2014).
McDermott, E. et al. DNA methylation profiling in inflammatory bowel disease provides new insights into disease pathogenesis. J Crohns Colitis 10, 77–86 (2016).
Harris, R. A. et al. DNA methylation-associated colonic mucosal immune and defense responses in treatment-naive pediatric ulcerative colitis. Epigenetics 9, 1131–1137 (2014).
Taman, H., Fenton, C. G., Anderssen, E., Florholmen, J. & Paulssen, R. H. DNA hypo-methylation facilitates anti-inflammatory responses in severe ulcerative colitis. PLOS ONE 16, e0248905 (2021).
Taman, H. et al. Genome-wide DNA methylation in treatment-naïve ulcerative colitis. J. Crohn’s Colitis 12, 1338–1347 (2018).
Hasler, R. et al. A functional methylome map of ulcerative colitis. Genome. Res. 22, 2130–2137 (2012).
Dunham, I. et al. An integrated encyclopedia of DNA elements in the human genome. Nature 489, 57–74 (2012).
Bryant, R., Sandborn, W. & Travis, S. Introducing vedolizumab to clinical practice: who, when, and how?. J. Crohn’s Colitis 9, 356–366 (2015).
D’Haens, G. Risks and benefits of biologic therapy for inflammatory bowel diseases. Gut 56, 725–732 (2007).
Neurath, M. F. & Travis, S. P. Mucosal healing in inflammatory bowel diseases: a systematic review. Gut 61, 1619–1635 (2012).
Planell, N. et al. Transcriptional analysis of the intestinal mucosa of patients with ulcerative colitis in remission reveals lasting epithelial cell alterations. Gut 62, 967–976 (2013).
Fenton, C. G., Taman, H., Florholmen, J., Sørbye, S. W. & Paulssen, R. H. Transcriptional signatures that define ulcerative colitis in remission. Inflamm. Bowel Dis. 27, 94–105 (2021).
Meng, W., Johnsen, K. M., Fenton, C. G., Florholmen, J. & Paulssen, R. H. Anti-apoptotic genes and non-coding RNAs are potential outcome predictors for ulcerative colitis. Funct. Integrat. Genom. 23, 165 (2023).
Magro, F. et al. Third European evidence-based consensus on diagnosis and management of ulcerative colitis. Part 1: definitions, diagnosis, extra-intestinal manifestations, pregnancy, cancer surveillance, surgery, and Ileo-anal pouch disorders. J. Crohns Colitis 11, 649–670 (2017).
Tursi, A. et al. Treatment of relapsing mild-to-moderate ulcerative colitis with the probiotic VSL#3 as adjunctive to a standard pharmaceutical treatment: a double-blind, randomized Placebo-Controlled Study. Am. J. Gastroenterol. 105, 2218–2227 (2010).
Bray, N. L., Pimentel, H., Melsted, P. & Pachter, L. Near-optimal probabilistic RNA-seq quantification. Nat. Biotechnol. 34, 525–527 (2016).
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).
Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).
Krueger, F. & Andrews, S. R. Bismark: a flexible aligner and methylation caller for Bisulfite-Seq applications. Bioinformatics 27, 1571–1572 (2011).
Korthauer, K., Chakraborty, S., Benjamini, Y. & Irizarry, R. A. Detection and accurate false discovery rate control of differentially methylated regions from whole genome bisulfite sequencing. Biostatistics 20, 367–383 (2019).
Kendall, M. G. A new measure of rank correlation. Biometrika 30, 81 (1938).
Team BC, M. B. TxDb.Hsapiens.UCSC.hg38.knownGene: annotation package for TxDb object(s). R package version 3.4.6 (2019).
Mi, H. & Thomas, P. PANTHER Pathway: an ontology-based pathway database coupled with dataanalysis tools. Methods Mol. Biol. 563, 123 (2009).
Yu, G., Wang, L.-G., Han, Y. & He, Q.-Y. Cluster profiler: an R Package for comparing biological themes among gene clusters. OMICS A J. Integrat. Biol. 16, 284–287 (2012).
Zheng, S. C., Breeze, C. E., Beck, S. & Teschendorff, A. E. Identification of differentially methylated cell types in epigenome-wide association studies. Nat. Methods 15, 1059–1066 (2018).
Tukey, J. W. Comparing individual means in the analysis of variance. Biometrics 5, 99 (1949).
Newman, A. M. et al. Robust enumeration of cell subsets from tissue expression profiles. Nat. Methods 2015(12), 453–457 (2015).
Paulssen, R. et al. Novel regulatory DNA methylation features of relevance for ulcerative colitis. Am. J. Gastroenterol. 113, S12 (2018).
Selvestrel, D. et al. DNA methylation of the TPMT gene and azathioprine pharmacokinetics in children with very early onset inflammatory bowel disease. Biomed. Pharmacother. 157, 113901 (2023).
Guderud, K. et al. Methotrexate treatment of newly diagnosed RA patients is associated with DNA methylation differences at genes relevant for disease pathogenesis and pharmacological action. Front. Immunol. 12, 713611 (2021).
Adamowicz, M., Milkiewicz, P. & Kempinska-Podhorodecka, A. 5-aminosalicylic acid inhibits the expression of oncomiRs and pro-inflammatory microRNAs: an in vitro study. J. Physiol. Pharmacol. 72, 529–535 (2021).
Lu, J. W., Rouzigu, A., Teng, L. H. & Liu, W. L. The construction and comprehensive analysis of inflammation-related ceRNA networks and tissue-infiltrating immune cells in ulcerative progression. BioMed Res. Int. 2021, 6633442 (2021).
Zhang, J., Han, X., Jiang, L., Han, Z. & Wang, Z. LncRNA CERNA2 is an independent predictor for clinical prognosis and is related to tumor development in gastric cancer. Int. J. Clin. Exp. Pathol. 11, 5783–5791 (2018).
Lin, J., Li, X., Zhang, F., Zhu, L. & Chen, Y. Transcriptome wide analysis of long non-coding RNA-associated ceRNA regulatory circuits in psoriasis. J. Cell. Mol. Med. 25, 6925–6935 (2021).
Wang, M., Ouyang, J. & Li, H. CERNA2: A predictor for clinical progression and poor prognosis in cervical carcinoma. J. Cell. Biochem. 120, 11216–11221 (2019).
Sahin, I. et al. Citron Rho-interacting kinase silencing causes cytokinesis failure and reduces tumor growth in multiple myeloma. Blood Adv. 3, 995–1002 (2019).
Meng, D. et al. Citron kinase (CIT-K) promotes aggressiveness and tumorigenesis of breast cancer cells in vitro and in vivo: preliminary study of the underlying mechanism. Clin. Transl. Oncol. 21, 910–923 (2019).
Berlin, C. et al. Alpha 4 beta 7 integrin mediates lymphocyte binding to the mucosal vascular addressin MAdCAM-1. Cell 74, 185–195 (1993).
Habtezion, A., Nguyen, L. P., Hadeiba, H. & Butcher, E. C. Leukocyte trafficking to the small intestine and colon. Gastroenterology 150, 340–354 (2016).
Butcher, E. C., Williams, M., Youngman, K., Rott, L. & Briskin, M. Lymphocyte trafficking and regional immunity. Adv. Immunol. 72, 209–253 (1999).
Dotan, I. et al. The role of integrins in the pathogenesis of inflammatory bowel disease: Approved and investigational anti-integrin therapies. Med. Res. Rev. 40, 245 (2020).
Nakache, M., Lakey-Berg, E., Streeter, P. R. & Butcher, E. C. The mucosal vascular addressin is a tissue-specific endothelial cell adhesion molecule for circulating lymphocytes. Nature 337, 179–181 (1989).
Riley, S. A., Mani, V., Goodman, M. J., Dutt, S. & Herd, M. E. Microscopic activity in ulcerative colitis: what does it mean?. Gut 32, 174–178 (1991).
Korelitz, B. I. Mucosal healing as an index of colitis activity: back to histological healing for future indices. Inflamm. Bowel Dis. 16, 1628–1630 (2010).
Xia, C., Braunstein, Z., Toomey, A. C., Zhong, J. & Rao, X. S100 proteins as an important regulator of macrophage inflammation. Front. Immunol. 8, 1908 (2018).
Petrey, A. C. & de la Motte, C. A. Hyaluronan in inflammatory bowel disease: cross-linking inflammation and coagulation. Mat. Biol. 78–79, 314 (2019).
Dokoshi, T. et al. Hyaluronidase inhibits reactive adipogenesis and inflammation of colon and skin. JCI Insight 3, e123072 (2018).
Han, X., Ding, S., Jiang, H. & Liu, G. Roles of macrophages in the development and treatment of gut inflammation. Front.Cell Dev. Biol. 9, 625423 (2021).
Karakaya, B. et al. ANXA11 rs1049550 associates with Löfgren’s syndrome and chronic sarcoidosis patients. Cells 11, 1557 (2022).
Fries, W. et al. Association between inflammatory bowel disease and sarcoidosis. Report of two cases and review of the literature. Scand. J. Gastroenterol. 30, 1221–1223 (1995).
Li, Y., Zhu, Z., Hou, X. & Sun, Y. LncRNA AFAP1-AS1 promotes the progression of colorectal cancer through miR-195–5p and WISP1. J. Oncol. 2021, (2021).
Elabd, N. S. et al. Long non-coding RNAs ASB16-AS1 and AFAP1-AS1: diagnostic, prognostic impact and survival analysis in colorectal cancer. Appl. Clin. Genetics 15, 97 (2022).
MacDermott, R. P. Chemokines in the inflammatory bowel diseases. J. Clin. Immunol. 19, 266–272 (1999).
Scarpa, M. et al. CD80-CD28 signaling controls the progression of inflammatory colorectal carcinogenesis. Oncotarget 6, 20058–20069 (2015).
Nakazawa, A. et al. Functional expression of costimulatory molecule CD86 on epithelial cells in the inflamed colonic mucosa. Gastroenterology 117, 536–545 (1999).
Sugimoto, K. Role of STAT3 in inflammatory bowel disease. World J. Gastroenterol. 14, 5110 (2008).
El Kasmi, K. C. et al. Cutting edge: A transcriptional repressor and corepressor induced by the STAT3-regulated anti-inflammatory signaling pathway. J. Immunol. 179, 7215–7219 (2007).
Garg, A. V. et al. MCPIP1 endoribonuclease activity negatively regulates interleukin-17-mediated signaling and inflammation. Immunity 43, 475–487 (2015).
Roy, A., Zhang, M., Saad, Y. & Kolattukudy, P. E. Antidicer RNAse activity of monocyte chemotactic protein-induced protein-1 is critical for inducing angiogenesis. Am. J. Physiol. Cell Physiol. 305, C1021–C1032 (2013).
Kerber, E. L. et al. The importance of hypoxia-inducible factors (HIF-1 and HIF-2) for the pathophysiology of inflammatory bowel Disease. Int. J. Mol. Sci. 21, 1–27 (2020).
Nanki, K. et al. Somatic inflammatory gene mutations in human ulcerative colitis epithelium. Nature 577, 254–259 (2020).
Waterman, M. et al. Predictors of outcome in ulcerative colitis. Inflamm. Bowel Dis. 21, 2097 (2015).
Ye, B. & Lu, Z. Role of TRIM22 in ulcerative colitis and its underlying mechanisms. Mol. Med. Rep. 26, 1–10 (2022).
Zheng, Z., Wang, L. & Pan, J. Interferon-stimulated gene 20-kDa protein (ISG20) in infection and disease: review and outlook. Intractable Rare Dis. Res. 6, 35–40 (2017).
Kanda, M. et al. Prognostic impact of expression and methylation status of DENN/MADD domain-containing protein 2D in gastric cancer. Gastric Cancer 18, 288–296 (2015).
Hibino, S. et al. Reduced expression of DENND2D through promoter hypermethylation is an adverse prognostic factor in squamous cell carcinoma of the esophagus. Oncol. Rep. 31, 693–700 (2014).
Kanda, M. et al. Downregulation of DENND2D by promoter hypermethylation is associated with early recurrence of hepatocellular carcinoma. Int. J. Oncol. 44, 44–52 (2014).
Tian, R. et al. A pharmacoinformatics analysis of artemisinin targets and de novo design of hits for treating ulcerative colitis. Front. Pharmacol. 13, 843043 (2022).
Pacis, A. et al. Gene activation precedes DNA demethylation in response to infection in human dendritic cells. Proc. Natl. Acad. Sci. USA 116, 6938–6943 (2019).
Acknowledgements
The authors thank Lars Daniel Håland at the Genomics Support Centre Tromsø (GSCT; (www.uit.no/helsefak/genomics) for technical assistance.
Funding
Open access funding provided by UiT The Arctic University of Norway (incl University Hospital of North Norway). This work was supported by institutional funding.
Author information
Authors and Affiliations
Contributions
W.M. performed statistical and functional analyses, visualization, and wrote parts of the manuscript. K.M.J. was involved in evaluating and providing clinical samples from patients in long-term remission. C.G.F. was involved in data analysis and visualization. H.T. was involved in library preparation and sequencing, J.F. was involved in evaluating and providing clinical samples from patients and healthy controls. R.H.P was involved in project inception, project design, data analysis, supervision, manuscript writing. All authors revised the manuscript and approved the last version of the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Meng, W., Fenton, C.G., Johnsen, KM. et al. DNA methylation fine-tunes pro-and anti-inflammatory signalling pathways in inactive ulcerative colitis tissue biopsies. Sci Rep 14, 6789 (2024). https://doi.org/10.1038/s41598-024-57440-0
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-024-57440-0
- Springer Nature Limited