Table 2 Seminal papers on DNA methylation in IBD, psoriasis and AS. Study design elements and key findings are outlined
Condition | Reference | Study design | Patients | Controls | Tissue | Measurement method | Key findings |
|---|---|---|---|---|---|---|---|
Inflammatory bowel disease | Hasler [33] | MZ CC | 20 UC discordant MZ twins 50 inflamed UC; 30 non-inflamed UC | 50 HC 25 inflamed disease controls 30 non-inflamed disease controls | Intestinal mucosa tissue | Illumina Methylation 27 BeadChip Custom tiling array; pyrosequencing | Integrated DEG data, MeDIP-seq data and Illumina Methylation 450K data was used to identify 61 IBD-associated loci harbouring DMP in cis of a DEG. All were novel candidate risk loci for IBD including SPINK4, THY1 and CFI. |
McDermott [34] | CC | 150 IBD patients; 24 paediatric IBD | 40 HC; 22 paediatric HC | PBMC; colonic mucosa tissue | Illumina Methylation 450K BeadChip; pyrosequencing | 3196 probes were significantly differentiated between CD patients and healthy controls, and 1418 probes between UC patients and controls. The most significant DMP for both groups was in the 5′UTR of TIFAB. | |
Ventham [35] | CC | 240 IBD patients | 74 symptomatic HC 117 HC | Whole blood, CD4+ T cells, CD8+ T cells, CD14+ monocytes | Illumina Methylation 450K BeadChip | Samples clustered by cell type separately from PBMCs. The top DMP in PBMCs was driven by changes in CD14+ monocytes, not either T cell subset. Conversely, some significant DMP in cell subsets were not detectable in whole PBMC. DMP were significantly associated with known IBD-risk loci. | |
Howell [36] | CC | 66 paediatric IBD | 30 paediatric HC | Bowel mucosal biopsies | Illumina Methylation 450K or EPIC BeadChip; pyrosequencing | Studied DNA methylation, gene expression and gut microbiota from a single cohort at multiple intestinal sites. Site-specific signatures were observed for DNA methylation and gene expression in the ascending colon and sigmoid colon compared to the terminal ileum. | |
Somineni [37] | CC | 164 paediatric CD | 74 paediatric HC | Peripheral blood | Illumina MethylationEPIC BeadChip | Previous findings were replicated including VMP1, RPS6KA2, ITGB2, and TXK, alongside known therapeutic targets (TNF, JAK3, IL12B, IL23A, and IL1R1). 194 DMP had a strong evidence of genetic effect. The Crohn’s methylation signature was a powerful predictor of disease (AUC = 0.91), but had no prognostic utility. | |
Psoriasis | Gervin [38] | MZ | 27 MZ pairs discordant for psoriasis | 27 HC | CD4+ cell and CD8+ cell from PBMCs | Illumina Methylation 27 BeadChip | DNA methylation highly correlated between monozygotic twins in CD4+ T cells and CD8+ T cells. No differences were identified between individual CpG sites or overall methylation per gene. When combined with gene expression data, cell-specific differences were identified, including IL13, ALOX5AP, PTHLH and TNFSF11. |
Zhou (1) [39] | CC | 114 psoriasis patients | 62 HC | Skin biopsies PP and PN | Illumina Methylation 450K BeadChip; Sequenom Epityper system | 129 SNP-CpG pairs achieved statistical significance between psoriatic and healthy control peripheral blood, constituting 28 unique meQTLs and 34 unique CpGs. 11 SNP-CpG pairs passed CIT constituting 3 unique CpG sites within C1orf106, DMBX1 and SIK3. | |
Zhou (2) [40] | CC | 114 Psoriasis patients (41 PP+PN) | 62 HC | skin biopsies; Previous methylation data | Illumina Methylation 450K BeadChip | 1514 DMP were identified between PP and PN, only 426 DMP were identified between PP and PN, and none between PN and NN. During replication, 9 sites reached significance (CYP2S1, ECE1, EIF2C2, MAN1C1, DLGAP4, S100A9, S100A7A and S100A5), but none replicated in PBMCs. | |
Pollock [41] | CC | 23 psoriasis patients without PsA; 13 PsA individuals | 18 HC | Sperm | Illumina Infinium Human Methylation 450K BeadChip | 2467 DMP identified between PsA and healthy controls, compared to 574 DMP between psoriasis and controls. DMR were enriched for the MHC complex. IL22 was associated with joint and skin involvement and was the only site to have correlated methylation levels between sperm and blood. | |
Ankylosing spondylitis | Aslani [42] | CC | 40 AS | 40 HC | PBMC | MSP | DNMT1 promoter was hypermethylated in cases compared to controls. No significant difference between HLA-B*27+ and HLA-B*27- patients, or disease activity scores. |
Karami [43] | CC | 50 AS | 50 HC | PBMC | MSP | Cases had decreased BCL11B expression associated with increased methylation in the promoter region. | |
Hao [44] | CC | 10 AS | 10 HC | PBMC | Illumina Methylation 450K BeadChip | 1915 CpGs were identified as differentially methylated between cases and controls. The most significant was HLA-DQB1, previously associated with AS radiographic severity and age of onset. Increased DNA methylation associated with decreased HLA-DQB1 expression. |